Latin-American Journal of Computing

A study on the impact of data balance on rainfall prediction through artificial neural networks using surface microwave radiometers

2024-07-08T14:49:26+00:00

The National Institute for Space Research (INPE) has been a partner in significant projects that conduct atmospheric investigations impacting various sectors, such as the Amazon Tall Tower Observatory (ATTO) project. Since 2009, the project has conducted studies on the interactions between climate and the Amazon forest. ATTO has played an essential role in providing large volumes of data obtained by meteorological sensors, contributing to a deeper understanding of the atmospheric dynamics of the region. In a landscape where Artificial Intelligence-based rainfall forecast models gain prominence, this study explores the imbalance of data from the ATTO Campina field experiment and its influence on short-term rainfall forecasts using Artificial Neural Networks (ANNs). Metrics such as MAE, RMSE, and POD, as well as FAR indices, were applied in the assessment and revealed the connection between data balance and forecast results. More balanced data or data with greater weights for different rainfall ranges yield better results. The study emphasizes the importance of reliable data for training rain forecast models, aiming to improve the dexterity of these models. This approach is fundamental to increase the reliability of these models in real environments.

Estimation of Spatially Dependent Coeficients in Heterogeneous Media in Diffusive Heat Transfer Problems

2024-07-08T14:49:26+00:00

This article addresses the solution to the inverse problem in a one-dimensional transient partial differential equation with a source term, commonly encountered in heat transfer modeling for diffusion problems. The equation is utilized in a dimensionless form to derive a more general solution that is applicable across various contexts. The Transition Markov Chain Monte Carlo (TMCMC) method is utilized to estimate spatially variable thermophysical properties within the equation. This approach involves transitioning between probability densities, gradually refining the prior distribution to approximate the posterior distribution. The results indicate the effectiveness of the TMCMC method in addressing this inverse problem, offering a robust methodology for estimating spatially variable coefficients.

ANN-MoC Method for Inverse Transient Transport Problems in One-Dimensional Domain

2024-07-08T14:49:26+00:00

Transport problems of neutral particles have important applications in engineering and medical fields, from safety and quality protocols to optical medical procedures. In this paper, the ANN-MoC approach is proposed to solve the inverse transient transport problem of estimating the absorption coefficient from scalar flux measurements at the boundaries of the model domain. The central idea is to fit an Artificial Neural Network (ANN) using samples generated by direct solutions computed by a Method of Characteristics (MoC) solver. The direct solver validation is performed on a manufactured solution problem. Two inverse problems are then presented for testing the ANN-MoC method. In the first, a homogeneous medium is assumed, and, in the second, the medium is heterogeneous with a piecewise constant absorption coefficient. We show that the method can achieve good estimates, with accuracy depending on that of the direct solver. We also include a test of sensibility by studying the propagation of noise on the input data. The results highlight the potential of the proposed method to be applied to a broader range of inverse transport problems.

Exploring Digital Twins of Nonlinear Systems through Meta-Modeling with Echo State Networks

2024-07-10T19:57:13+00:00

monitoring, and control rely on precise dynamic models that can capture the inherent nonlinearities of chemical systems. However, rigorous modeling of complex industrial processes can be computationally demanding. Meta modeling using machine learning methodologies offers a viable approach to generate computationally efficient surrogate representations. Specifically, Echo State Networks (ESNs) are a promising neural network approach for meta-modeling nonlinear dynamical systems. ESNs simplify training through fixed input weights while they focus learning on output weights. This study explores the development of ESN-based digital twins for a nonlinear dynamic process. An ESN is employed to construct a meta-model of a simulated continuously stirred tank reactor with biochemical kinetic. The network was trained on input-output data obtained from the simulation of an ordinary differential equation system, and the performance was evaluated both in-sample and out-of-sample. The results indicate that the ESN meta-model can successfully approximate the underlying dynamics, accurately capturing temporal evolution. A closed-loop digital twin deployment using the ESN surrogate also showed reliable behavior. This work presents initial steps toward developing digital twins of chemical processes using ESN-driven meta-modeling. The findings suggest ESNs can effectively generate computationally efficient surrogate representations of nonlinear dynamical systems. Such digital twins hold promise for online process monitoring and optimized control of industrial plants.

Forensic Investigation in Robots

2024-07-08T14:49:26+00:00

Integrating robots into industrial automation has led to a revolutionary transformation in executing complex tasks, harnessing precision and efficiency. The Robot Operating System (ROS) has played a significant role in driving this advancement. ROS Bag files in robots are crucial for preserving data, as they provide a format for recording and playing back ROS message data. These files serve as a comprehensive log of a robot's sensory inputs and operational activities, enabling detailed analysis and reconstruction of the robot's interactions and performance over time. However, there have been instances where security considerations were overlooked, giving rise to concerns about unauthorized access, data theft, and malicious actions. This research investigates the forensic potential of data generated by robots, with a particular focus on ROS Bag data. By analyzing ROS Bag data, we aim to uncover how such information can be used in forensic investigations to reconstruct events, diagnose system failures, and verify compliance with operational protocols. The components of the ROS ecosystem were examined, identifying the challenges in parsing ROS Bag files and underscoring the need for specialized tools. This analysis highlights the security risks associated with plain text communication within legacy ROS systems, emphasizing the importance of encryption. While providing valuable insights, this research calls for further exploration, tool development, and enhanced security practices in robotics and digital forensics, aiming to lay the foundation for effective crime resolution involving robots.

A Comparative Study Between the Brazilian Stock Market and Cryptocurrencies

2024-07-08T14:49:26+00:00

The Brazilian Stock Market has been experiencing an increase in trading volume, and this shows an improvement in indices. This phenomenon is due to the adoption of Corporate Governance practices, improvement in institutional environments, and greater liquidity in national markets. In this scenario, blockchain technology has become popular in recent years, with various applications, ensuring transaction identification, authenticity, reliability, transparency, equity, and interoperability, along with the emergence of smart contracts. However, the most well-known cryptocurrency is Bitcoin, followed by Ethereum, which was the first to allow the use of smart contracts, and Solana, created in 2018, already holds the fourth position, with great expectations for future growth. The popularization of this asset class may represent an investment opportunity; on the other hand, there is research on its possible relationship with other markets and assets, such as gold, the dollar, or even the Dow Jones index. However, the literature on this subject lacks broader research regarding the Brazilian economy, which, being less stable than those markets known as strong, may present different results. This is the aim of the research to compare three cryptocurrencies (Bitcoin, Ethereum, and Solana) with the Brazilian stock market by means of the non-parametric statistical test Kolmogorov-Smirnov.

Editorial

2024-07-08T14:49:26+00:00

We are pleased to share Volume 11, Issue 2 of the Latin American Journal of Computing (LAJC) with you. This edition includes a selection of pioneering research articles that demonstrate the latest advancements in the computer science field. Each paper included in this volume represents rigorous academic research and innovative problem-solving methods. We believe that the insights and discoveries presented here will significantly contribute to the field, stimulate insightful discussions, and inspire future innovations.

This issue begins with three articles that explore advanced methodologies in process monitoring, heat transfer, and robotics. The first article investigates the use of Echo State Networks (ESNs) to create digital twins for nonlinear dynamic chemical processes, demonstrating the potential of ESNs in generating efficient surrogate models for real-time process monitoring and control. The second article addresses the inverse problem in heat transfer modeling using the Transition Markov Chain Monte Carlo method, showcasing its effectiveness in estimating spatially variable thermophysical properties. Next, Janarthanan et al. explore the potential of data generated by robots, specifically focusing on ROS Bag files used in the Robot Operating System (ROS). The study highlights security concerns, such as unauthorized access and data theft, due to plain text communication in legacy ROS systems.

This issue also delves into the critical applications of artificial intelligence and machine learning in various scientific and industrial domains. The fourth article presents the ANN-MoC approach for solving inverse transient transport problems, showcasing its potential in engineering and medical fields by accurately estimating absorption coefficients from scalar flux measurements. Next, another study explores the impact of data balance on short-term rainfall forecasts using Artificial Neural Networks (ANNs) with data from the Amazon Tall Tower Observatory (ATTO). This research emphasizes the necessity of balanced data to improve the accuracy and reliability of meteorological models, highlighting the broader implications for environmental monitoring and prediction. Additionally, the volume includes an innovative fault classification model for industrial processes, merging Decision Trees with Genetic Programming to enhance preventive and corrective measures.

Finally, we explore financial markets and technological advancements. One article compares the Brazilian stock market with cryptocurrencies like Bitcoin, Ethereum, and Solana, using the Kolmogorov-Smirnov test to examine their relationships and potential investment opportunities. The last study uses machine learning and the Grey Wolf Optimization meta-heuristic to predict Brazil's electricity demand, showcasing advanced regression models for accurate energy consumption forecasting.

We hope that the diverse range of topics and innovative approaches presented in this volume will inspire your own research endeavors. The advancements in computational intelligence, machine learning, and data analysis showcased here underscore the transformative potential of these technologies in addressing real-world challenges. As we continue to explore the frontiers of computer science, we invite you to join us in pushing the boundaries of knowledge within our scientific community. Together, we can drive progress and make meaningful contributions to the field.

Classification of Failure Using Decision Trees Induced by Genetic Programming

2024-07-08T14:49:26+00:00

Fault classification in industrial processes is of paramount importance, as it allows the implementation of preventive and corrective measures before catastrophic failures occur, which can result in significant repair costs and production loss, for example. Therefore, the purpose of this study was to develop a classification model by merging the concepts of Decision Trees with Genetic Programming. To accomplish this, the proposed model randomly generates a set of decision trees using the adapted Tennessee Eastman dataset. The generation of these trees does not rely on classical construction logic; instead, they employ an approach where the structure and characteristics of the trees are randomly determined and adjusted throughout the evolutionary process. This approach enables a broader exploration of the search space and may lead to diverse solutions. The results obtained were moderate, largely due to the high number of target classes for classification (21 classes), resulting in the creation of complex trees. The average accuracy on the test data was 0.75, indicating the need to implement new alternatives and enhancements in the algorithm to improve the results.

Electricity Energy Demand Prediction Using Computational Intelligence Techniques

2024-07-08T14:49:26+00:00

Energy is an important pillar for the economic development of a country. The demand for electricity is something that continues to grow, one of the contributing factors is the emergence of various technological equipment and the consequent use by the population. There are several resources that can be exploited to generate electricity, with hydroelectric power stations being one of the most used resources. As electrical energy cannot be stored, there is a need to estimate its consumption, looking for a way to meet this energy demand. In this context, this study seeks to apply machine learning techniques, using the Grey Wolf Optimization (GWO) meta-heuristic to optimize regression models, to predict the demand for electricity in Brazil, and it aims to estimate how much energy should be produced. For the predictions, the period between the years 2017 to 2022 was used, totaling around 2,190 samples. The methodology involves pre-processing, crossvalidation, parameters optimization and regression. The results show that Random Forest performed well in the experiments carried out, presenting a coefficient of determination (R²) of 0.8751, Root Mean Squared Error (RMSE) of 0.0554 and Mean Absolute Error (MAE) of 0.0348 in the best model.