Search

This paper is aimed to develop a hybrid model (PCA-ANN) to predict the extractive content in eucalyptus wood clones. The input variables to the virtual analyzer are planting parameters available from the forest inventory, therefore the prediction does not rely on laboratory analysis of the wood samples, affording a quick estimate of the extractive contents. This study further bridges the literature gap on the investigation of the cause of variability of extractive content in eucalyptus wood. The PCA-ANN was identified from experimental data to predict and monitor the extractive content, since laboratory measurements can take several days and become available only after wood processing. The experimental data contained information on ten species of eucalyptus clones from five regions in the extreme south of Bahia, Brazil. Principal Component Analysis (PCA) firstly assessed the impact of planting variables on the extractives content. The variability of the data was represented by eight principal components and the variables that mostly contribute to the extractive content are: potential acidity, iron, saturation of aluminum, magnesium, pH, base saturation, remaining phosphorus, zinc, manganese and copper. The artificial neural network (ANN) with the 8 principal components in the input layer showed that the PCA could effectively reduce the dimensionality of the data. For practical purposes, though, the ANN with 10 input variables and 16 neurons in the hidden layer, presenting an average relative deviation of 1.5%, is recommended. The prediction of the extractive content is essential to allow preventive management practices toward the improvement of yield and quality of the cellulosic pulp. [ABSTRACT FROM AUTHOR]

The large-scale integration into electrical systems of intermittent power-generation sources, such as wind power plants, requires greater efforts and knowledge from operators to keep these systems operating efficiently. These sources require reliable output power forecasts to set up the optimal operating point of the electrical system. In previous research, the authors developed an evolutionary approach algorithm called RCDESIGN to optimize the hyperparameters and topology of Echo State Networks (ESN), and applied the model in different time series forecasting, including wind speed. In this paper, RCDESIGN was modified in some aspects of the genetic algorithm, and now it optimizes an ESN with augmented states (ESN-AS) and has been called RCDESIGN-AS. The evolutionary algorithm allows the search for the best parameters and topology of the recurrent neural network to be performed simultaneously. In addition, RCDESIGN-AS has the important characteristic of requiring little computational effort and processing time since it is not necessary for the eigenvalues of the reservoir weight matrix to be reduced and also due to the fact that the augmented states make it possible to reduce the number of neurons in the reservoir. The method was applied for wind speed forecasting with a 24-h ahead horizon using real data of wind speed from five cities in the Northeast Region of Brazil. All results obtained with the proposed method overcame forecasting performed by the persistence method, obtaining prediction gains ranging from 60% to 80% in relation to this reference method. In some datasets, the proposed method also yielded better results than the traditional ESN, showing that RCDESIGN-AS can be a powerful tool for wind-speed forecasting and possibly for other types of time series. [ABSTRACT FROM AUTHOR]

Summary: An adaptation of the portfolio theory (PT) is proposed in this article, denoted as PrevPT, "Previsão" (in Portuguese) by PT, to integrate the three artificial neural networks, namely multilayer perceptron (MLP) backpropagation, radial basis function (RBF), and self‐organizing map (SOM), based forecasting techniques, aiming to analyze the impact of wind speed forecasting errors and achieve more accurate results. In its first use, the PT goal was to maximize a financial return, at any risk, through the diversification of securities or investments that are not positively correlated. Based on the development of PrevPT, which was used until this work only for solar forecasting, the proposed technique is applied in this paper to integrate and improve the results of individual wind forecasts. Four‐year wind speed data (January 2007 to December 2010) from two different locations (Algeciras, Spain and Petrolina, Brazil) were used. Our methodology develops a topology that integrates the forecasts obtained by MLP, RBF, and SOM aiming to obtain smaller forecast errors. By diversifying the forecasted asset, when one of the assets has negative prediction errors, another compensates for them and, thus, the total or partial cancellation of the errors occurs. PrevPT obtains a mean absolute percentage error of 1.13% for Spain and 2.35% for Brazil. PrevPT surpassed the results obtained by the three techniques applied individually in the two locations. The main innovations of the methodology are the significant reduction of errors and optimization of resource planning, and the beneficial features compared to other predictor integration techniques. [ABSTRACT FROM AUTHOR]

Cachaça is a distilled spirit made from sugarcane, exclusively produced in Brazil, and appreciated worldwide. This paper seeks to evaluate the sensory characteristics of 24 nonaged artisanal cachaça samples from Salinas (Minas Gerais, Brazil) through descriptive analysis, as well as chemometrically treat the obtained data based on principal components analysis (PCA) and Kohonen's neural network. The attributes (23) were divided between aroma (11) and flavor (12). PCA does not show good differentiation of nonaged cachaça samples. On the other hand, by using Kohonen's neural network it was possible to group samples according to their aroma and flavor characteristics in 9 and 10 distinct groups, respectively. A reduced number of descriptors could be used to describe the flavor of cachaça samples (alcohol, acidic, sweet, bitter, citric, tar, and burning), as significant correlations (R > 0.70, p <.05) exist among them with fruity, bagasse, fermented sugarcane juice, and astringent descriptors. This diminution on descriptors numbers could be able to reduce the workload of the judging panel with no losses to the sample' sensory characterization. The use of Kohonen's network chemometric treatment for treat sensory data showed to be a better alternative that PCA approach in this study. Practical applications: The growth in production and the appreciation of cachaça in the domestic and foreign markets have directed the production of the drink in Brazil with a focus on its quality and added value, seeking to obtain international recognition and increase exports. In this sense, the lexicon development and sensory characterization of samples from Salinas‐MG is very important to compare with commercial and artisanal samples from other regions in future studies. This work created a sensory profile for samples from this region, which despite its national and international popularity, does not exist yet. Finally, this is the first study that describes a lexicon development and sensory characterization of cachaças from Salinas‐MG and describes data with artificial neural networks, a potential approach for further studies. [ABSTRACT FROM AUTHOR]

This paper aims to provide alternative approaches for automatic classification of subsurface hydrocarbon-bearing regions from 2D seismic images driven by multi-layer perceptron neural networks (MLPs) (a kind of artificial neural network) and convolutional neural networks (CNNs). The first approach is based on a standard MLP whose features are controlled by Haralick's textural descriptors; the second one is developed with a multiple-layer CNN. Both techniques are studied to identify geologic "leads", instead of delineating other structures of the porous medium, such as salt bodies or seismic faults. The outcomes obtained from each approach are evaluated for a dataset of seismic images corresponding to the offshore SEAL Basin in Brazil's northeastern. Performance indicators (accuracy, recall, precision, F-measure and loss) are computed to verify training and validation of the network learning capabilities. It is shown that for both MLP and CNN configurations, good agreement is achieved in blind testing qualitatively and quantitatively. • Automatic classification of hydrocarbon accumulations in 2D seismic images. • Straightforward detection of the location of geologic leads. • Image processing through a multilayer convolutional neural network. • Feature extraction from several Haralick's textural descriptors. • Classification with good agreement in blind testing. [ABSTRACT FROM AUTHOR]

This paper aims to evaluate and compare the mixed effects modeling and artificial neural networks in order to estimate the taper of eucalyptus in integrated Crop-Livestock-Forestry (iCLF) systems. The data were collected in an experimental area of iCLF, implanted by the Brazilian Company of Farming Research – EMBRAPA Agrossilvipastoril, located in the municipality of Sinop, Mato Grosso State, Brazil. To reach the proposed aim, 165 trees with 51 months of age were scaled for the taper modeling with mixed effects models and artificial neural networks. The performance of these techniques was evaluated through precision measurements and graphical analysis. Mixed effects modeling and artificial neural networks are efficient and recommended in the estimative of taper of eucalyptus in integrated Crop-Livestock-Forestry system; however, despite both evaluated techniques present accurate results in predicting the taper of the sampled trees, the artificial neural network predicts values with greater precision than the modeling of mixed effects. [ABSTRACT FROM AUTHOR]

Precipitation, in short periods of time, is a phenomenon associated with high levels of uncertainty and variability. Given its nature, traditional forecasting techniques are expensive and computationally demanding. This paper presents a soft computing technique to forecast the occurrence of rainfall in short ranges of time by artificial neural networks (ANNs) in accumulated periods from 3 to 7 days for each climatic season, mitigating the necessity of predicting its amount. With this premise it is intended to reduce the variance, rise the bias of data and lower the responsibility of the model acting as a filter for quantitative models by removing subsequent occurrences of zeros values of rainfall which leads to bias the and reduces its performance. The model were developed with time series from ten agriculturally relevant regions in Brazil, these places are the ones with the longest available weather time series and and more deficient in accurate climate predictions, it was available 60 years of daily mean air temperature and accumulated precipitation which were used to estimate the potential evapotranspiration and water balance; these were the variables used as inputs for the ANNs models. The mean accuracy of the model for all the accumulated periods were 78% on summer, 71% on winter 62% on spring and 56% on autumn, it was identified that the effect of continentality, the effect of altitude and the volume of normal precipitation, have an direct impact on the accuracy of the ANNs. The models have peak performance in well defined seasons, but looses its accuracy in transitional seasons and places under influence of macro-climatic and mesoclimatic effects, which indicates that this technique can be used to indicate the eminence of rainfall with some limitations. [ABSTRACT FROM AUTHOR]

This paper empirically investigates the main drivers of airline network concentration in an air transport market subject to rapid growth. We consider the Brazilian air transport industry of the 2000s, in which network concentration rapidly increased and was followed by a period of massive flight delays and cancelations, which resulted in the “big blackout” of 2006–2007. We develop an econometric model of network concentration, accounting for demand, cost and competition variables that may affect the propensity of carriers to concentrate flights and passenger connections on a few airports of a network. The main focus of the paper is on the relation between networks leading to the problems of the blackout episode. We investigate the dynamic pattern of the evolution of concentration before and after the abnormal period of operations and find that concentration began to rise at least six quarters before, and persisted at a high level until two quarters after the blackout – and then plunged steeply toward the end of the decade. We believe that our analysis contributes to an improved understanding of the behavior of air transport systems subject to network concentration and congestion. With respect to methodology, we suggest and employ the use of alternative measures of network concentration to check the robustness and validity of our results. [ABSTRACT FROM AUTHOR]

Estimating the amount of materials necessary for deploying electrical systems in buildings is a difficult task since these systems are complex and highly correlated with the design of other building systems. In Brazil, the usual method for estimating the consumption of electrical materials relies on constant rates per electrical points, incurring in large estimation errors. Furthermore, these rates can only be used in advanced project phases. This paper proposes a novel method for estimating the consumption of electrical materials, based on artificial neural networks (ANNs), by using information from early project stages. A dataset with project attributes was used to construct the ANN estimation models. In order to evaluate these new models, their estimates were compared to results from linear regression and constant rate models. Results showed that the ANN models have better performance when compared to other methods, supporting that ANNs are well suited for nonlinear and multidimensional problems. This investigation, therefore, supports that the method proposed could also be used by construction companies for estimating the consumption of other materials. [ABSTRACT FROM AUTHOR]

Perform forecasts is an activity that allows the industrial sector or services scale their needs to better attend the customers. For that, forecasters are always searching for more accurate predictions. Thus this paper presents a study through the data series of the percentage of the amount of outpatient procedures payment not approved by the Unified Health System (SUS) in relation to the total reported to that system, a research to compare the forecast techniques performance of this demand. The methods used in this study were: model of exponential smoothing and artificial neural networks, average forecast combination and regression forecast combination. It was possible to observe that the regression combination method showed mild superiority over the other techniques. [ABSTRACT FROM AUTHOR]

• An innovative framework for energy benchmarking of buildings is proposed. • Entropy and cluster analysis reduced uncertainty in setting parameters of archetypes. • The machine learning technique used as a benchmarking tool showed high performance. • The benchmarking applied in actual buildings showed a high inefficiency ratio. Energy benchmarking of buildings has an important role in improving energy performance by establishing a reference for the energy efficiency of the building stock. The simulation of archetypes followed by a generalisation model has been widely used to obtain benchmarks. However, even though archetypes summarise the main features of the building stock, the uncertainties must be accounted for in the modelling process. Moreover, testing the response of the benchmarking model using the actual building stock data supports the reliability of the method. This paper aims to propose an innovative framework to reduce the uncertainty of archetypes for benchmarking buildings. A standard framework for data compiling is proposed and an assessment of the uncertainty of variables using entropy and cluster analysis allowed to obtain representative archetypes. An Artificial Neural Network (ANN) was used as a predictive tool, and it was applied to benchmark a sample of actual buildings. Also, the simulation outcomes were used to determine energy end-uses according to the climatic zones. The framework proposition is presented alongside with a practical application. The result is an unprecedented benchmarking model: the archetype considers more variation in parameters of the building stock with higher uncertainties. Additionally, the modelling process showed to be robust for combining different datasets, and the ANN achieved high-performance metrics. Conclusion indicates the potential of using the framework for other typologies. Moreover, the framework demonstration is used for a school stock in Brazil, showing a trend to the inefficiency while a specific case study was explored, showing the potential of the method to find faults in the building energy use. [ABSTRACT FROM AUTHOR]

• A novel strategy using only sample points to delineate polygons of soil classes. • GEOBIA is a promising approach for digital soil mapping. • The GEOBIA approach improves soil class prediction. The use of techniques to calibrate soil prediction models based on legacy maps is restricted to areas where conventional survey and soil mapping were performed. It is necessary to seek alternatives to calibrate soil predictive models without a legacy soil map. The problem is the difficulty to delineate polygons of soil classes based on soil sampling points. This paper presents a novel digital soil mapping strategy using only georeferenced points of soil profiles to delineate detailed polygons of soil classes by the Geographic Object-Based Image Analysis (GEOBIA) approach and Artificial Neural Networks (ANN) models. The main objective was to evaluate the integration of the GEOBIA approach at different segmentation levels with ANN's models and sampling points to produce a digital soil map of the Vale dos Vinhedos region, in Rio Grande do Sul, Brazil. From a Digital Terrain Model, 10 predictive variables were extracted. From the RapidEye remote sensing image the following spectral variables were extracted: Normalized Difference Vegetation Index and Normalized Difference Water Index. A new set of variables was produced using principal component analysis after standardization of the original variables. The multiresolution segmentation algorithm was used to create image objects at different segmentation levels. Four segmentation levels were tested with scale parameter (SP) 1, 2, 5 and 10; and the Cheesboard segmentation (CS) was used to transform the original pixels into polygons. Information on types of soil was obtained from 163 georeferenced soil profiles. Five ANN's structures were implemented: four for the segmented data (SP01, SP02, SP05 and SP10) and one for the per-pixel approach (CS); 16 repetitions of each ANN were performed. With the best models for each segmentation level, extrapolations of soil classes were performed for the entire study area and validated with a detailed legacy soil map. The per-pixel approach model obtained the worst result. The best result in the extrapolation of soil types found was with the object-based approach (SP1). In our understanding, the scientific community can profit from the proposed methodology particularly when it only has georeferenced points of soil classes and not a legacy map, i.e., immediately after a field survey without the need for photointerpretation for delineation of the soil mapping units. The use of GEOBIA for digital mapping of soil classes seems to be a promising approach for future research. [ABSTRACT FROM AUTHOR]

This paper estimates intersectoral labor efficiencies and its covariances in Brazil from 2003 to 2016 using a two-stage neural network model. It investigates whether there is a positive relationship between education and productivity. At the first stage, a static Markov chain regression provides labor quantitative efficiency (volume) and labor value efficiency (value added). The second stage runs a dynamic regression model between each estimated efficiency and the social contextual variables to unveil endogeneities from covariances among the variables set. Strong covariances are found between the efficiencies and fertility rate, suggesting that there is a relevant gap between the productive sector and worker qualification, leading to lower levels of efficiency in Brazil, as shown in the theoretical model. The economy is unable to allocate efficiently the stock of qualified workers. • Usage of NDEA model to evaluate Brazilian labor productivity. • Usage of Markov neural chains to unveil the endogeneities between social variables and efficiency indexes. • Proposition of economic model of intersectoral efficiency based on structural factors and social variables. • Dynamic model based on two sectors – tradables and non-tradables. [ABSTRACT FROM AUTHOR]

Forecasting photovoltaic electricity generation is one of the key components to reducing the impacts of solar power natural variability, nurturing the penetration of renewable energy sources. Machine learning is a well-known method that relies on the principle that systems can learn from previously measured data, detecting patterns which are then used to predict future values of a target variable. These algorithms have been used successfully to predict incident solar irradiation, but the results depend on the specificities of the studied location due to the natural variability of the meteorological parameters. This paper presents an extensive comparison of the three ML algorithms most used worldwide for forecasting solar radiation, Support Vector Machine (SVM), Artificial Neural Network (ANN), and Extreme Learning Machine (ELM), aiming at the best prediction of daily solar irradiance in a São Paulo context. The largest dataset in Brazil for meteorological parameters, containing measurements from 1933 to 2014, was used to train and compare the results of the algorithms. The results showed good approximation between measured and predicted global solar radiation for the three algorithms; however, for São Paulo, the SVM produced a lower Root-Mean-Square Error (RMSE), and ELM, a faster training rate. Using all 10 meteorological parameters available for the site was the best approach for the three algorithms at this location. [ABSTRACT FROM AUTHOR]

In recent years, renewable and sustainable energy sources have attracted the attention of various investors and stakeholders, such as energy sector agents and even consumers. It is perplexing to observe and anticipate the required levels of photovoltaic generation, which are inherent tasks for such rapid insertion into the electric grid. This distributed/renewable generation must be integrated in a coordinated way such that there is no negative impact on the electric performance of the grid, increasing in the complexity of energy management. In this article, a methodology for photovoltaic generation forecasting is addressed for a horizon of one week ahead, using a new approach based on an artificial neural network (ANN) ensemble. Two main questions will be explored with this approach: how to select the ANNs, and how to combine them in the ensemble. The design of experiments (DOE) approach is applied to the photovoltaic time series factors and ANN factors. Then, a cluster analysis is performed to select the best networks. From this point on, a mixture (MDE) is employed to determine the ideal weights for the ensemble formation. The methodology is detailed throughout the paper and, based on the combination of forecasts, the photovoltaic generation was estimated for a specific panel set located in the state of Minas Gerais, Brazil, reaching the value of 4.7% for the weekly mean absolute percentage error. The versatility of the proposed method allowed the change of the number of factors to be used in the experimental arrangement, the forecast model, and the desired forecast horizon, and consequently enhancing the forecasting determination. • An ANN ensemble is created for medium-term photovoltaic generation forecasting. • ANNs are selected by using a new design of experiments approach. • The ANN ensemble is built by using a new mixture design of experiments approach. • We predicted the photovoltaic generation for a panel set located in Brazil. [ABSTRACT FROM AUTHOR]

The availability of freshwater is becoming a global concern. Because agricultural consumption has been increasing steadily, the mapping of irrigated areas is key for supporting the monitoring of land use and better management of available water resources. In this paper, we propose a method to automatically detect and map center pivot irrigation systems using U-Net, an image segmentation convolutional neural network architecture, applied to a constellation of PlanetScope images from the Cerrado biome of Brazil. Our objective is to provide a fast and accurate alternative to map center pivot irrigation systems with very high spatial and temporal resolution imagery. We implemented a modified U-Net architecture using the TensorFlow library and trained it on the Google cloud platform with a dataset built from more than 42,000 very high spatial resolution PlanetScope images acquired between August 2017 and November 2018. The U-Net implementation achieved a precision of 99% and a recall of 88% to detect and map center pivot irrigation systems in our study area. This method, proposed to detect and map center pivot irrigation systems, has the potential to be scaled to larger areas and improve the monitoring of freshwater use by agricultural activities. [ABSTRACT FROM AUTHOR]

In this paper, laser-induced breakdown spectroscopy (LIBS) combined with artificial neural network (ANN) was investigated to classify four species of wood samples (Africa rosewood, Brazil bubinga, Myanmar padauk, and Pterocarpus erinaceus). The wood samples were ablated by laser pulses to generate plasma emission, which was measured by a spectrometer and transmitted into a computer for further data analysis. The feature spectral data were selected out based on loadings of principal component analysis (PCA) and normalized using the sum of all feature spectra data. The ANN model was built based on the feature spectral data to classify the wood species. The relationship between correct classification rate (CCR) and settings of ANN was discussed. The CCR of ANN model for test set data achieved 100% with multilayer perceptron network and Broyden–Fletcher–Goldfarb–Shanno iterative algorithm. This result was also compared with the CCRs of PLS-DA, KNN, and SIMCA model for test set (82.5%, 95.83%, and 51.67%, respectively). Using the ratio between feature variables to recognize the species of wood was also discussed. The experimental results demonstrated that LIBS integrated with ANN could be applied for analyzing and recognizing wood species. [ABSTRACT FROM AUTHOR]