Search

Accurate land valuation is necessary for tax purposes, land resources allocation, real estate management and urban development and planning. Since various factors from different domains affect land prices through non-linear relationships, automating the land valuation process on a large scale is a complex task. Advanced technologies in big data analysis and artificial intelligence have demonstrated superior capabilities in knowledge extraction in such cases. Accordingly, this paper develops and compares the performance of four Automated Valuation Models (AVMs) based on machine learning and deep learning techniques utilizing physical, geographical, socio-economic, environmental, legal and planning factors in Melbourne Metropolitan, Australia. According to the results, the eXtreme Gradient Boosting (XGBoost) method outperforms other algorithms of Support Vector Regression (SVR), random forest and Deep Neural Network (DNN). This method has achieved the coefficient of determination (R2) of 0.862, Mean Absolute Percentage Error (MAPE) of 0.139, and normalized Root Mean Square Error (nRMSE) of 0.281. The achieved high accuracy is due to incorporating a wide range of driving factors and applying innovative feature selection and hyperparameter tuning procedures evaluating various possible feature sets and hyperparameters. Accordingly, this paper can contribute to research, governmental and industry-based activities in terms of developing AVMs for mass land valuation. • Automated land valuation models using machine learning and deep learning • Comprehensive feature set and hybrid feature selection process for robust model performance • Spatial mapping of land values in large urban areas • Contribution to fair housing policies and urban development by accurate land valuation • Practical insights for housing affordability strategies in terms of land management [ABSTRACT FROM AUTHOR]

Power grid damage and blackouts are increasing with climate change. Load forecasting methods that integrate climate resilience are therefore essential to facilitate timely and accurate network reconfiguration during periods of extreme stress. Our paper proposes a generalised Wildfire Resilient Load Forecasting Model (WRLFM) to predict electricity load based on operational data of a Distribution Network (DN) in Australia during wildfire seasons in 2015–2020. We demonstrate that load forecasting during wildfire seasons is more challenging than during non-wildfire seasons, motivating an imperative need to improve forecast performance during wildfire seasons. To develop the robust WRLFM, comprehensive comparative analyses were conducted to determine proper Machine Learning (ML) forecast structures and methods for incorporating multiple factors. Bi-directional Gated Recurrent Unit (Bi-GRU) and Vision Transformer (ViT) were selected as they performed the best among all 13 recently trending ML methods. Multi-factors were incorporated to contribute to forecast performance, including input sequence structures, calendar information, flexible correlation-based temperature conditions, and categorical Fire Weather Index (FWI). High-resolution categorical FWI was used to build a forecasting model with climate resilience for the first time, significantly enhancing the average stability of forecast performances by 42%. A sensitivity analysis compared data set patterns and model performances during wildfire and non-wildfire seasons. The improvement rate of load forecasting performance during wildfire seasons was more than two times greater than in non-wildfire seasons. This indicates the significance and effectiveness of applying the WRLFM to improve forecast accuracy under extreme weather risks. Overall, the WRLFM reduces the Mean Absolute Percentage Error (MAPE) of the forecast by 14.37% and 20.86% for Bi-GRU and ViT-based models, respectively, achieving an average forecast MAPE of around 3%. • Novel climate-resilient load forecast model for use during extreme wildfire seasons. • Bi-GRU and ViT outperform 11 recently trending machine learning models. • Categorical FWI with high resolution enhances forecast stability by 42%. • Optimised calendar, temperature and FWI input reduce errors by 17.6%. • Forecast improvement during wildfires is twice that in non-wildfire seasons. [ABSTRACT FROM AUTHOR]

Although traditional physical models have been used to analyse groundwater systems, the emergence of novel machine learning models can improve the accuracy of the predictions. Deep learning has been prominent in environmental and climate change problems. In this paper, we present a framework for utilising deep learning models to predict groundwater levels based on nearby streamflow and rainfall data. We address the missing data problem using a Bayesian linear regression model within the deep learning framework. Our deep learning framework utilises models such as long-short term memory (LSTM) networks and convolutional neural networks (CNN) for multi-step ahead time series prediction. We examine the fluctuations in groundwater levels at various boreholes located near Middle Creek in New South Wales, Australia. We use the National Collaborative Research Infrastructure Strategy (NCRIS) groundwater database and utilise Bayesian linear regression to impute missing data. We investigate the accuracy of the selected models for individual and regional basins and univariate and multivariate strategies. Our results show that the LSTM-based regional model with multivariate strategy using rainfall data provided the best accuracy. • We predict groundwater levels based on stream flow and rainfall data. • Our framework uses Bayesian data imputation and deep learning for multi-step prediction. • We examine the groundwater fluctuations at various boreholes located in Australia. • Results show a fine-tuned regional model with rainfall data yielded the best accuracy. • We find that LSTM models outperformed CNNs using the multivariate approach. [ABSTRACT FROM AUTHOR]

In response to the high uncertainty associated with residential net load due to the coupling of distributed photovoltaic generation and user demand, this paper proposed a novel cluster-based stacking ensemble learning model for net load interval prediction. Firstly, the k-means algorithm is employed to discover the similarity in user electricity consumption patterns. Then, a RIME optimization algorithm with local enhancement (LRIME) is developed to optimize the parameters and weights of the base learners in stacking ensemble learning. Subsequently, base learners with strong predictive capabilities and significant diversity are chosen as the first-layer predictive models, extreme learning machine (ELM) is utilized as the second-layer predictive model, ultimately resulting in the proposed stacking ensemble learning prediction model. And utilizing the bootstrap method to fit the volatility of point predictions, different prediction intervals are obtained at varying confidence levels, aiming to quantify the integrated uncertainty in photovoltaic generation and load. Testing on the open Ausgrid electricity load data in Australia provided robust validation of the proposed method's effectiveness. In comparison with other outstanding prediction models, the proposed ensemble model can effectively capture the uncertainty in integrating photovoltaic generation and user load. • A RIME optimization algorithm with local enhancement (LRIME) is developed. • A stacking ensemble model is proposed for residential net load interval prediction. • The proposed model demonstrates higher forecasting accuracy and reliability. [ABSTRACT FROM AUTHOR]

The increased penetration of renewable energy sources (RESs) as an effective mechanism to reduce carbon emissions leads to an increased weather dependency for power and energy systems. This has created dynamic operation and degradation phenomena, which affect the lifetime estimation of the assets operated with RESs. For the reliable and efficient operation of RES it is crucial to monitor the health of its constituent components and feature selection is a crucial step for building robust and accurate health monitoring approaches. In this context, this paper presents a probabilistic feature selection approach, which probabilistically weights and selects features through a heuristic and iterative process for an improved asset lifetime estimation. Power transformers are key power grid assets and they are used to demonstrate the validity and impact of the proposed approach. The approach is tested on two different photovoltaic power plants operated in Spain and Australia. Results consistently show that the proposed feature-selection approach reduces the prediction error and consistently selects relevant features. The approach has been applied to transformer lifetime estimation, but it can be generally applied to assist in the lifetime estimation of other components operated in RESs. Part of the studies presented here as well as source codes are all open-source under the GitHub repository https://github.com/iramirezg/FeatureSelection. • Probabilistic feature selection approach for improved asset lifetime estimation. • Integration of environmental features for improved renewable-operated asset lifetime. • Systematic and robust feature weighting methodology. • Improved transformer lifetime estimation including sensor and environmental data. • Validated on two real photovoltaic power plant case studies. [ABSTRACT FROM AUTHOR]

The peak period of an energy-generating wave is one of the most important parameters that describe the spectral shape of the oceanic wave, as this indicates the duration for which the waves prevail with respect to their maximum extractable energy. In this paper, a half-hourly peak wave energy period (T P) forecast model is constructed using a suite of statistically significant lagged inputs based on the partial auto-correlation function with an extreme learning machine model developed and its predictive utility is benchmarked against deep learning models, i.e., convolutional neural network (CNN/CovNet) and recurrent neural network (RNN) models and other traditional M5tree, Conditional Maximization based Multiple Linear Regression (MLR-ECM) and MLR models. The objective model (ELM) vs. the comparison models (CNN, RNN, M5tree, MLR-ECM, and MLR) were trained and validated independently on the test dataset obtained from coastal zones of eastern Australia that have a high potential for implementation of wave energy generation systems. The outcomes ascertain that the ELM model can generate significantly accurate predictions of the half-hourly peak wave energy period, providing a good level of accuracy relative to deep learning models in selected coastal study zones. The study establishes the practical usefulness of the ELM model as being a noteworthy methodology for the applications in renewable and sustainable energy resource management systems. • A real-time peak wave energy period forecasting model is constructed using ELM model. • The advance ELM is benchmarked against a deep learning and other traditional models. • The advance ELM generates better predictions than deep learning models in this region. • ELM model provides significant energy management implications for ocean waves. [ABSTRACT FROM AUTHOR]

Property valuation plays a significant role in urban economics and is of great importance to various stakeholders who interact and shape the city, including property owners, buyers, banks, land developers, real estate agents, local councils and government planning authorities. In the literature, various predictive models have been proposed to automate the calculation of property value, most of which endeavour to factor in the combination of property characteristics, market factors and location-based attributes associated with individual properties use large citywide databases. At the same time, it has been widely acknowledged that regional sub-areas have impacts on property price prediction. Therefore, this paper aims to investigate the performance of various techniques on sub-areas using the Greater Sydney Region as the study area. The sub-area in this paper is defined as the statistical areas (SAs) as defined by the Australian Bureau of Statistics. In particular, two different SA geographies (SA4, SA3) along with the City Level are adopted to understand the spatial dependence which occurs at different levels. With real-world transaction records and data collected from a diverse range of sources, various methods including the traditional hedonic price model (HPM) and popular machine learning (ML) approaches are implemented and evaluated for property price prediction. Two different property markets for residential property are modelled, being for housing stock and apartment (unit) stock. Experimental results show that Random Forest and Gradient Boosting-based methods outperform other approaches in most scenarios and that the high spatial resolution property sub-area (SA3) improved the performance in terms of overall model accuracy. This research provides insights into how sub-area machine learning models can be employed in real estate to characterize property price, and helps understand the influential factors in different local geographical areas for policy-making. • Property valuation is modelled with property attributes, locational and socioeconomic factors by a variety of machine learning algorithms. • Models are implemented on three different geographical area levels. • Both in-time and out-of-time validation are implemented. • House and Apartments are both valued. • Valuation carried out on smaller regions can improve the predictive performance of most modelling techniques. • The study illustrates that the variables' contributions to the property price vary for different geographical areas. [ABSTRACT FROM AUTHOR]

In this paper, a systematic machine learning strategy is proposed to classify wood properties based on a contact ultrasonic testing results. As such, several aspects of the wood material including the type of wood (hardwood or softwood), the direction of an ultrasonic test with respect to the growth rings of the wood, and, whether the wood is damaged or intact are investigated. As a pre-processor, the Variational Mode Decomposition technique is applied to the nonlinearly modulated ultrasonic signals, and the centre frequencies of the decomposition results are taken as features for Machine learning Algorithms (MLAs). Then with each of the MLAs, hyperparameter settings were optimised and technical aspects of the feature engineering are discussed. Best results were achieved using Ensemble classifiers, SVM, and KNN using three features/decomposition. To explore the physics behind the nonlinear problem, the relative false discovery rate obtained from the confusion matrix associated with applying the MLAs is proposed as a metric. We show that different features are capable of exploring different aspects of the problem better. The techniques of this paper can be applied to quality assessment of wood materials. The paper also demonstrates the capability of MLAs in exploring some physics of such problems. Moreover, billets with natural imperfections harvested from the site Collie in WA, Australia are classified to demonstrate the applicability of the proposed approach in real world problems. The result of 94% 5-fold cross-validation accuracy indicates the effectiveness of the proposed approach. • Contact ultrasonic technique is used for the wood quality assessment. • The Variational Mode Decomposition is used to decompose ultrasonic signals. • time–frequency features of the decomposed signals are used as features. • Machine learning algorithms are used to solve classification problems. • Some physics of the problem has been explored from the machine learning results. [ABSTRACT FROM AUTHOR]

• Developed an approach to approximate the demand of a carsharing station. • Improved the estimation accuracy of the model by including social media data and spatio-temporal correlated data. • Applied a spatio-temporal autocorrelation model. • Developed a correlation analysis to account for the interactions among variables. As an emerging urban mobility service, carsharing has become increasingly popular worldwide. To understand customers' needs and optimize the design of service networks, the usage of carsharing vehicles whose trips are recorded by the operators has been applied in research estimating carsharing demand. However, as a form of spatio-temporal correlated data, the underlying spatio-temporal information included in such carsharing records has not been investigated in existing models of carsharing demand. Meanwhile, due to the supply limitation of carsharing stations, some demand cannot be fulfilled and thus remains unrecorded in the operational data. Unrealized demand may lead to underestimation of carsharing demand and therefore an incorrect vehicle deployment strategy by the service providers. In view of these issues, this paper develops an innovative approach to estimating the actual demand at a carsharing station with operational data from GoGet , the largest carsharing company in Australia. The accuracy of the estimation is improved by adding spatio-temporal correlated variables as well as variables from emerging data sources such as social media. To explore the latent space-and-time correlated information, spatio-temporal autoregressive and moving-average models have been applied. Based on the results of the analysis, the paper also provides recommendations related to the operation policies of the service providers. [ABSTRACT FROM AUTHOR]

Local and state governments depend on small area population forecasts to make important decisions concerning the development of local infrastructure and services. Despite their importance, current methods often produce highly inaccurate forecasts. Recent years have witnessed promising developments in time series forecasting using Machine Learning across a wide range of social and economic variables. However, limited work has been undertaken to investigate the potential application of Machine Learning methods in demography, particularly for small area population forecasting. In this paper we describe the development of two Long-Short Term Memory network architectures for small area populations. We employ the Keras Tuner to select layer unit numbers, vary the window width of input data, and apply a double training and validation regime which supports work with short time series and prioritises later sequence values for forecasts. These methods are transferable and can be applied to other data sets. Retrospective small area population forecasts for Australia were created for the periods 2006–16 and 2011–16. Model performance was evaluated against actual data and two benchmark methods (LIN/EXP and CSP-VSG). We also evaluated the impact of constraining small area population forecasts to an independent national forecast. Forecast accuracy was influenced by jump-off year, constraining, area size, and remoteness. The LIN/EXP model was the best performing method for the 2011-based forecasts whilst deep learning methods performed best for the 2006-based forecasts, including significant improvements in the accuracy of 10 year forecasts. However, benchmark methods were consistently more accurate for more remote areas and for those with populations <5000. • A novel deep learning method for small area forecasts is presented. • Forecast accuracy was influenced by jump-off year, constraining, area size, and remoteness. • Deep learning methods performed best for the 5 and 10 year 2006-based forecasts. • The LIN/EXP model was the best performing method for the 5 year 2011-based forecasts. • Benchmark methods were consistently more accurate for more remote areas and for those with populations <5000. [ABSTRACT FROM AUTHOR]

Predicting electricity demand data is considered an essential task in decisions taking, and establishing new infrastructure in the power generation network. To deliver a high-quality electricity demand prediction, this paper proposes a hybrid combination technique, based on a deep learning model of Convolutional Neural Networks and Echo State Networks, named as CESN. Daily electricity demand data from four sites (Roderick, Rocklea, Hemmant and Carpendale), located in Southeast Queensland, Australia, have been used to develop the proposed hybrid prediction model. The study also analyzes five other machine learning-based models (support vector regression, multilayer perceptron, extreme gradient boosting, deep neural network, and Light Gradient Boosting) to compare and evaluate the outcomes of the proposed deep learning approach. The results obtained in the experimental study showed that the proposed hybrid deep learning model is able to obtain the highest performance compared to other existing models developed for daily electricity demand data forecasting. Based on the statistical approaches utilized in this study, the proposed hybrid approach presents the highest prediction accuracy among the compared models. The obtained results showed that the proposed hybrid deep learning algorithm is an excellent and accurate electricity demand forecasting method, which outperformed the state of the art algorithms that are currently used in this problem. [Display omitted] • A Hybrid deep learning model is proposed for daily electricity demand prediction. • Convolutional Neural Networks and Echo State Networks are integrated into a single forecasting algorithm. • The proposed deep learning-based approach is excellent in electricity demand prediction. • The proposed algorithm has practical use in inter-daily electricity demand prediction. [ABSTRACT FROM AUTHOR]

• Formulate the logistic regression model with minimum classification error function. • Adopt classification maximum likelihood criterion to infer label information. • Compare results of the proposed model with others and show the effectiveness. • Investigate the most significant contributing factors to hit-and-run crashes. • Conduct sensitivity analysis of related parameters. • Provide implications for policies and counter-measures for hit-and-run crashes. Hit-and-run crashes not only degrade the morality, but also result in delays of medical services provided to victims. However, class imbalance problem exists as the number of hit-and-run crashes is much smaller than that of non-hit-and-run crashes. The missing label problem also exists in the crash analysis due to reasons like data barrier such that the information hidden in the unlabelled samples has not been effectively utilised. In this paper, a cost-sensitive semi-supervised logistic regression (CS3LR) model is proposed for hit-and-run analysis, in order to tackle class-imbalanced data distribution and missing label problem, based on the crash dataset of Victorian, Australia (2013–2019). By performing label estimation with logistic regression jointly utilising both labelled and unlabelled data with pseudo labels in a well-designed cost-sensitive semi-supervised maximum likelihood framework, the proposed model can obtain an unbiased likelihood parameter for hit-and-run prediction and analysis. Comparing the experimental results of CS3LR model with two logistic regression models and seven machine learning methods, better performance of CS3LR model is demonstrated. The most significant contributing factors to hit-and-run crashes extracted by CS3LR with only 10% labelled data show a high degree of consistency with the true contributing factors obtained by the supervised cost-sensitive logistic regression with complete hit-and-run labels. The effects of class-weighted ratio and hyper-parameter λ on the performance of hit-and-run crash prediction model have also been analysed. The results can further provide recommendations and implications on the policies and counter-measures for preventing hit-and-run collisions and crimes. The methodology proposed in this paper can also be employed to analyse crash data with other types of missing labels, such as crash severity. [ABSTRACT FROM AUTHOR]

Modern power systems are experiencing the challenge of high uncertainty with the increasing penetration of renewable energy resources and the electrification of heating systems. In this paradigm shift, understanding electricity users' demand is of utmost value to the retailers, aggregators, and policymakers. However, behind-the-meter (BTM) equipment and appliances at the household level are unknown to the other stakeholders mainly due to privacy concerns and tight regulations. In this paper, we seek to identify residential consumers based on their BTM equipment, mainly rooftop photovoltaic (PV) systems and electric heating, using imported/purchased energy data from utility meters. To solve this problem with an interpretable, fast, secure, and maintainable solution, we propose an integrated method called Interpretable Refined Motifs And binary Classification (IRMAC). The proposed method comprises a novel shape-based pattern extraction technique, called Refined Motif (RM) discovery, and a single-neuron classifier. The first part extracts a sub-pattern from the long time series considering the frequency of occurrences, average dissimilarity, and time dynamics while emphasising specific times with annotated distances. The second part identifies users' types with linear complexity while preserving the transparency of the algorithms. With the real data from Australia and Denmark, the proposed method is tested and verified in identifying PV owners and electrical heating system users. The performance of the IRMAC is studied and compared with various state-of-the-art methods. The proposed method reached an accuracy of 96% in identifying rooftop PV users and 94.4% in identifying electric heating users, which is comparable to the best solution based on deep learning, while the speed of the inference model is a thousand times faster. Last but not least, the proposed method is a transparent algorithm, which can tackle the concerns regarding the agnostic decision-making process when policies prohibit some machine learning methods. [ABSTRACT FROM AUTHOR]

Conventional housing price prediction methods rarely consider the spatiotemporal non-stationary problem in a large data volumes. In this study, four machine learning (ML) models are used to explore the impacts of various features – i.e., property attributes and neighborhood quality - on housing price variations at different geographical scales. Using a 32-year (1984–2016) housing price dataset of Metropolitan Adelaide, Australia, this research relies on 428,000 sale transaction records and 38 explanatory variables. It is shown that non-linear tree-based models, such as Decision Tree, have perform better than linear models. In addition, ensemble machine learning techniques, such as Gradient-Boosting and Random Forest, are better at predicting future housing prices. A spatiotemporal lag (ST-lag) variable was added to improve the prediction accuracy of the models. The study demonstrates that ST-lag (or similar spatio-temporal indicator) can be a useful moderator of spatio-temporal effects in ML applications. This paper will serve as a catalyst for future research into the dynamics of the Australian property market, utilizing the benefits of cutting-edge technologies to develop models for business and property valuation at various geographical levels. • Spatio-temporal non-stationary is significant in explanation of the variations in housing price; • A proposed spatiotemporal lag is incorporated to increase the accuracy of the models; • Non-linear tree-based models have better performance than the linear model. • Ensemble ML techniques are powerful methods to produce better predictive performance. [ABSTRACT FROM AUTHOR]

More timely and accurate air quality forecasting could contribute to better public health protection and air pollution prevention. Particulates are a significant indicator for measuring the degree of air pollution. This paper reports on research to model an early warning tool for coarse particulates when assessing the impact of the 12 satellite-derived and ground-based meteorological pollutants out of 30 pollutants considered using hourly Australian data from January 2018–December 2020. A one-dimensional convolutional neural network (CNN) was integrated with a one-directional fully gated recurrent unit (GRU) to forecast consecutive hours' air quality. The CNN model acts as a spatial feature extractor, whereas the new generation GRU makes it computationally efficient. The resultant hybrid ' CNN-GRU' is then comprehensively benchmarked outperforming an ensemble of six other deep learning models. The proposed model's efficacy is indicated at the four most air polluted Australian postcodes in the testing phase. A detailed error analysis with visual and statistical metrics for air quality forecasting ascertains the proposed model's countermeasure to reduce harm and loss. The practical tool is immensely beneficial and can be widely deployed to the regions of public health concern where air pollution is a significant hazard. [Display omitted] • A novel early warning hybrid PM10 forecasting framework designed for Australia. • 30 satellite and ground-based meteorological pollutants were considered with 7 modeling approaches. • CNN integrated with GRU outperformed the other models. • Early warning AI model show potential in the health, environment sector, and decision-making. [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]

• Novel ensemble framework which improves short-term load forecasting accuracy. • Model divides the problem into several subtasks based on peak and off-peak conditions. • Classification and regression models were developed to solve each subtask. • Framework is validated on real-world operational demand across regions in Australia. • Model has a substantially higher accuracy (up to 25.4% in MAE) over other methods. [Display omitted] To keep the balance between electricity demand and supply as well as infrastructure planning, it is important to accurately forecast the electricity demand. This has become a challenging task due to increasing share of renewable energy and prosumers (i.e. consumers who produce electricity) in the electricity grid. This paper develops a cooperative ensemble framework which divides the forecasting problem into several subtasks based on peak and off-peak conditions. Each subtask is then solved using multiple forecasting models that include classification and regression. The developed framework is finally validated on real-world operational demand across the National Electricity Market (NEM) of Australia. The performance is comprehensively compared against various state-of-the-art techniques in the field, which indicates up to 25.4% mean absolute error (MAE) improvement. [ABSTRACT FROM AUTHOR]

• A hybrid short-term load forecasting method is proposed. • A data preprocessing module is conducted based on de-noising technique and fuzzy time series. • To balance the forecasting accuracy and stability, a multi-objective optimization algorithm is applied to optimize the BPNN model. Countries around the globe have introduced renewable energies (RE) and minimized the dependency of fossil resources in power systems to address extensive environmental risks. However, such large-scale energy transitions pose a great challenge to power systems due to the volatility of RE. Meanwhile, power demand is increasing over time and it shows temporal characteristics, such as seasonal and peak-valley patterns. Whether the future power system with a larger proportion of RE can meet the surging but fluctuated electricity demand remains problematic. Previous studies on short-term load forecasting focused more on forecasting accuracy than stability. Further, there is a relative paucity of research into temporal patterns. In order to fill in these research gaps, this paper proposes a fuzzy theory-based machine learning model for workdays and weekends short-term load forecasting. Fuzzy time series (FTS) is applied for data mining and back propagation (BP) neural network is used as the main predictor for short-term load forecasting. To exploit the trade-offs between forecasting stability and accuracy, multi-objective optimization is applied to modify the parameters of BP. Moreover, an interval forecasting architecture with several statistical tests is constructed to address forecasting uncertainties. Short-term load data from Victoria in Australia is selected as a case study. Results demonstrate that the proposed method can significantly boost forecasting stability and accuracy, and help strategy making in the field of energy and electricity system management and planning. [ABSTRACT FROM AUTHOR]