Search

The transition to a low-carbon economy is one of the main challenges of our time. In this context, solar energy, along with many other technologies, has been developed to optimize performance. For example, solar trackers follow the sun's path to increase the generation capacity of photovoltaic plants. However, several factors need consideration to further optimize this process. Important variables include the distance between panels, surface reflectivity, bifacial panels, and climate variations throughout the day. Thus, this paper proposes an artificial intelligence-based algorithm for solar trackers that takes all these factors into account—mainly weather variations and the distance between solar panels. The methodology can be replicated anywhere in the world, and its effectiveness has been validated in a real solar plant with bifacial panels located in northeastern Brazil. The algorithm achieved gains of up to 7.83% on a cloudy day and obtained an average energy gain of approximately 1.2% when compared to a commercial solar tracker algorithm. [ABSTRACT FROM AUTHOR]

Increasingly stringent pollutant emission regulations and a customer demand for a high-fuel economy drive the modern automotive industry to hurriedly solve the problem of decarbonization and powertrain efficiency, leading R&D towards alternative powertrain solutions and fuels. Electrification, today, plays the biggest role in the topic, with Mild Hybrid Electrified Vehicles (MHEVs) being the most cost-effective architectures, displaying dominance in smaller markets such as Brazil. One of the biggest challenges for HEVs' development is the complexity of the hybrid control system, knowing when to actuate the electric machine, and the optimum power delivery, plus the gearshift schedule becomes a hard optimization problem that plays a key role in powertrain efficiency and cost savings for the customer. This paper proposes the implementation of a genetic algorithm (GA) as a machine learning-based control strategy to determine the torque split and the gear engaged for each driving condition of an MHEV operation, aiming to optimize fuel consumption. A quasi-static model of the vehicle was developed in Matlab/Simulink version 2022b, the virtual vehicle was then tested following the FTP75 and HWFET driving cycles. Simulation results indicate that the control decisions taken by the GA are qualitatively coherent for all operation conditions, and even quantitatively coherent in some cases, and that the software has the potential to be used as a control strategy outside the simulation environment, in future steps of development. [ABSTRACT FROM AUTHOR]

The Coronavirus Disease 2019 (COVID-19) pandemic has had social and clinical effects over the healthcare system. Globally, over 314.019.135 confirmed cases and 5.507.370 deaths have been recorded at 12 January 2022, based on the Dashboard updated by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University. The continuous rise in the number of asymptomatic, pre-symptomatic, and symptomatic patients has developed efficient and accessible models for prediction using Open Source Libraries and Cloud Services. This paper proposes different machine-learning algorithms to analyse the COVID-19 OpenData Resources from Mexico and Brazil, which represent the two major populations, economics and affected by disease countries in Latin America. This model uses only the COVID-19 patient's geographical, social conditions, economics conditions, clinical risk factors, symptom reports, and demographic data to predict recovery and death. The model of Mexico has an accuracy of 93% and the perceived mean of recall and the precision (F1 Score) of 0.79 on the dataset used. On the other hand, the model of Brazil has an accuracy of 69% and F1 score of 0.75 on the dataset studied. The result considers data from patients under the age of 0 and 120 years. The contribution of the work is the application of Big Data technologies and Machine Learning algorithms using Open Resource Libraries and Amazon Web Services (AWS) with the vision to improve the clinical diagnosis, even infectious disease with mathematical approaches. [ABSTRACT FROM AUTHOR]

The global energy market has significantly developed in recent years; proof of this is the creation and promotion of smart grids and technical advances in energy commercialization and transmission. Specifically in the Brazilian context, with the recent modernization of the electricity sector, energy trading prices, previously published on a weekly frequency, are now available on an hourly domain. In this context, the definition and forecasting of prices become increasingly important factors for the economic and financial viability of energy projects. In this scenario of changes in the local regulatory framework, there is a lack of publications based on the new hourly prices in Brazil. This paper presents, in a pioneering way, the Dynamic Time Scan Forecasting (DTSF) method for forecasting hourly energy prices in Brazil. This method searches for similarity patterns in time series and, in previous investigations, showed competitive advantages concerning established forecasting methods. This research aims to test the accuracy of the DTSF method against classical statistical models and machine learning. We used the short-term prices of electricity in Brazil, made available by the Electric Energy Commercialization Chamber. The new DTSF model showed the best predictive performance compared to both the statistical and machine learning models. The DTSF performance was superior considering the evaluation metrics utilized in this paper. We verified that the predictions made by the DTSF showed less variability compared to the other models. Finally, we noticed that there is not an ideal model for all predictive 24 steps ahead forecasts, but there are better models at certain times of the day. [ABSTRACT FROM AUTHOR]

Studies comparing large-scale assessment data among educational systems have been an important tool for understanding the differences in how education is delivered worldwide. Many of these studies do not go beyond reporting average student scores in a particular educational system. A more unbiased analysis would avoid the simple use of gross performance and consider educational system contexts. A common approach is to estimate effectiveness by the residuals of parametric linear models. These models rely upon strong assumptions regarding the data-generating process, and are limited to handling extensive datasets. To address this issue, our paper provides a new approach based on machine learning models. The new approach is flexible, allows paired comparison, and is model-independent. An analysis conducted in Brazil verifies the suitability of the method to explore differences in effectiveness between Brazilian educational administrative units at the regional and state levels from 2009 to 2019. Our results are consistent with the existing literature, but the methodology produced a number of new findings that were not observed in studies using more traditional approaches. [ABSTRACT FROM AUTHOR]

This paper proposes a new hybrid SVR-ANN model for water demand forecasting. Where an adaptation of the methodology proposed by Zhang (Neurocomputing 50:159–175, 2003) is used to decompose the time series of 10 reservoirs that supply the Metropolitan Region of Salvador (RMS). The data used are from the historical consumption from January 2017 to February 2022, obtained from the local supply company, Empresa Baiana de Águas e Saneamento, and meteorological data obtained from the National Institute of Meteorology of Brazil. The results demonstrated the feasibility of using the proposed model, compared to other traditional models such as the multilayer perceptron (MLP), support vector regression (SVR), short long-term memory (LSTM) and autoregressive integrated moving average (ARIMA). [ABSTRACT FROM AUTHOR]

The delineation of management zones is one of the ways to enable the spatially differentiated management of plots using precision agriculture tools. Over the years, the spatial variability of data collected from soil and plant sampling started to be replaced by data collected by proximal and orbital sensors. As a result, the variety and volume of data have increased considerably, making it necessary to use advanced computational tools, such as machine learning, for data analysis and decision-making support. This paper presents a methodology used to establish management zones (MZ) in precision agriculture by analyzing data obtained from soil sampling, proximal sensors and orbital sensors, in experiments carried out in four plots featuring soybean–cotton and soybean–corn crops, in Mato Grosso and Paraná states, Brazil. Four procedures were evaluated, using different input data sets for the MZ delineation: (I) soil attributes, including clay content, apparent electrical conductivity or fertility, along with elevation, yield maps and vegetation indices (VIs) captured during the peak crop biomass period; (II) soil attributes in conjunction with VIs demonstrating strong correlations; (III) solely VIs exhibiting robust correlation with soil attributes and yield; (IV) VIs selected via random forests to identify the importance of the variable for estimating yield. The results showed that the VIs derived from satellite images could effectively replace other types of data. For plots where the natural spatial variability can be easily identified, all procedures favor obtaining MZ maps that allow reductions of 40% to 70% in yield variance, justifying their use. On the other hand, in plots with low natural spatial variability and that do not have reliable yield maps, different data sets used as input do not help in obtaining feasible MZ maps. For areas where anthropogenic activities with spatially differentiated treatment are already present, the exclusive use of VIs for the delineation of MZs must be carried out with reservations. [ABSTRACT FROM AUTHOR]

[Display omitted] • Assessment of the status and trends of PV DG in Brazil based on big data processing; • Application of machine learning algorithms to estimate the installed power of individual systems; • Application of forecasting models to predict the installed power of PV DG over time in Brazil and its regions; • Application of the DEA method to rank Brazilian states in terms of efficiency in PV DG deployment; • The paper is expected to be valuable for regulatory agencies, researchers, and PV aggregator companies. The deployment of photovoltaic (PV) distributed generation (DG) has been increasing substantially in Brazil. In this context, the present paper exploits a robust dataset provided by the Brazilian Electricity Regulatory Agency (ANEEL) to evaluate the status and trends concerning PV DG. As of Nov/2022, such a dataset consists of more than 1,4 million lines (one for each system) and thirty columns (various information concerning the systems). For an in-depth assessment, three items are addressed: (i) the application of machine learning algorithms to estimate the installed power of individual systems based on other information available in the dataset, (ii) the application of forecasting models to predict the installed power of PV DG over time in Brazil and its regions, and (iii) the application of the data envelopment analysis (DEA) method to rank Brazilian states in terms of efficiency in PV DG deployment. While items (i)-(iii) present distinct specific goals, they use the same dataset and provide essential insights concerning PV DG in Brazil. In item (i), elastic net (EN), decision tree (DT), random forest (RF), extra tree (ET), AdaBoost (AB), and gradient boosting (GB) are applied to select the most accurate algorithm. In item (ii), autoregressive integrated moving average (ARIMA), seasonal autoregressive integrated moving average (SARIMA), Holt-Winters exponential smoothing (HWES), Bass diffusion model (BDM), and multilayer perceptron artificial neural network (MLP-ANN) are applied. In item (iii), output-oriented DEA CCR and BCC are applied. Concerning item (i), results demonstrate that estimating the installed power of individual systems is not very simple. Nonetheless, the machine learning algorithms imply a significant accuracy increase compared to taking the dataset's average as the estimation, especially the DT and RF algorithms with a coefficient of determination of 29%. Regarding item (ii), results demonstrate that the installed capacity in Brazil and its regions is expected to increase approximately linearly in the horizon of months but with weekly seasonal characteristics. Moreover, the most accurate model depends on the region, i.e., the ANN is superior for the Northeast (MAPE of 3.25%) and South (MAPE of 1.28%) regions, the HWES method is superior for Brazil (MAPE of 0.51%) and the Southeast region (MAPE of 2.26%), and the BDM is superior for the Mid-west (MAPE of 1.95%) and North (MAPE of 3.37%) regions. Concerning item (iii), results demonstrate that DEA CCR is more representative of the Brazilian reality since DEA BCC assigns several states as efficient, even if they showcase a very low installed capacity, particularly small states. DEA CCR indicates that Minas Gerais, Rio Grande do Sul, and Mato Grosso are the most efficient states in PV DG deployment relative to their inputs, closely followed by São Paulo with an efficiency of 95%. Furthermore, states located in the North and Northeast region require more attention from ANEEL since their efficiencies are generally low. A literature review demonstrated that this paper is the first of its kind in exploiting so thoroughly and robustly the dataset provided by ANEEL. Therefore, the paper is expected to assist ANEEL in the task of properly regulating the sector and also assist researchers and PV aggregator companies through in-depth insights. Moreover, the conducted procedures can also be applied to other countries with available datasets, thus ensuring high applicability for the research. [ABSTRACT FROM AUTHOR]

User eXperience (UX) has been used to achieve improvements in digital information systems based on how people perceive them. In particular, this paper establishes a framework that employs methods for eye and mouse tracking, keyboard input, self-assessment questionnaire and artificial intelligence algorithms to evaluate user experience and categorize users in terms of performance profiles. The results obtained with this framework are artifacts that can be used to support customizations of the User Interface (UI) on the websites. Moreover, the established framework is generic and flexible and can be applied to any information system, such as the case study shown in the website of the Federal Revenue of Brazil (RFB). The main objectives of this paper are as follows: (i) to set out a powerful UX framework based on three tracking techniques – the AIT2-UX; (ii) to provide the T2-UXT to collect, collate, process and visualize data obtained from users' interactions (iii) to use and compare machine learning algorithms with the classification of user performance profiles; (iv) to use the artifacts generated by the framework to manually customize the UI with the website. [ABSTRACT FROM AUTHOR]

The COVID-19 pandemic continues to have a destructive effect on the health and well-being of the global population. A vital step in the battle against it is the successful screening of infected patients, together with one of the effective screening methods being radiology examination using chest radiography. Recognition of epidemic growth patterns across temporal and social factors can improve our capability to create epidemic transmission designs, including the critical job of predicting the estimated intensity of the outbreak morbidity or mortality impact at the end. The study's primary motivation is to be able to estimate with a certain level of accuracy the number of deaths due to COVID-19, managing to model the progression of the pandemic. Predicting the number of possible deaths from COVID-19 can provide governments and decision-makers with indicators for purchasing respirators and pandemic prevention policies. Thus, this work presents itself as an essential contribution to combating the pandemic. Kalman Filter is a widely used method for tracking and navigation and filtering and time series. Designing and tuning machine learning methods are a labor- and time-intensive task that requires extensive experience. The field of automated machine learning Auto Machine Learning relies on automating this task. Auto Machine Learning tools enable novice users to create useful machine learning units, while experts can use them to free up valuable time for other tasks. This paper presents an objective method of forecasting the COVID-19 outbreak using Kalman Filter and Auto Machine Learning. We use a COVID-19 dataset of Ceará, one of the 27 federative units in Brazil. Ceará has more than 235,222 confirmed cases of COVID-19 and 8850 deaths due to the disease. The TPOT automobile model showed the best result with a 0.99 of R 2 score. [ABSTRACT FROM AUTHOR]

The subduction and export of Antarctic Intermediate Water (AAIW) is important for the heat, freshwater, carbon, and nutrient budgets of the world's oceans. Three types of AAIW are identified by applying an unsupervised machine learning approach to individual Argo profiles. The likely sources and pathways of these AAIWs are indicated by the locations and magnitudes of the salinity and oxygen extrema. The Southeast Pacific AAIW forms north of the Subantarctic Front (SAF) in the west corner of the Drake Passage and is exported north into the subtropical gyre. The South Pacific AAIW forms in the South Pacific along the SAF and is transported through the Drake Passage, thereby becoming a saliter and denser AAIW. The circumpolar AAIW, the coldest and freshest AAIW, has a circumpolar source and is injected into the ocean interior in the Brazil‐Malvinas confluence. The results clarify different types of AAIW have different origins. Plain Language Summary: The formation and export of Antarctic Intermediate Water (AAIW) strongly influence the global climate system by redistributing heat, freshwater, carbon, and nutrients from high latitudes to the tropics. Thanks to an increasing number of observations from autonomous instruments and Argo floats, our understanding of the sources and pathways of AAIW has deepened in recent decades. Even so, there is controversy about the formation of AAIW. Unlike previous studies using traditional property‐driven classification, we apply an unsupervised classification technique to automatically classify the Argo temperature and salinity profiles and identify three types of AAIWs from the regimes in the Southern Ocean. The classification results are consistent with previous studies. Moreover, we find that different types of AAIW form in different regions. This paper shows that the machine learning technique benefits us in dealing with a large data set and identifying AAIW classes that is hard to distinguish by the traditional property‐driven classification method. Key Points: An unsupervised learning approach, the profile classification model (PCM), is applied to classify Argo temperature and salinity profilesPCM identifies three types of Antarctic Intermediate Water (AAIW) in the Southern Ocean and reveals their locations and boundariesThe Polar Frontal Zone in the South Pacific sector is established as the source region for AAIW formation [ABSTRACT FROM AUTHOR]

Mineral prospectivity mapping (MPM), like other geoscience fields, is subject to a variety of uncertainties. When data about unfavorable sites to find deposits (i.e., drill intersections to barren rocks) is lacking in MPM using machine learning (ML) methods, the synthetic generation of negative datasets is required. As a result, techniques for selecting point locations to represent negative examples must be employed. Several approaches have been proposed in the past; however, one can never be certain that the points chosen are true negatives or, at the very least, optimal for training. As a consequence, methodologies that account for the uncertainty of the generation of negative datasets in MPM are needed. In this paper, we compare two criteria for selecting negative examples and quantify the uncertainty associated with this process by generating 400 potential maps for each of the three ML methods utilized (200 maps for each criterion), which include random forest (RF), support vector machine (SVM), and k-nearest neighbors (KNC). The results showed that applying a geological constraint to the creation of negative datasets reduced prediction uncertainty and improved overall model performance but produced larger areas of very high probability (i.e., >0.9) when compared to using only the spatial distribution of known deposits and occurrences as a constraint. SHAP values were used to find approximations for the importance of features in nonlinear methods, and kernel density estimations were used to examine how they varied depending on the negative dataset used to train the ML models. Prospectivity models for magmatic-hydrothermal gold deposits were generated using data from the shuttle radar terrain mission, gamma-ray, magnetic lineaments, and proximity to dykes. The Juruena Mineral Province, situated in Northern Mato Grosso, Brazil, represented the case study for this work. [ABSTRACT FROM AUTHOR]

Electricity generation in Brazil heavily depends on hydroelectric power, making it vulnerable to fluctuations due to its reliance on weather patterns. Accurately forecasting water inflow into hydroelectric plants is vital for the National Electric System Operator to make decisions regarding the monthly scheduling and operation of the power system. In this paper, an evaluation of predicted flows for a 14-day horizon are evaluated for the Tucuruíhydroelectric plant, located in the Tocantins river in the North of Brazil. The temporal fusion transformer (TFT), long short-term memory (LSTM), and temporal convolutional networks (TCN) are compared. The findings demonstrate that the TFT is a more suitable alternative than LSTM and TCN models for predicting inflows for the next 14 days. The TFT model is hypertuned by Optuna to achieve an optimized structure (h-TFT). The h-TFT had a mean absolute percentage error of 13.1 and a Nash–Sutcliffe of 0.96, outperforming its initial version and even the bidirectional LSTM model in benchmarking. Based on the error results, h-TFT showed promise for flow forecasting and provides insights into decision-making processes in the Brazilian electricity sector. • Hypertuning the TFT improves its performance in median-term forecasting. • The h-TFT outperforms LSTM, and TCN models for time series forecasting. • Deep learning applied for scheduling operation of the electrical power system. • Energy planning based on the inflow forecasting of hydroelectric power plants. [ABSTRACT FROM AUTHOR]

Knowledge and monitoring of invasive species are fundamental measures to determine the short- and long-term effect on invaded ecosystems, in addition to developing strategies to control the problem or its specific solution. In this context, the lionfish is an invasive species that worries managers and scientists of fisheries and marine conservation, this is due to the affected area that spread starting from the east coast of the United States to the coasts of Brazil and it is recently extending to include the Mediterranean Sea. The diet of the invasive fish is small species of fish, crustaceans and invertebrates; the consequent damage is the decrease of food for species at the next level of the food chain and the lack of species to keep coral reefs healthy. In this paper, we propose a lionfish detection system that will be installed in an autonomous underwater vehicle, as part of a monitoring strategy that will allow real-time determination of the number of Lionfish, their location and without human intervention. We compared two detection systems, namely YOLOv4 and SSD-Mobilenet-v2, by training with cross-validation and evaluation with the test set we obtained the best model with 63.66% recall, 89.79% precision, and 79.15% mAP with images in the natural environment, implemented on NVIDIA's Jetson Nano embedded system. [ABSTRACT FROM AUTHOR]

School dropout is a relevant socio-economic problem across the globe. Predictive models have been developed to determine the likelihood of students dropping out of their studies precociously to overcome such a problem. Academic systems, which gather data from many students, are potential sources for datasets that feed dropout prediction algorithms, thus leading to general improvements in education quality. Despite successful past attempts to predict dropout, several works depict small datasets with features that are hard to reproduce. Furthermore, predicting whether a student will drop out is not enough to diagnose and prevent the problem as it is also necessary to provide potential justifications for the dropout. This paper proposes an approach for creating and enriching a dataset for dropout prediction, which has been applied for dropout prediction using data from 19 schools in Brazil. With this dataset and using classifiers and model explaining techniques, our experiments achieved Area Under the Precision–Recall Curve (AUC-PR) scores of up to 89.5%, Precision up to 95%, Recall up to 93%, and Kolmogorov–Smirnov (KS) rates up to 97% when predicting dropout at different year moments. This study also shows differences when predicting dropouts in different educational stages, such as preschool and secondary education, with the former being more complex than the latter. In addition to the high recognition rates, our proposal identifies potential reasons for student dropout, which are relevant for educational institutions to take preemptive actions. • Dropout prediction is performed with educational, financial, and public data. • Dropout analysis is conducted in different educational stages and time of the year. • Explainable machine learning allows reasoning behind potential dropouts. [ABSTRACT FROM AUTHOR]

• IoT and machine learning promote soil moisture forecast for smart irrigation. • Our approach is fully data-driven and independent of domain-knowledge features. • Results come from twelve fields on four farms, from eight crop types in four years. • Boosting outperforms neural networks, and blended models improve performance. • Poor data quality is a challenge for machine learning soil moisture forecast. The rise of the Internet of Things allowed higher spatial–temporal resolution soil moisture data captured through in situ sensing. Such abundance of data enables machine learning-based soil moisture forecast as an alternative to traditional mechanistic approaches for irrigation water need estimation. This paper develops a guideline for soil moisture forecast modeling based on machine learning, tested in a real case analysis comprehending eight crop types in twelve fields from four farms distributed over diverse climatic scenarios in Brazil. Instead of a single value, we predict the following days' minimum and maximum values as targets to monitor risks of extreme soil moisture values. Furthermore, modeling soil moisture directly in volumetric water content (VWC) is better than modeling soil matric potential (SMP) to later convert in soil moisture VWC. We test several algorithms and find out that LightGBM outperforms linear regression, decision tree, random forest, multilayer perceptron, LSTM, and StemGNN. Also, blending predictions via algorithm ensemble provides an additional accuracy gain. For model training and accuracy measurement, we use weighted datasets to privilege rare but critical data points. We show that soil moisture forecast reaches its maximum performance considering only past soil moisture, a context-aware index, and a precipitation forecast. Finally, we demonstrate that traditional domain-knowledge features - such as evapotranspiration, crop phenology, and soil hydraulic behavior - are not relevant to improving SM forecast performance. Consequently, our paper suggests full data-driven approaches for irrigation water need estimation, observed some care regarding data quality. [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]

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]

• An underfrequency load shedding scheme for high dependability and security tolerant to induction motors dynamics. • Improper operation of under frequency load shedding methods due to the inertia of induction motors. • Patterns classification of under frequency load shedding operate and block cases. • Use of machine learning method of discriminant function analysis for electric power systems protection. The load shedding scheme is a last resort to prevent power system collapse when the available generation is less than the load demand. The traditional load shedding schemes monitors the frequency and voltage by underfrequency and undervoltage relays and provides load cutoff until the frequency returns to a stable point of operation. The reduction of rotating inertia demands faster load shedding to meet the stability requirements. In some electrical power systems in Brazil, it was found that, in the event of opening a transmission line upstream of the load shedding substation, some under frequency relay may operate improperly. This paper proposes a new method to avoid the unwanted operation of load shedding schemes when the utility substation (source) is out of service. The proposed method monitors negative sequence voltage, frequency and its rate of change, enabling faster load shedding (high dependability) without unwanted (high security) operation when compared with traditional methods. Events were classified by linear discriminant analysis and tested in the IEEE 9 bus test system. The tests consisted of several disturbances such as loss of generation and line outages at different levels of feeder loading, reactive compensation and distributed generation. The proposed method reached a 100% rate of correct classification. [ABSTRACT FROM AUTHOR]

Labeling of samples is a recurrent and time-consuming task in data analysis and machine learning and yet generally overlooked in terms of visual analytics approaches to improve the process. As the number of tailored applications of learning models increases, it is crucial that more effective approaches to labeling are developed. In this paper, we report the development of a methodology and a framework to support labeling, with an application case as background. The methodology performs visual active learning and label propagation with 2D embeddings as layouts to achieve faster and interactive labeling of samples. The framework is realized through SoundscapeX, a tool to support labeling in soundscape ecology data. We have applied the framework to a set of audio recordings collected for a Long Term Ecological Research Project in the Cantareira-Mantiqueira Corridor (LTER CCM), localized in the transition between northeastern São Paulo state and southern Minas Gerais state in Brazil. We employed a pre-label data set of groups of animals to test the efficacy of the approach. The results showed the best accuracy at 94.58 % in the prediction of labeling for birds and insects; and 91.09 % for the prediction of the sound event as frogs and insects. [ABSTRACT FROM AUTHOR]

In sucker-rod pumping wells, due to the lack of an early diagnosis of operating condition or sensor faults, several problems can go unnoticed. These problems can increase downtime and production loss. In these wells, the diagnosis of operation conditions is carried out through downhole dynamometer cards, via pre-established patterns, with human visual effort in the operation centers. Starting with machine learning algorithms, several papers have been published on the subject, but it is still common to have doubts concerning the difficulty level of the dynamometer card classification task and best practices for solving the problem. In the search for answers to these questions, this work carried out sixty tests with more than 50,000 dynamometer cards from 38 wells in the Mossoró, RN, Brazil. In addition, it presented test results for three algorithms (decision tree, random forest and XGBoost), three descriptors (Fourier, wavelet and card load values), as well as pipelines provided by automated machine learning. Tests with and without the tuning of hypermeters, different levels of dataset balancing and various evaluation metrics were evaluated. The research shows that it is possible to detect sensor failures from dynamometer cards. Of the results that will be presented, 75% of the tests had an accuracy above 92% and the maximum accuracy was 99.84%. [ABSTRACT FROM AUTHOR]