Search

The aim of this paper is to explore what aspects create experiential value for wine tourists. We synthesize the extant literature into four dimensions for wine tourism value creation, namely, product-related aspects; sensory and affective experiential aspects; cognitive, educational experiential aspects; and social-relational experiential value-creating aspects. So far, most studies merely discuss product-related aspects whilst insights on experiential value are less known. Using online review data from wine tourists in Australia, we develop a novel deep neural network-based framework using an innovative AI-based exploratory design. Results of the case study reveal that in addition to product-related aspects, sensory-and education-related experiential aspects are also highly important for value creation in wine tourism. Theoretical and practical implications, as well as ideas for future research are discussed. [ABSTRACT FROM AUTHOR]

A Ground Coupled Heat Pump (GCHP) is a highly energy efficient heating, ventilation, and air conditioning (HVAC) system that utilises the ground as the heat source when heating and as the heat sink when cooling. This paper investigates GCHP systems with horizontal Ground Heat Exchangers (GHEs) in the rural industry, exemplifying the technology for poultry (chicken) sheds in Australia. This investigation aims to provide an Artificial Neural Network (ANN) model that can be used for GCHP design at various locations with different climates. To this extent, a Transient System Simulation Tool (TRNSYS) model for a typical horizontal GHE applied in a rural farm was first verified. Using this model, over 700,000 hourly performance data items were obtained, covering over 80 different yearly loading patterns under three different climate conditions. The simulated performance data was then used to train the ANN. As a result, the trained ANN can predict the performance of GCHP systems with identical (multiple) GHEs even under climatic conditions (and locations) that have not been specifically trained for. Unlike other works, the newly introduced ANN model is accurate even with limited types of input data, with high accuracy (less than 5% error in most cases tested). This ANN model is 100 times computationally faster than TRNSYS simulations and 10,000 times faster than finite element models. [ABSTRACT FROM AUTHOR]

Simple Summary: In radiotherapy treatment, organs at risk and target volumes are contoured by the clinicians to prepare a dosimetry plan. In retrospective data, these structures are not often standardised to universal names across the patients plans, which is required to enable data mining and analysis. In this paper, a new method was proposed and evaluated to automatically standardise radiotherapy structures names using machine learning algorithms. The proposed approach was deployed over a dataset with 1613 patients collected from Liverpool & Macarthur Cancer Therapy Centres, New South Wales, Australia. It was concluded that machine learning techniques can standardise the dosimetry plan structures, taking into consideration the integration of multiple modalities representing each structure during the training process. In progressing the use of big data in health systems, standardised nomenclature is required to enable data pooling and analyses. In many radiotherapy planning systems and their data archives, target volumes (TV) and organ-at-risk (OAR) structure nomenclature has not been standardised. Machine learning (ML) has been utilised to standardise volumes nomenclature in retrospective datasets. However, only subsets of the structures have been targeted. Within this paper, we proposed a new approach for standardising all the structures nomenclature by using multi-modal artificial neural networks. A cohort consisting of 1613 breast cancer patients treated with radiotherapy was identified from Liverpool & Macarthur Cancer Therapy Centres, NSW, Australia. Four types of volume characteristics were generated to represent each target and OAR volume: textual features, geometric features, dosimetry features, and imaging data. Five datasets were created from the original cohort, the first four represented different subsets of volumes and the last one represented the whole list of volumes. For each dataset, 15 sets of combinations of features were generated to investigate the effect of using different characteristics on the standardisation performance. The best model reported 99.416% classification accuracy over the hold-out sample when used to standardise all the nomenclatures in a breast cancer radiotherapy plan into 21 classes. Our results showed that ML based automation methods can be used for standardising naming conventions in a radiotherapy plan taking into consideration the inclusion of multiple modalities to better represent each volume. [ABSTRACT FROM AUTHOR]

The increase in atmospheric greenhouse gases over the last 200 years has caused an increase in ocean acidity levels. Documenting how the ocean has changed is critical for assessing how these changes could impact marine ecosystems and for the management of marine resources. We use present day ocean carbon observations from shelf and offshore waters around Australia, combined with neural network mapping of CO2, to estimate the current seasonal and regional distributions of carbonate chemistry (pH and aragonite saturation state). These predicted changes in carbonate chemistry are combined with atmospheric CO2 concentration changes since to reconstruct pH and aragonite saturation state changes over the last 140 years (1870-2013). The comparison with data collected at Integrated Marine Observing System National Reference Station sites located on the shelf around Australia shows both the mean state and seasonality for the present day is well represented by our reconstruction, with the exception of sites such as the Great Barrier Reef. Our reconstruction predicts that since 1870 an average decrease in aragonite saturation state of 0.48 and of 0.09 in pH has occurred in response to increasing oceanic uptake of atmospheric CO2. Our reconstruction shows that seasonality is the dominant mode of variability, with only small interannual variability present. Large seasonal variability in pH and aragonite saturation state occur in Southwestern Australia driven by ocean dynamics (mixing) and in the Tasman Sea by seasonal warming (in the case of aragonite saturation state). The seasonal and historical changes in aragonite saturation state and pH have different spatial patterns and suggest that the biological responses to ocean acidification are likely to be non-uniform depending on the relative sensitivity of organisms to shifts in pH and saturation state. This new historical reconstruction provides an important to link to biological observations to help elucidate the consequences of ocean acidification. [ABSTRACT FROM AUTHOR]

Annual turnover of home care workers represents a huge loss of revenue and is a key source of inefficiency in the home health care industry. In this article, we propose a data-driven approach to monitor employee churn and to capture the evolution of employee intent to leave. Unlike most papers in the literature, we use machine learning techniques to analyze over 2 million visits in the US, Canada, and Australia between 2016 and 2019. Results show that the gap between the number of hours worked and in the contract is the most important factor to predict employee intent to leave, which means an employee should be given as many hours as requested in the contract to improve retention. Secondary results show that having diverse shift lengths and continuity in services and patients seem to be associated with less turnover. [ABSTRACT FROM AUTHOR]

Design flood estimations at ungauged catchments are a challenging task in hydrology. Regional flood frequency analysis (RFFA) is widely used for this purpose. This paper develops artificial intelligence (AI)-based RFFA models (artificial neural networks (ANN) and support vector machine (SVM)) using data from 181 gauged catchments in South-East Australia. Based on an independent testing, it is found that the ANN method outperforms the SVM (the relative error values for the ANN model range 33–54% as compared to 37–64% for the SVM). The ANN and SVM models generate more accurate flood quantiles for smaller return periods; however, for higher return periods, both the methods present a higher estimation error. The results of this study will help to recommend new AI-based RFFA methods in Australia. [ABSTRACT FROM AUTHOR]

There is a long history of coastal erosion caused by frequent storm surges in the coastal regions of Australia, which imposes great threats to communities and infrastructures alongside the beach. Old Bar Beach, New South Wales, Australia, is one such hotspot famous for its extreme coastal erosion. To apply remedial measures such as beach nourishment effectively and economically, estimating/reconstructing the subsurface hydrogeology over the coastal areas is essential. A geophysical tool such as a ground-penetrating radar (GPR) which works on the principle of reflecting electromagnetic (EM) waves, can be conveniently deployed to delineate the soil and rock profiling, water-table depth, bedrock depth, and the subsurface structural features. Here, DeepLabv3+ architecture based newly developed deep convolutional neural networks (DCNNs) were used to establish an inherent non-linear relationship between the GPR data and the EM wave velocity. The presented DCNNs have a lesser number of layers, a lesser number of trainable (learnable) parameters, a high convergence rate and, at the same time, achieve prediction accuracy comparable to that of well-established DeepLabv3+ networks, having high trainable parameters and a relatively low convergence rate. Here, firstly the DCNNs were trained and validated on small 1D datasets. Each dataset contains a 1D GPR trace and a corresponding EM velocity model. The DCNNs turned out to be quite promising in the 1D case, with training, validation, and testing accuracy of approximately 95%, 94%, and 95%, respectively. Secondly, 1D trained weights were applied to 2D synthetic GPR data for EM velocity prediction, and the accuracy of prediction achieved was approximately 95%. Seeing the excellent performance of the DCNNs in the 2D prediction case using 1D trained weights, a large amount of 1D synthetic datasets (approximately 1.2 million) were generated and gaussian noise was added to it to replicate the real field scenario. Thirdly, topographically corrected GPR data acquired over the Old Bar Beach were inverted using the DCNNs trained on 1.2 million 1D synthetic datasets to obtain the subsurface high-resolution, high-precision EM velocity, and ε r distribution information to understand the hydrogeology over the beach. The findings presented in this paper agree well with the previous hydrogeological studies carried out using GPR. Our findings show that DCNNs, along with GPR, can be successfully used in coastal environments for the quick and accurate hydrogeological investigation required for the implementation of coastal erosion mitigation methods such as beach nourishment. [ABSTRACT FROM AUTHOR]

This paper presents the results of a research project which investigated permeability prediction for the Precipice Sandstone of the Surat Basin. Machine learning techniques were used for permeability estimation based on multiple wireline logs. This information improves the prediction of CO2 injectivity in this formation. Well logs and core data were collected from 5 boreholes in the Surat Basin, where extensive core data and complete sets of conventional well logs exist for the Precipice Sandstone. Four different machine learning (ML) techniques, including Random Forest (RF), Artificial neural network (ANN), Gradient Boosting Regressor (GBR), and Support Vector Regressor (SVR), were independently trained with a wide range of hyper-parameters to ensure that not only is the best model selected, but also the right combination of model parameters is selected. Cross-validation for 20 different combinations of the seven available input logs was used for this study. Based on the performances in the validation and blind testing phases, the ANN with all seven logs used as input was found to give the best performance in predicting permeability for the Precipice Sandstone with the coefficient of determination (R2) of about 0.93 and 0.87 for the training and blind data sets respectively. Multi-regression analysis also appears to be a successful approach to calculate reservoir permeability for the Precipice Sandstone. Models with a complete set of well logs can generate reservoir permeability with R2 of more than 90%. [ABSTRACT FROM AUTHOR]

This paper presents the development of the Wavelet Artificial Neural Networks (WANN) model to forecast seasonal rainfall in Queensland, Australia, using the Inter-decadal Pacific Oscillation (IPO), Southern Oscillation Index (SOI), and Nino3.4 climate indices as predictors. Eight input sets with different combinations of predictive variables from 1908 to 2016 were considered to develop forecast models for ten selected rainfall stations in Queensland, Australia. The outcomes of WANN modeling are compared with Artificial Neural Networks (ANN). Moreover, the skillfulness of the WANN in comparison to the current climate prediction system used by the Australian Community Climate Earth-System Simulator–Seasonal (ACCESS–S) and climatology forecasts are investigated. Besides, the WANN predictions are compared with two other conventional approaches like autoregressive integrated moving average (ARIMA) and multiple linear regression (MLR) for further investigations. The comparisons indicated that the WANN achieves the lower average root mean square error (RMSE) in all the stations with 112.2mm compared to ANN with 178.9mm, ACCESS-S with 281.8mm, climatology prediction with 279.7mm, MLR with 195.1mm, and ARIMA with 187.7mm. The WANN seasonal rainfall forecasts are more accurate than the ANN, ACCESS-S, Climatology, MLR, and ARIMA by 37%, 60%, 53%, 42%, and 40%, respectively. It was also found that the ACCESS-S underestimates the extreme seasonal rainfall during the testing period up to 80%, while it is limited to 21% for the WANN among the selected stations. The results show that the WANN model outperforms the MLR, ARIMA, climatology, ACCESS-S, and ANN forecasts in all the selected stations. [ABSTRACT FROM AUTHOR]

The assessment of the stope stability is of great importance in the underground mine design process. The instability of stopes leads to serious economic and safety problems in mining operations. This paper aims to assist in the prediction of stope stability by making use of machine learning algorithms. For this purpose, a hybrid grid search-based Artificial Neural Network method is proposed. In the experiments, a total of 215 stope cases which have been collected from six underground mining operations located across Australia are used to prove the validity of the proposed method. Then, the performance of the proposed method is compared with k-nearest neighbour (kNN), naive Bayes (NB), support vector machine (SVM), decision tree (DT) and the stability graph method. Furthermore, several performance measures, such as accuracy, precision, specificity, recall, f-measure, and g-mean are considered during the performance comparison of the methods. The accuracy values are obtained as 91.63%, 81.86%, 77.21%, 76.74%, 73.95%, and 69.77% with the proposed hybrid grid search-based ANN method, SVM, DT, NB, kNN, and the stability graph method, respectively. The empirical results from the experiments indicated that the proposed hybrid approach outperformed the other aforementioned methods, which confirms that the proposed method is a useful tool to predict stope stability. [ABSTRACT FROM AUTHOR]

This paper investigates the expected results of the current COVID-19 outbreak to arrivals of Chinese tourists to the USA and Australia. The growing market share of Chinese tourism and the fact that the county was the first to experience the pandemic make China a suitable proxy for predictions on global tourism. We employ data from the 2003 SARS outbreak to train a deep learning artificial neural network named Long Short Term Memory (LSTM). The neural network is calibrated for the particulars of the current pandemic. Our findings, which are cross-validated using backtesting, suggest that recovery of arrivals to pre-crisis levels can take from 6 to 12 months and this can have significant adverse effects not only on the tourism industry but also on other sectors that interact with it. [ABSTRACT FROM AUTHOR]

• Develops a deep Bayesian survival model that integrates deep neural network and Bayesian inference into the survival analysis for rail useful lifetime modelling. • Demonstrates survival analysis is an effective way to handle censored data problems in rail useful lifetime modelling. • Conducts extensive evaluation to verify the effectiveness of this model, using Concordance index, root mean square error, and a proposed CW-index. • The estimated lifetimes obtained by this model are the closest to real lifetimes compared to other commonly used models. • This model can provide the confidence interval of the estimated lifetime, which is more useful than point estimates in decision-making and maintenance planning of railroad systems. Reliable estimation of rail useful lifetime can provide valuable information for predictive maintenance in railway systems. However, in most cases, lifetime data is incomplete because not all pieces of rail experience failure by the end of the study horizon, a problem known as censoring. Ignoring or otherwise mistreating the censored cases might lead to false conclusions. Survival approach is particularly designed to handle censored data for analysing the expected duration of time until one event occurs, which is rail failure in this paper. This paper proposes a deep Bayesian survival approach named BNN-Surv to properly handle censored data for rail useful lifetime modelling. The proposed BNN-Surv model applies the deep neural network in the survival approach to capture the non-linear relationship between covariates and rail useful lifetime. To consider and quantify uncertainty in the model, Monte Carlo dropout, regarded as the approximate Bayesian inference, is incorporated into the deep neural network to provide the confidence interval of the estimated lifetime. The proposed approach is implemented on a four-year dataset including track geometry monitoring data, track characteristics data, various types of defect data, and maintenance and replacement (M&R) data collected from a section of railway tracks in Australia. Through extensive evaluation, including Concordance index (C-index) and root mean square error (RMSE) for evaluating model performance, as well as a proposed CW-index for evaluating uncertainty estimations, the effectiveness of the proposed approach is confirmed. The results show that, compared with other commonly used models, the proposed approach can achieve the best concordance index (C-index) of 0.80, and the estimated rail useful lifetimes are closer to real lifetimes. In addition, the proposed approach can provide the confidence interval of the estimated lifetime, with a correct coverage of 81% of the actual lifetime when the confidence interval is 1.38, which is more useful than point estimates in decision-making and maintenance planning of railroad systems. [ABSTRACT FROM AUTHOR]

This paper presents development of an artificial intelligence (AI)‐based model, genetic expression programming (GEP) to predict long‐term streamflow using large‐scale climate drivers as predictors. GEP is chosen over artificial neural networks (ANNs) model, as ANN is a black‐box model, whereas GEP is able to explain the developed forecast models with mathematical expressions. As a case study, 12 streamflow measuring stations were selected from four different regions of New South Wales (NSW) in eastern Australia. A number of climate indices, Pacific Decadal Oscillation (PDO), Indian Ocean Dipole (IOD), El Nino Southern Oscillation (ENSO) and ENSO Modoki index (EMI), were selected as candidate predictors based on the findings of some preliminary studies. Higher predictabilities of the GEP‐based models are evident from the Pearson correlation (r) values ranging between 0.57 and 0.97, which are mostly about twice the values achieved by multiple linear regression (MLR) models in the preliminary study. Performances of the developed models were assessed using standard statistical measures such as root relative squared error (RRSE), relative absolute error (RAE), root mean square error (RMSE), mean absolute error (MAE), Nash–Sutcliffe efficiency (NSE) and Pearson correlation (r) values. The developed models are able to predict spring streamflow up to 5 months in advance with significantly high correlation values. [ABSTRACT FROM AUTHOR]

The main objective of this paper is to present a systematic analysis and review of the state of the art regarding the prediction of absenteeism and temporary incapacity using machine learning techniques. Moreover, the main contribution of this research is to reveal the most successful prediction models available in the literature. A systematic review of research papers published from 2010 to the present, related to the prediction of temporary disability and absenteeism in available in different research databases, is presented in this paper. The review focuses primarily on scientific databases such as Google Scholar, Science Direct, IEEE Xplore, Web of Science, and ResearchGate. A total of 58 articles were obtained from which, after removing duplicates and applying the search criteria, 18 have been included in the review. In total, 44% of the articles were published in 2019, representing a significant growth in scientific work regarding these indicators. This study also evidenced the interest of several countries. In addition, 56% of the articles were found to base their study on regression methods, 33% in classification, and 11% in grouping. After this systematic review, the efficiency and usefulness of artificial neural networks in predicting absenteeism and temporary incapacity are demonstrated. The studies regarding absenteeism and temporary disability at work are mainly conducted in Brazil and India, which are responsible for 44% of the analyzed papers followed by Saudi Arabia, and Australia which represented 22%. ANNs are the most used method in both classification and regression models representing 83% and 80% of the analyzed works, respectively. Only 10% of the literature use SVM, which is the less used method in regression models. Moreover, Naïve Bayes is the less used method in classification models representing 17%. [ABSTRACT FROM AUTHOR]

This paper presents and evaluates artificial neural network (ANN) models used for macrocell path-loss prediction. Measurement data obtained by utilizing the IS-95 pilot signal from a commercial code-division multiple-access (CDMA) mobile network in rural Australia are used to train and evaluate the models. A simple neuron model and feed-forward networks with different numbers of hidden layers and neurons are evaluated regarding their training time, prediction accuracy, and generalization properties. Furthermore, different backpropagation training algorithms, such as gradient descent and Levenberg--Marquardt, are evaluated. The artificial neural network inputs are chosen to be distance to base station, parameters easily obtained from terrain path profiles, land usage, and vegetation type and density near the receiving antenna. The path-loss prediction results obtained by using the ANN models are evaluated against different versions of the semi-terrain based propagation model Recommendation ITU-R P.1546 and the Okumura--Hata model. The statistical analysis shows that a non-complex ANN model performs very well compared with traditional propagation models with regard to prediction accuracy, complexity, and prediction time. The average ANN prediction results were 1) maximum error: 22 dB; 2) mean error: 0 dB; and 3) standard deviation: 7 dB. A multilayered feedforward network trained using the standard backpropagation algorithm was compared with a neuron model trained using the Levenberg--Marquardt algorithm. It was found that the training time decreases from 150 000 to 10 iterations, while the prediction accuracy is maintained. [ABSTRACT FROM AUTHOR]

Purpose: Improving valuation accuracy, especially for sale and acquisition purposes, remains one of the key targets of the global real estate research agenda. Among other recommendations, it has been argued that the use of technology-based advanced valuation methods can help to narrow the gap between asset valuations and actual sale prices. The purpose of this paper is to investigate the property valuation methods being adopted by Australian valuers and the factors influencing their level of awareness and adoption of the methods. Design/methodology/approach: An online questionnaire survey was conducted to elicit information from valuers practising in Australia. They were asked to indicate their level of awareness and adoption of the different property valuation methods. Their response was analysed using frequency distribution, χ2 test and mean score ranking. Findings: The results show that the traditional methods of valuation, namely, comparative, investment and residual, are the most adopted methods by the Australian valuers, while advanced valuation methods are seldom applied in practice. The results confirm that professional bodies, sector of practice and educational institutions are the three most important drivers of awareness and adoption of the advanced valuation methods. Practical implications: There is a need for all the property valuation stakeholders to synergise and transform the property valuation practice in a bid to promote the awareness and adoption of advanced valuation methods, (e.g. hedonic pricing model, artificial neural network, expert system, fuzzy logic system, etc.) among valuers. These are all technology-based methods to improve the efficiency in the prediction process, and the valuer still needs to input reliable transaction data into the systems. Originality/value: This study provides a fresh and most recent insight into the current property valuation methods adopted in practice by valuers practising in Australia. It identifies that the advanced valuation methods could supplement the traditional valuation methods to achieve good practice standard for improving the professional valuation practice in Australia so that the valuation profession can meet the industry's expectations. [ABSTRACT FROM AUTHOR]

Microgrids are extensively integrated into electrical systems due to their many technical, economic, and environmental advantages. However, they encounter a challenge as they experience high-frequency fluctuations caused by the stochastic nature of renewable energy generation, electric loads, and the presence of Electric Vehicles (EVs). Therefore, various techniques, algorithms, and controllers have been introduced to ensure effective Load Frequency Control (LFC) and maintain a stable power system in microgrids. These methods aim to ensure that the system's frequency remains stable and within an acceptable range, especially when faced with changing load demands and other factors. This paper presents a novel enhanced control approach, Particle Swarm Optimization-Trained Artificial Neural Network (PSO-TANN), to optimize the load frequency model of a microgrid with vehicle-to-grid integration. The results are then compared under various scenarios, including renewable energy integration, EV charging and discharging dynamics, and varying load demands. The comparative analysis involves assessing the performance of the conventional Proportional–Integral–Derivative (PID) controller, the PSO-PID controller, and the newly proposed controlling technique. The suggested controller attains 99.904% efficiency with a negligible mean squared error of 1.1112 × 10−7, decreasing the integrated time absolute error to 1.0 × 10−4. It shows rapid response, precise targeting, and quick peak output ability, with marginal overshoot and undershoot, and a transient time of 28.5626 s, efficiently controlling microgrid frequency. Stability analysis validates the effectiveness of the proposed PSO-TANN controller in ensuring stability within the microgrid's LFC system during uncertainties and disturbances. This establishes resilience, diminishes settling time, and maintains reliable performance while controlling frequency. • An investigation into an islanded microgrid's Load Frequency Control (LFC) model framework has been conducted, comprising diesel generator, solar PV system, wind turbine, and EVs operating in vehicle-to-grid mode to supply power to electric loads, consequently impacting the frequency dynamics of the microgrid system. • A novel enhanced control approach, Particle Swarm Optimization-Trained Artificial Neural Network (PSO-TANN) has been implemented to optimize the load frequency model of a microgrid. • An extensive comparative analysis with other controllers, including PID and PSO-PID, is conducted across diverse scenarios involving renewable energy integration and EV charging and discharging dynamics under various load disturbances to reduce settling time, control system frequency, and ensure reliable performance. • The stability analysis validates the efficiency of the proposed PSO-TANN controller in ensuring reliability during uncertainties and disturbances, maintaining overall robustness and stability of the LFC system within the microgrid. [ABSTRACT FROM AUTHOR]

Recent failures of bridges worldwide highlight the need for robust structural health surveillance systems to provide alerts for early damage and prevent severe loss. Structural health monitoring (SHM) continues to evolve and gain wide adoption among asset managers to mitigate catastrophic failure risks of structural systems. With the prolific integration of SHM with artificial intelligence and internet of things systems becoming prevalent, several data-driven approaches have gained popularity in SHM. However, it is challenging to obtain ground-truth-labelled damage samples for supervised machine learning methods. A class of shallow machine learning methods known as unsupervised methods (e.g. the one-class support vector machine (OCSVM)), which only uses unlabelled data from a healthy state, has been shown to achieve promising results. However, shallow machine learning methods require laborious expert-crafted feature inputs for optimal performance. Inspired by the automatic feature-extraction ability of deep neural networks, unsupervised methods (deep learning models trained without supervisory ground truth labels) such as autoencoder (AE)-based methods were explored to detect structural damage. Results obtained on both real-world data from Sydney Harbour Bridge, one of Australia's iconic structures, and data collected from laboratory specimens demonstrate the effectiveness of the AE-based method over conventional shallow methods such as the OCSVM. [ABSTRACT FROM AUTHOR]

This paper focuses on the development and testing of the genetic algorithm (GA)-based regional flood frequency analysis (RFFA) models for eastern parts of Australia. The GA-based techniques do not impose a fixed model structure on the data and can better deal with nonlinearity of the input and output relationship. These nonlinear techniques have been applied successfully in many hydrologic problems; however, there have been only limited applications of these techniques in RFFA problems, particularly in Australia. A data set comprising of 452 stations is used to test the GA for artificial neural networks (ANN) optimization known as GAANN. The results from GAANN were compared with the results from back-propagation for ANN optimization known as BPANN. An independent testing shows that both the GAANN and BPANN methods are quite successful in RFFA and can be used as alternative methods to check the validity of the traditional linear models such as quantile regression technique. [ABSTRACT FROM AUTHOR]

This paper focuses on predicting Australia's outbound international airline passenger demand using an artificial neural network (ANN) modelling method. The modelling in the study was based on annual data for the period 1993 to 2016. The model was developed using the input parameters of world GDP, world population growth, world jet fuel prices, world air fares (proxy for air travel cost), Australia's tourism attractiveness, outbound flights, Australia's unemployment levels, the Australian and United States foreign exchange rate and three dummy variables (Sydney Olympics, 9/11 and the 2006 Commonwealth Games). The artificial neural network (ANN) used multi-layer perceptron (MLP) architecture that compromised a multi-layer feed-forward network and the sigmoid and linear functions were used as activation functions with the feed forward-back propagation algorithm. The ANN was applied during training, testing and validation and had 8 inputs, 1 neurons in the hidden layers and 1 neuron in the output layer. The data was randomly divided into three data sets; training, testing and model validation. The best-fit model was selected according to four goodness-of-fit measures: mean absolute error (MAE), mean square error (MSE), root mean square errors (RMSE), AND mean absolute percentage errors (MAPE). The highest R-value obtained from the ANN model is 0.99733, demonstrating that the ANN provided a high predictive capability. [ABSTRACT FROM AUTHOR]

The accuracy of photovoltaic (PV) power forecasting techniques relies not only on high-quality spatiotemporal data but also on an efficient feature-mining methodology. In this study, a spatiotemporal power forecasting model based on the broad learning system (BLS) and the improved backtracking search optimization algorithm (IBSOA) is proposed. The objective is to enhance the accuracy of PV power predictions while reducing the time-intensive training process associated with an extensive set of broad learning system parameters. The spatiotemporal attributes of historical data from multiple PV sites are clustered using a self-organizing map. The clustering analysis explores the spatiotemporal correlation among five photovoltaic (PV) power stations for each season between 2017 and 2018. Subsequently, the IBSOA is employed to optimize the hyperparameters of the BLS model, particularly the mapping and enhancement nodes. By utilizing hyperparameter optimization, a BSOA-based broad learning model is introduced to achieve superior accuracy. The results are assessed using the proposed method in comparison with three popular optimization algorithms: 1) genetic algorithm (GA), 2) bird swarm algorithm (BSA), and 3) backtracking search optimization algorithm (BSOA). All scenarios are validated and compared using PV plant data from the DKA center in Australia. The root-mean-square error (RMSE) indicators of the proposed prediction method are consistently lower than the worst-case scenario in each season, decreasing by 3.2283 kW in spring, 3.9159 kW in summer, 1.3425 kW in autumn, and 1.4058 kW in winter. Similarly, the mean absolute percentage error (MAPE) exhibits a reduction compared to the worst case, with a decreases of 0.882% in spring, 1.2399% in summer, 1.803% in autumn, and 1.087% in winter. The comprehensive results affirm that the proposed method surpasses alternative optimization techniques, delivering high-quality power forecasts for the given case study. [ABSTRACT FROM AUTHOR]

Energy efficiency of domestic refrigerators has markedly improved over recent years. Despite these improvements, low-income households in Australia still tend to own older and less energy-efficient refrigerators. Replacement of old refrigerators with energy-efficient appliances presents an important opportunity to reduce both greenhouse gas emissions and the electricity bills of households for sustainable development. This paper presents the measurement and verification of energy savings from a sample of 16 household refrigerator replacements under the Appliance Replacement Offer of the New South Wales Government, Australia. In-home monitoring and laboratory testing were conducted for both old and new refrigerators. A key challenge for this study was to estimate the energy savings of the replacement in each household based on limited in-home monitoring data. The annual energy savings from the refrigerator replacement in the households were estimated using the artificial neural networks (ANN) models for indoor air temperature prediction and the empirical models for the refrigerator energy consumption estimation. The experimental and estimation results both showed that most of the households achieved substantial energy savings from the replacement. The average annual energy cost savings and the average payback period of the refrigerator replacements were 53% and 4.3 years, respectively. It was also found that the energy savings could be compromised if the size of the new refrigerator was much larger than the old unit or if the efficiency of the old refrigerator was reasonably good. • Energy savings from 16 household refrigerator replacements were presented. • Both in-home monitoring and laboratory test were conducted to determine energy savings. • Artificial neural network models were used to predict indoor air temperature. • The average annual energy cost saving of the refrigerator replacements was 53%. [ABSTRACT FROM AUTHOR]

The exacerbation of wildfires, attributed to the effects of climate change, presents substantial risks to ecological systems, infrastructure, and human well-being. In the context of the Sustainable Development Goals (SDGs), particularly those related to climate action, prioritizing the assessment and management of the occurrence and intensity of extensive wildfires is of utmost importance. In recent times, there has been a significant increase in the frequency and severity of widespread wildfires worldwide, affecting several locations, including Australia, Italy, and the United States of America. The presence of complex phenomena marked by limited predictability leads to significant negative impacts on biodiversity and human lives. The utilization of satellite-derived data with neural networks, such as convolutional neural networks (CNNs), is a potentially advantageous approach for augmenting the monitoring capabilities of wildfires. This research examines the generalization capability of four neural network models, namely the fully connected (FC), one-dimensional (1D) CNN, two-dimensional (2D) CNN, and three-dimensional (3D) CNN model. Each model's performance, as measured by accuracy, recall, and F1 scores, is assessed through K-fold cross-validation. Subsequently, T-statistics and p-values are computed based on these metrics to conduct a statistical comparison among the different models, allowing us to quantify the degree of similarity or dissimilarity between them. By using training data from Australia and Sicily, the performances of the trained model are evaluated on the test dataset from Oregon. The results are promising, with cross-validation on the training dataset producing mean precision, recall, and F1 scores ranging between approximately 0.97 and 0.98. Especially, the fully connected model has superior generalization capabilities, whilst the 3D CNN offers more refined and less distorted classifications. However, certain issues, such as false fire detection and confusion between smoke and shadows, persist. The aforementioned methodologies offer significant perspectives on the capabilities of neural network technologies in supporting the detection and management of wildfires. These approaches address the crucial matter of domain transferability and the associated dependability of predictions in new regions. This study makes a valuable contribution to the ongoing efforts in climate change by assisting in monitoring and managing wildfires. [ABSTRACT FROM AUTHOR]

Mean sea level rise is a significant emerging risk from climate change. This research paper is based on the use of artificial intelligence models to assess and predict the trend on mean sea level around northern Australian coastlines. The study uses sea-level times series from four sites (Broom, Darwin, Cape Ferguson, Rosslyn Bay) to make the prediction. Multivariate adaptive regression splines (MARS) and artificial neural network (ANN) algorithms have been implemented to build the prediction model. Both models show high accuracy (R2 > 0.98) and low error values (RMSE < 27%) overall. The ANN model showed slightly better performance compared to MARS over the selected sites. The ANN performance was further assessed for modelling storm surges associated with cyclones. The model reproduced the surge profile with the maximum correlation coefficients ~0.99 and minimum RMS errors ~4 cm at selected validating sites. In addition, the ANN model predicted the maximum surge at Rosslyn Bay for cyclone Marcia to within 2 cm of the measured peak and the maximum surge at Broome for cyclone Narelle to within 7 cm of the measured peak. The results are comparable with a MARS model previously used in this region; however, the ANN shows better agreement with the measured peak and arrival time, although it suffers from slightly higher predictions than the observed sea level by tide gauge station. [ABSTRACT FROM AUTHOR]

Short-term load forecasting is an essential instrument in power system planning, operation, and control. Many operating decisions are based on load forecasts, such as dispatch scheduling of generating capacity, reliability analysis, and maintenance planning for the generators. Overestimation of electricity demand will cause a conservative operation, which leads to the start-up of too many units or excessive energy purchase, thereby supplying an unnecessary level of reserve. On the other hand, underestimation may result in a risky operation, with insufficient preparation of spinning reserve, causing the system to operate in a vulnerable region to the disturbance. [ABSTRACT FROM PUBLISHER]

Load forecasting is a challenging task in power markets that require attention in generating accurate and stable load to deal with planning and management strategies. In past few years, several intelligence-based models have been introduced for precise load forecast. Among them, artificial neural network (ANN) seems more effective and capable to handle the non-linear behavior of load and generates an accurate forecast. However, it suffers from overfitting problem thus reducing the accuracy of load forecasts. To overcome this problem, a hybrid methodology namely ANN-IEAMCGM-R, for short-term load forecast is proposed in this paper. ANN is integrated with an enhanced evolutionary algorithm (IEAMCGM-R) to find optimal network weights. This evolutionary algorithm is composed of improved environmental adaptation method with real parameters (IEAM-R) and our proposed Controlled Gaussian Mutation (CGM) method to bring greater diversity within the population resulting in a higher convergence of solutions. The electric load data from the New England Power Pool (NEPOOL, ISO New England) and Australian Energy Market Operator (New South Wales (NSW), Australia) have been used to illustrate the efficacy of the proposed hybrid methodology. Results show that the proposed hybrid methodology generates higher accuracy than other state-of-the-art algorithms. • Proposal of Controlled Gaussian Mutation for enhancing the diversity of IEAM-R. • Proposal of enhanced optimization algorithm, namely IEAMCGM-R. • A hybrid methodology is presented for load forecasting. • Experimental results show that ANN-IEAMCGM-R generates least forecasting error. • Statistical test results ensures the statistical signi_cance of forecasted load. [ABSTRACT FROM AUTHOR]

Abstract To support alternative forms of energy resources, the prediction of global incident solar radiation (I rad) is critical to establish the efficacy of solar energy resources as a free and clean energy, and to identify and screen solar powered sites. Solar radiation data for construction of energy feasibility studies are not available in many locations due to the absence of meteorological stations, especially in remote or regional sites. To surmount the challenge in solar energy site identification, the universally gridded data integrated into predictive models used to generate reliable I rad forecasts can be considered as a viable medium for future energy utilization. The objective of this paper is to review, develop and evaluate a suite of machine learning (ML) models based on the artificial neural network (ANN) versus several other kinds of data-driven models such as support vector regression (SVR), Gaussian process machine learning (GPML) and genetic programming (GP) models for the prediction of daily I rad generated through the European Centre for Medium Range Weather Forecasting (ECMWF) Reanalysis fields. The performance of the ML models are benchmarked against several statistical tools: auto regressive moving integrated average (ARIMA), Temperature Model (TM), Time series and Fourier Series (TSFS) models. To train these models, 87 different predictor variables from the ERA-Interim reanalysis dataset (01-January-1979 to 31-December-2015) were extracted for 5 solar-rich metropolitan sites (i.e., Brisbane, Gold Coast, Sunshine Coast, Ipswich and Toowoomba, Australia) targeted against surface level I rad available from the measured Scientific Information for Land Owners dataset. For daily forecast models, a total of the 20 most important predictors related to the I rad dataset were screened with nearest component analysis: " fsrnca " feature selection, and partitioned into training (70%), validation (15%) and testing (15%) sets for model design. To benchmark the ANN, TSFS and TM models were developed with Fourier series and regression analysis, respectively and the statistical performance was benchmarked with root mean square error (RMSE), mean absolute error (MAE), Nash-Sutcliffe efficiency (E NS), Willmott's Index (WI), Mean Bias error (MBE), Legates and McCabe Index (E 1), and relative MAE , RMSE and diagnostic plots. The performance of ANN was significantly better than the other models (SVR, GPML, GP, TM), resulting in lower RMSE (1.715–2.27 MJm−2/day relative to 2.14–5.90 MJm−2/day), relative RMSE (9.07–12.47 vs 10.98–29.15), relative RMAE (7.97–11.74 vs 9.27–33.96) and larger WI, E NS and E 1 (0.938–0.967 vs. 0.462–0.955, 0.935–0.872 vs. 0.355–0.915, 0.672–0.783 vs. 0.252–0.740). Additionally, models assessed with predictors grouped into El Niño, La Niña and the positive, negative and neutral periods of Indian Ocean Dipole, affirmed the merits of ANN model (RRMSE ≤ 11%). Seasonal analysis showed that ANN was an elite tool over SVR, GPML and GP for I rad prediction. The study concludes that an ANN approach integrated with ECMWF fields, incorporating physical interactions of I rad with atmospheric data, is an efficacious alternative to forecast solar energy and assist with energy modelling for solar-rich sites that have diverse climatic conditions to further support clean energy utilization. Graphical abstract Image 1 Highlights • ANN, SVR, GPML and GP models were designed for prediction of daily solar radiation with reanalysis data. • Determinstic model ARIMA, temperature model and time series with Fourier transform model was used for benchmarking. • Nearest component analysis (fsrnca) was used for feature selection to get optimum inputs. • ANN outperform all ML model and deterministic models. • Solar energy potential is assessed for 5 cities of Queensland, Australia. [ABSTRACT FROM AUTHOR]

In this paper, a deep learning long-short-term memory (LSTM) method is applied to the forecasting of the critical frequency of the ionosphere F2 layer (foF2). Hourly values of foF2 from 10 ionospheric stations in China and Australia (based on availability) from 2006 to 2019 are used for training and verifying. While 2015 and 2019 are exclusive for verifying the forecasting accuracy. The inputs of the LSTM model are sequential data for the previous values, which include local time (LT), day number, solar zenith angle, the sunspot number (SSN), the daily F10.7 solar flux, geomagnetic the Ap and Kp indices, geographic coordinates, neutral winds, and the observed value of foF2 at the previous moment. To evaluate the forecasting ability of the deep learning LSTM model, two different neural network forecasting models: a back-propagation neural network (BPNN) and a genetic algorithm optimized backpropagation neural network (GABP) were established for comparative analysis. The foF2 parameters were forecasted under geomagnetic quiet and geomagnetic disturbed conditions during solar activity maximum (2015) and minimum (2019), respectively. The forecasting results of these models are compared with those of the international reference ionosphere model (IRI2016) and the measurements. The diurnal and seasonal variations of foF2 for the 4 models were compared and analyzed from 8 selected verification stations. The forecasting results reveal that the deep learning LSTM model presents the optimal performance of all models in forecasting the time series of foF2, while the IRI2016 model has the poorest forecasting performance, and the BPNN model and GABP model are between two of them. [ABSTRACT FROM AUTHOR]

This paper presents the results from a study on the application of an artificial neural network (ANN) model for regional flood frequency analysis (RFFA). The study was conducted using stream flow data from 88 gauging stations across New South Wales (NSW) in Australia. Five different models consisting of three to eight predictor variables (i.e., annual rainfall, drainage area, fraction forested area, potential evapotranspiration, rainfall intensity, river slope, shape factor and stream density) were tested. The results show that an ANN model with a higher number of predictor variables does not always improve the performance of RFFA models. For example, the model with three predictor variables performs considerably better than the models using a higher number of predictor variables, except for the one which contains all the eight predictor variables. The model with three predictor variables exhibits smaller median relative error values for 2- and 20-year return periods compared to the model containing eight predictor variables. However, for 5-, 10-, 50- and 100-year return periods, the model with eight predictor variables shows smaller median relative error values. The proposed ANN modelling framework can be adapted to other regions in Australia and abroad. [ABSTRACT FROM AUTHOR]

Forecast models of solar radiation incorporating cloud effects are useful tools to evaluate the impact of stochastic behaviour of cloud movement, real-time integration of photovoltaic energy in power grids, skin cancer and eye disease risk minimisation through solar ultraviolet (UV) index prediction and bio-photosynthetic processes through the modelling of solar photosynthetic photon flux density (PPFD). This research has developed deep learning hybrid model (i.e., CNN-LSTM) to factor in role of cloud effects integrating the merits of convolutional neural networks with long short-term memory networks to forecast near real-time (i.e., 5-min) PPFD in a sub-tropical region Queensland, Australia. The prescribed CLSTM model is trained with real-time sky images that depict stochastic cloud movements captured through a total sky imager (TSI-440) utilising advanced sky image segmentation to reveal cloud chromatic features into their statistical values, and to purposely factor in the cloud variation to optimise the CLSTM model. The model, with its competing algorithms (i.e., CNN, LSTM, deep neural network, extreme learning machine and multivariate adaptive regression spline), are trained with 17 distinct cloud cover inputs considering the chromaticity of red, blue, thin, and opaque cloud statistics, supplemented by solar zenith angle (SZA) to predict short-term PPFD. The models developed with cloud inputs yield accurate results, outperforming the SZA-based models while the best testing performance is recorded by the objective method (i.e., CLSTM) tested over a 7-day measurement period. Specifically, CLSTM yields a testing performance with correlation coefficient r = 0.92, root mean square error RMSE = 210.31 μ mol of photons m−2 s−1, mean absolute error MAE = 150.24 μ mol of photons m−2 s−1, including a relative error of RRMSE = 24.92% MAPE = 38.01%, and Nash Sutcliffe's coefficient ENS = 0.85, and Legate and McCabe's Index LM = 0.68 using cloud cover in addition to the SZA as an input. The study shows the importance of cloud inclusion in forecasting solar radiation and evaluating the risk with practical implications in monitoring solar energy, greenhouses and high-value agricultural operations affected by stochastic behaviour of clouds. Additional methodological refinements such as retraining the CLSTM model for hourly and seasonal time scales may aid in the promotion of agricultural crop farming and environmental risk evaluation applications such as predicting the solar UV index and direct normal solar irradiance for renewable energy monitoring systems. [ABSTRACT FROM AUTHOR]

This paper presents a new hybrid approach for Global Solar Radiation (GSR) prediction problems, based on deep learning approaches. Predictive models are useful ploys in solar energy industries to optimize the performance of photovoltaic power systems. Specifically, in this work we develop a new 4-phase hybrid CXGBRFR framework, which includes a deep learning Convolutional Neural Network, an Extreme Gradient Boosting with Random Forest Regression, and a Harris Hawks Optimization for initial feature selection. The proposed system has been tested in a problem of daily GSR prediction at six solar farms in Australia. Data from three global climate models (GCM) (CSIRO-BOM ACCESS1-0, MOHC Hadley-GEM2-CC and MRI MRI-CGCM3) have been considered as predictive (input) variables for the proposed approach. The variables from these GCMs contain enough information to obtain an accurate prediction of the GSR at each solar farm. The performance of the proposed approach is compared against different deep and shallows learning approaches: Deep Belief Network, Deep Neural Network, Artificial Neural Network, Extreme Learning Machine and Multivariate Auto-Regressive Spline models. We show that the proposed approach exhibits an excellent performance in GSR prediction, against all alternative approaches in all solar farms considered. [Display omitted] • We develop hybrid deep learning model to predict solar radiation • We optimize the algorithm with Harris Hawks, gradient boosting and random forest • We adopt global climate model and observed predictors as learning model inputs • We build an expert system with relevant knowledge in context of solar application • Our methods benefit solar photovoltaic power monitoring in solar engineering problems. [ABSTRACT FROM AUTHOR]

Cocoa is an important commodity crop, not only to produce chocolate, one of the most complex products from the sensory perspective, but one that commonly grows in developing countries close to the tropics. This paper presents novel techniques applied using cover photography and a novel computer application (VitiCanopy) to assess the canopy architecture of cocoa trees in a commercial plantation in Queensland, Australia. From the cocoa trees monitored, pod samples were collected, fermented, dried, and ground to obtain the aroma profile per tree using gas chromatography. The canopy architecture data were used as inputs in an artificial neural network (ANN) algorithm, with the aroma profile, considering six main aromas, as targets. The ANN model rendered high accuracy (correlation coefficient (R) = 0.82; mean squared error (MSE) = 0.09) with no overfitting. The model was then applied to an aerial image of the whole cocoa field studied to produce canopy vigor, and aroma profile maps up to the tree-by-tree scale. The tool developed could significantly aid the canopy management practices in cocoa trees, which have a direct effect on cocoa quality. [ABSTRACT FROM AUTHOR]

Solar energy predictive models designed to emulate the long-term (e.g., monthly) global solar radiation (GSR) trained with satellite-derived predictors can be employed as decision tenets in the exploration, installation and management of solar energy production systems in remote and inaccessible solar-powered sites. In spite of a plethora of models designed for GSR prediction, deep learning, representing a state-of-the-art intelligent tool, remains an attractive approach for renewable energy exploration, monitoring and forecasting. In this paper, algorithms based on deep belief networks and deep neural networks are designed to predict long-term GSR. Deep learning algorithms trained with publicly-accessible Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data are tested in Australia's solar cities to predict the monthly GSR: single hidden layer and ensemble models. The monthly-scale MODIS-derived predictors (2003–2018) are adopted, with 15 diverse feature selection approaches including a Gaussian Emulation Machine for sensitivity analysis used to select optimal MODIS-predictor variables to simulate GSR against ground-truth values. Several statistical score metrics are adopted to comprehensively verify surface GSR simulations to ascertain the practicality of deep belief and deep neural networks. In the testing phase, deep learning models generate significantly lower absolute percentage bias (≤3%) and high Kling–Gupta efficiency (≥97.5%) values compared to the single hidden layer and ensemble model. This study ascertains that the optimal MODIS input variables employed in GSR prediction for solar energy applications can be relatively different for diverse sites, advocating a need for feature selection prior to the modelling of GSR. The proposed deep learning approach can be adopted to identify solar energy potential proactively in locations where it is impossible to install an environmental monitoring data acquisition instrument. Hence, MODIS and other related satellite-derived predictors can be incorporated for solar energy prediction as a strategy for long-term renewable energy exploration. [ABSTRACT FROM AUTHOR]

Wildfires not only severely damage the natural environment and global ecological balance but also cause substantial losses to global forest resources and human lives and property. Unprecedented fire events such as Australia's bushfires have alerted us to the fact that wildfire prediction is a critical scientific problem for fire management. Therefore, robust, long‐lead models and dynamic predictions of wildfire are valuable for global fire prevention. However, despite decades of effort, the dynamic, effective, and accurate prediction of wildfire remains problematic. There is great uncertainty in predicting the future based on historical and existing spatiotemporal sequence data, but with advances in deep learning algorithms, solutions to prediction problems are being developed. Here, we present a dynamic prediction model of global burned area of wildfire employing a deep neural network (DNN) approach that produces effective wildfire forecasts based on historical time series predictors and satellite‐based burned area products. A hybrid DNN that combines long short‐term memory and a two‐dimensional convolutional neural network (CNN2D‐LSTM) was proposed, and CNN2D‐LSTM model candidates with four different architectures were designed and compared to construct the optimal architecture for fire prediction. The proposed model was also shown to outperform convolutional neural networks (CNNs) and the fully connected long short‐term memory (FcLSTM) approach using the refined index of agreement and evaluation metrics. We produced monthly global burned area spatiotemporal prediction maps and adequately reflected the seasonal peak in fire activity and highly fire‐prone areas. Our combined CNN2D‐LSTM approach can effectively predict the global burned area of wildfires 1 month in advance and can be generalized to provide seasonal estimates of global fire risk. [ABSTRACT FROM AUTHOR]

Measuring the predictability and complexity of 2D data (image) series using entropy is an essential tool for evaluation of systems' irregularity and complexity in remote sensing and geophysical mapping. However, the existing methods have some drawbacks related to their strong dependence on method parameters and image rotation. To overcome these difficulties, this study proposes a new method for estimating two-dimensional neural network entropy (NNetEn2D) for evaluating the regularity or predictability of images using the LogNNet neural network model. The method is based on an algorithm for converting a 2D kernel into a 1D data series followed by NNetEn2D calculation. An artificial test image was created for the study. We demonstrate the advantage of using circular instead of square kernels through comparison of the invariance of the NNetEn2D distribution after image rotation. Highest robustness was observed for circular kernels with a radius of R = 5 and R = 6 pixels, with a NNetEn2D calculation error of no more than 10%, comparable to the distortion of the initial 2D data. The NNetEn2D entropy calculation method has two main geometric parameters (kernel radius and its displacement step), as well as two neural network hyperparameters (number of training epochs and one of six reservoir filling techniques). We evaluated our method on both remote sensing and geophysical mapping images. Remote sensing imagery (Sentinel-2) shows that brightness of the image does not affect results, which helps keep a rather consistent appearance of entropy maps over time without saturation effects being observed. Surfaces with little texture, such as water bodies, have low NNetEn2D values, while urban areas have consistently high values. Application to geophysical mapping of rocks to the northwest of southwest Australia is characterized by low to medium entropy and highlights aspects of the geology. These results indicate the success of NNetEn2D in providing meaningful entropy information for 2D in remote sensing and geophysical applications. [ABSTRACT FROM AUTHOR]

Accurate atmospheric correction (AC) is a prerequisite for quantitative ocean colour remote sensing and remains a challenge in particular over coastal waters. Commonly AC algorithms are validated by establishing a mean retrieval error from match-up analysis, which compares the satellite-derived surface reflectance with concurrent ground radiometric observations. Pixel-based reflectance uncertainties however, are rarely provided by AC algorithms and those for the operational Ocean and Land Colour Instrument (OLCI) marine reflectance product are not yet recommended for use. AC retrieval errors and uncertainties directly determine the quality with which ocean colour products can be estimated from the marine surface reflectance. Increasingly there is also the need for reflectance uncertainty products to be used as data assimilation inputs into biogeochemical models. This paper describes the development of a new coastal AC algorithm for Sentinel-3 OLCI that provides pixel-based estimation of the inherent model inversion uncertainty and sensor noise propagation. The algorithm is a full-spectral model-based inversion of radiative transfer (RT) simulations in a coupled atmosphere–ocean system using an ensemble of artificial neural networks (ANN) that were initialized differently during the training process, but composed of the same network architecture. The algorithm has been validated against in-situ radiometric observations across a wide range of optical water types, and has been compared with the latest EUMETSAT operational Level 2 processor IPF-OL-2 v7.01. In this analysis we found that the ensemble ANN showed improved performance over the operational Level 2 processor with a band-averaged (412–708 nm) mean absolute percentage error (MAPE) of 16% compared to 37% and a four-times lower band-averaged bias of -0.00045 sr-1. In the ensemble inversion process we account for three uncertainty components: (1) the total model variance that describes the variance of the data from the different ANNs, (2) the prediction variance of the mean, which is based on calculations of the RT simulations and (3) the instrument noise variance of the mean by propagating the OLCI spectral signal-to-noise ratios (SNR). To study algorithm performance and to quantify the contribution of the different uncertainty components to the total uncertainty, we applied the algorithm to an optically complex full resolution (FR) test scene covering coastal waters of the Great Barrier Reef, Australia. The uncertainties associated with the instrument noise variance were found to be two orders of magnitude lower than the uncertainty components of the prediction and total model variances. The overall largest uncertainty component in our uncertainty framework is attributed to the total model inversion error from averaging the responses of the slightly different adapted networks in the ensemble. The algorithm is made publicly available as a Python/C plugin for the Sentinel Application Platform (SNAP). • An ANN AC-algorithm that provides pixel-based uncertainty estimation was developed. • The algorithm shows advanced performance compared to the operational processor. • An analysis of the uncertainty components for different water types is provided. • Uncertainty maps provide a useful tool for future algorithm improvement. • The algorithm is publicly available as a plugin for the Sentinel Application Platform. [ABSTRACT FROM AUTHOR]

Abstract: Maturity index (MI) is a key determinant of pea softness and ultimately retail value. Pea seed development goes through the optimal market stage for human consumption about a week before harvest. MI increases rapidly during the last 3–4 days prior to the optimal harvest which is when there is a need for better forecasting capability. Extensive field sampling is currently used to track MI in each of the individual paddocks, though it has limited ability to predict MI more than a day ahead. We developed an Artificial Neural Network (ANN) model that complements field sampling by forecasting the MI trend several days ahead. It was built using historical harvest information along with weather and climate forecasts. We implement and evaluate the ANN in a large pea growing region in Tasmania, Australia, and this paper highlights key results. The ANN produced an average error of 31.8MI units when forecasting MI at harvest with a 7-day lead time versus the current manual method which produced an average error of 36.6MI units for a lead time of 2 days. This means the model provides the ability to not only harvest peas closer to their ideal MI but also plan harvesting and transport logistics with a much greater lead time. [Copyright &y& Elsevier]

Abstract: With the introduction of restructuring into the electric power industry, the price of electricity has become the focus of all activities in the power market. Electricity price forecast is key information for electricity market managers and participants. However, electricity price is a complex signal due to its non-linear, non-stationary, and time variant behavior. In spite of performed research in this area, more accurate and robust price forecast methods are still required. In this paper, a new forecast strategy is proposed for day-ahead price forecasting of electricity markets. Our forecast strategy is composed of a new two stage feature selection technique and cascaded neural networks. The proposed feature selection technique comprises modified Relief algorithm for the first stage and correlation analysis for the second stage. The modified Relief algorithm selects candidate inputs with maximum relevancy with the target variable. Then among the selected candidates, the correlation analysis eliminates redundant inputs. Selected features by the two stage feature selection technique are used for the forecast engine, which is composed of 24 consecutive forecasters. Each of these 24 forecasters is a neural network allocated to predict the price of 1h of the next day. The whole proposed forecast strategy is examined on the Spanish and Australia’s National Electricity Markets Management Company (NEMMCO) and compared with some of the most recent price forecast methods. [Copyright &y& Elsevier]

This paper presents a computational approach for predicting the Australian stock market index -- AORD using multi-layer feed-forward neural networks from the time series data of AORD and various interrelated markets. This effort aims to discover an effective neural network or set of adaptive neural networks for this prediction purpose, which can exploit or model various dynamical swings and inter-market influences discovered from professional technical analysis and quantitative analysis. Within a limited range defined by our empirical knowledge, three aspects of effectiveness on data selection are considered: effective inputs from the target market (AORD) itself, a sufficient set of interrelated markets, and effective inputs from the interrelated markets. Two traditional dimensions of the neural network architecture are also considered: the optimal number of hidden layers, and the optimal number of hidden neurons for each hidden layer. Three important results were obtained: A 6-day cycle was discovered in the Australian stock market during the studied period: the time signature used as additional inputs provides useful information; and a basic neural network using six daily returns of AORD and one daily returns of SP500 plus the day of the week as inputs exhibits up to 80% directional prediction correctness. [ABSTRACT FROM AUTHOR]

Background: Non‐nutritive sucking habits, bottle feeding, and facial hyperdivergency have been suggested as the influencing factors of anterior open bite (AOB). There was inconsistent reporting of prevalence and a gap of knowledge in the literature. Aim: The aim of this study was to investigate the prevalence of AOB in 7‐ to 12‐year‐olds, with a special interest to characterise the determinants of this malocclusion. Design: A sample of 203 children aged 7 to 12 years were selected from a primary school in regional Australia. The legal guardian of each child completed a self‐administered questionnaire. Data were assessed using methods of univariate statistics and neural analysis. Results: The prevalence of AOB and thumb sucking was 24.1% and 23.2%, respectively. AOB was associated with the habit (correlation = 0.754) and duration (correlation = 0.574) of thumb sucking. Age, gender, birth order, feeding modality, sleep patterns, history of orthodontic treatment and tonsil, adenoid or grommet surgery, and parents' level of education were not related to AOB (correlation absolute value ≤ 0.474). Conclusions: Thumb sucking, specifically for prolonged duration, increases the risk of development of AOB. Cessation of thumb sucking habits should be encouraged at an early age to avoid the development of AOB. [ABSTRACT FROM AUTHOR]

Purpose: This study aims to explore the adoption of contact tracing apps through a hybrid analysis of the collected data using structural equation modelling (SEM) and artificial neural networks (ANN), leading to the identification of the critical determinants for the adoption of contact tracing apps in Australia. Design/methodology/approach: A research model is developed within the background of the unified theory of acceptance and use of technology (UTAUT) and the privacy calculus theory (PCT) for investigating the adoption of contact tracing apps. This model is then tested and validated using a hybrid SEM-ANN analysis of the survey data. Findings: The study shows that effort expectancy, perceived value of information disclosure and social influence are critical for adopting contact tracing apps. It reveals that performance expectancy and perceived privacy risks are indirectly significant on the adoption through the influence of perceived value of information disclosure. Furthermore, the study finds out that facilitating condition is insignificant to the adoption of contact tracing apps. Practical implications: The findings of the study can lead to the formulation of targeted strategies and policies for promoting the adoption of contact tracing apps and inform future epidemic control for better emergency management. Originality/value: This study is the first attempt in integrating UTAUT and PCT for exploring the adoption of contact tracing apps in Australia. It combines SEM and ANN for analysing the survey data, leading to better understanding of the critical determinants for the adoption of contact tracing apps. [ABSTRACT FROM AUTHOR]

Identification of the extent of heat altered coals is important for coal mining and resource estimation because alteration directly affects key economic properties such as ash content, volatile matter, specific energy, sulphur, free swell index and beneficiation characteristics. The most reliable method to identify altered coal is through geochemical analysis of drill core samples; however this method is time consuming and costly. Cheaper alternative methods to identify altered coal is by macroscopic observation of non-core drill cuttings, but this is less reliable than core. Geophysical wireline log data is also used to identify altered coal, but with variable success. Over the past decades, Machine Learning methods have proven popular to automatically classify a variety of lithotypes from geophysical log data. However rarely has there been focus on predicting altered coal. In this paper we apply the most recent machine learning methods which include gradient boosted machines, random forests and artificial neural networks to automatically predict altered and non-altered lithotypes using geophysical log data. We use a massive data set comprising of 1230 samples from 263 boreholes from a highly intruded deposit in the Bowen Basin, Eastern Australia. To do this, we calculate each sample's distance from intrusion and predict their relative density from geophysical log inputs including gamma ray, caliper and compensated density. We then train our machine learning methods on 80% of the data to predict alteration class using the calculated distance to intrusion, predicted relative density, and geophysical log gamma ray, as primary inputs. We evaluated each of these machine learning methods on the remaining 20% of the data to determine the best performing model. Finally, using the best performing model we further split the altered and non-altered classification into lithotypes using a decision tree based on geological knowledge of the case study area. The results indicate that of the machine learning algorithms the random forest produced the best results with only 11 misclassifications across the entire data set of 1230 samples which represents <1% error. • Prediction of heat affected coals with a massive dataset and machine learning methods in the Bowen Basin. • Prediction of relative density and classification of altered and non-altered coals. • Lithology group definition based on a decision tree with <1% error. [ABSTRACT FROM AUTHOR]

Context: The contribution of forest ecosystem services to human well-being varies over space following the dynamics in forest cover. Use of machine learning models is increasing in projecting forest cover changes and investigating the drivers, yet references are still lacking for selecting machine learning models for spatial projection of forest cover patterns. Objectives: We assessed the ability of nonparametric machine learning techniques to project the spatial distribution of forest cover and identify its drivers using a case study of Tasmania, Australia. Methods: We developed, evaluated, and compared the performance of four nonparametric machine learning models: support vector regression (SVR), artificial neural networks (ANN), random forest (RF), and gradient boosted regression trees (GBRT). Results: The results demonstrated that RF far outperformed the other three models in both fitting and projection accuracy, and required less computional costs. GBRT outperformed SVR and ANN in projection accuracy. However, RF exhibited serious overfitting due to the full growth of its decision trees. The influence rankings of explanatory variables on spatial patterns of forest cover were different under the four models. Land tenure type and rainfall were identified among the top four most influential variables by all four models. The ranking produced by the RF model was significantly different with topographic factors associated with land clearing and production costs (elevation and distance to timber facilities) being the two most influential variables. Conclusions: We encourage practitioners to consider nonparametric machine learning methods, especially RF, when facing problems of complex environmental data modelling. [ABSTRACT FROM AUTHOR]

Although it is conceptually assumed that global models are relatively ineffective in modelling the highly unstable structure of chaotic hydrologic dynamics, there is not a detailed study of comparing the performances of local and global models in a hydrological context, especially with new emerging machine learning models. In this study, the performance of a local model (k-nearest neighbour, k-nn) and, as global models, several recent machine learning models – artificial neural network (ANN), least square-support vector regression (LS-SVR), random forest (RF), M5 model tree (M5), multivariate adaptive regression splines (MARS) – was analysed in multivariate chaotic forecasting of streamflow. The models were developed for Australia's largest river, the River Murray. The results indicate that the k-nn model was more successful than the global models in capturing the streamflow dynamics. Furthermore, coupled with the multivariate phase-space, it was shown that the global models can be successfully used for obtaining reliable uncertainty estimates for streamflow. [ABSTRACT FROM AUTHOR]

Bananas are the world's most popular fruit and an important staple food source. Recent outbreaks of Panama TR4 disease are threatening the global banana industry, which is worth an estimated $8 billion. Current methods to map land uses are time- and resource-intensive and result in delays in the timely release of data. We have used existing land use mapping to train a U-Net neural network to detect banana plantations in the Wet Tropics of Queensland, Australia, using high-resolution aerial photography. Accuracy assessments, based on a stratified random sample of points, revealed the classification achieves a user's accuracy of 98% and a producer's accuracy of 96%. This is more accurate compared to existing (manual) methods, which achieved a user's and producer's accuracy of 86% and 92% respectively. Using a neural network is substantially more efficient than manual methods and can inform a more rapid respond to existing and new biosecurity threats. The method is robust and repeatable and has potential for mapping other commodities and land uses which is the focus of future work. [ABSTRACT FROM AUTHOR]

Joorabchi, A. Zhang, H. and Blumenstein, M, 2007. Application of artificial neural networks in flow discharge prediction for the Fitzroy River, Australia. Journal of Coastal Research, SI 50 (Proceedings of the 9th International Coastal Symposium), 287 – 291. Gold Coast, Australia, ISSN 0749.0208 Prediction of flow discharge, and in particular floods, in rivers is one of the basic and key information in regards to operation and management of the river systems. The Fitzroy River, one of the largest Australian river systems, has a historical recording of heavy floods and there is a concern for the people of that area to have a clear prediction of the stream discharge to avoid damages. In this paper a feed-forward artificial neural network (ANN) model has been used to forecast the daily flow discharge of the Fitzroy River up to four days ahead. The feed-forward neural network uses error Back-propagation learning algorithm. A cross validation method is applied to prevent the over-fitting problem. The network uses multiple inputs including the daily values of discharge. The network output consists of four neurons in respect to the number of forecasted days. A suitable number of inputs for time-series data were selected by trial and error. Two different multi-layer networks were compared to find the optimised network. The results show an accurate forecasting of flow discharge during flood events. However, the neural network overestimates during low discharge with a mean value of 80 (m3/s). [ABSTRACT FROM AUTHOR]

Paleovalleys are buried ancient river valleys that often form productive aquifers, especially in the semiarid and arid areas of Australia. Delineating their extent and hydrostratigraphy is however a challenging task in groundwater system characterization. This study developed a methodology based on the deep learning super-resolution convolutional neural network (SRCNN) approach, to convert electrical conductivity (EC) estimates from an airborne electromagnetic (AEM) survey in South Australia to a high-resolution binary paleovalley map. The SRCNN was trained and tested with a synthetic training dataset, where valleys were generated from readily available digital elevation model (DEM) data from the AEM survey area. Electrical conductivities typical of valley sediments were generated by Archie's law, and subsequently blurred by down-sampling and bicubic interpolation to represent noise from the AEM survey, inversion and interpolation. After a model training step, the SRCNN successfully removed such noise, and reclassified the low-resolution, converted unimodal but skewed EC values into a high-resolution paleovalley index following a bimodal distribution. The latter allows us to distinguish valley from non-valley pixels. Furthermore, a realistic spatial connectivity structure of the paleovalley was predicted when compared with borehole lithology logs and a valley bottom flatness indicator. Overall the methodology permitted us to better constrain the three-dimensional paleovalley geometry from AEM images that are becoming more widely available for groundwater prospecting. [ABSTRACT FROM AUTHOR]

Rainfall data provide an essential input for most hydrologic analyses and designs for effective management of water resource systems. However, in practice, missing values often occur in rainfall data that can ultimately influence the results of hydrologic analysis and design. Conventionally, stochastic interpolation methods such as kriging are the most frequently used approach to estimate the missing rainfall values where the variogram model that represents spatial correlations among data points plays a vital role and significantly impacts the performance of the methods. In the past, the standard variogram models in ordinary kriging were replaced with the universal function approximator-based variogram models, such as artificial neural networks (ANN). In the current study, applicability of genetic programming (GP) to derive the variogram model and use of this GP-derived variogram model within ordinary kriging for spatial interpolation was investigated. Developed genetic programming-based ordinary kriging (GPOK) was then applied for estimating the missing rainfall data at a rain gauge station using the historical rainfall data from 19 rain gauge stations in the Middle Yarra River catchment of Victoria, Australia. The results indicated that the proposed GPOK method outperformed the traditional ordinary kriging as well as the ANN-based ordinary kriging method for spatial interpolation of rainfall. Moreover, the GP-derived variogram model is shown to have advantages over the standard and ANN-derived variogram models. Therefore, the GP-derived variogram model seems to be a potential alternative to variogram models applied in the past and the proposed GPOK method is recommended as a viable option for spatial interpolation. [ABSTRACT FROM AUTHOR]

This study has proposed and empirically tested two Adaptive Neuro-Fuzzy Inference System (ANFIS) models for the first time for predicting Australia's domestic low cost carriers’ demand, as measured by enplaned passengers (PAX Model) and revenue passenger kilometres performed (RPKs Model). In the ANFIS, both the learning capabilities of an artificial neural network (ANN) and the reasoning capabilities of fuzzy logic are combined to provide enhanced prediction capabilities, as compared to using a single methodology. Sugeno fuzzy rules were used in the ANFIS structure and the Gaussian membership function and linear membership functions were also developed. The hybrid learning algorithm and the subtractive clustering partition method were used to generate the optimum ANFIS models. Data was normalized in order to increase the model's training performance. The results found that the mean absolute percentage error (MAPE) for the overall data set of the PAX and RPKs models was 1.52% and 1.17%, respectively. The highestR2-value for the PAX model was 0.9949 and 0.9953 for the RPKs model, demonstrating that the models have high predictive capabilities. [ABSTRACT FROM AUTHOR]