Wilson Castillo-Rojas

Dr. Wilson Castillo-Rojas

University of Atacama, Copiapó, Atacama, Chile
orcid

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
27 July 2021 in Education Sciences
Verified
Journal Article
Open access
PDF file
CBCarmen Burgos
Carmen Burgos
Wilson Castillo-RojasWilson Castillo-Rojas
Wilson Castillo-Rojas
Eloy Lopez MenesesEloy Lopez Meneses
Eloy Lopez Meneses
JDJaviera Martínez Díaz
Javiera Martínez Díaz
The objective of this study is to characterize Latent Classes emerging from the analysis of the level of digital competences, use and consumption of applications and/or services through the Internet. For this purpose, the results of the survey Basic Digital Competences (Competencias Básicas Digitales-COBADI®) applied to university students, with more than 60 categorical variables, were considered. A total of 4762 undergraduate and graduate students from five Spanish universities participated in this survey: Complutense University of Madrid (UCM), Pablo de Olavide University (UPO), Almeria University (UAL), National University of Distance Education (UNED) and Rey Juan Carlos University (URJC). The application of the questionnaire was done through the Internet, from the Institute for Research in Social Sciences and Education of University of Atacama—Chile. The methodology used is mixed, because the questions of the questionnaire provide qualitative information that can be interpreted and elaborated from the results. It is also quantitative because basic statistical techniques are used for the exploratory analysis of the data, and later Latent Class Analysis (LCA), to complement the description of the data set and the variables considered in the study, thus allowing us to group the classes of variables that do not appear explicitly in the set of observed variables, but which nevertheless affect them. The results of the study show that regardless of the gender and age range of the participants, there are four clearly differentiated groups or classes in the use and consumption of ICTs in different ways for their activities, both personal and academic, which allows for identifying different developments of digital competences. This study allows establishing a baseline in order to be able to elaborate later, in the development of the digital competences currently needed, which should be developed by university students.

18 April 2023 in International Journal of Photoenergy
Verified
Journal Article
Open access
PDF file
Wilson Castillo-RojasWilson Castillo-Rojas
Wilson Castillo-Rojas
JBJuan Bekios-Calfa
Juan Bekios-Calfa
CHCésar Hernández
César Hernández
In recent years, photovoltaic energy has become one of the most implemented electricity generation options to help reduce environmental pollution suffered by the planet. Accuracy in this photovoltaic energy forecasting is essential to increase the amount of renewable energy that can be introduced to existing electrical grid systems. The objective of this work is based on developing various computational models capable of making short-term forecasting about the generation of photovoltaic energy that is generated in a solar plant. For the implementation of these models, a hybrid architecture based on recurrent neural networks (RNN) with long short-term memory (LSTM) or gated recurrent units (GRU) structure, combined with shallow artificial neural networks (ANN) with multilayer perceptron (MLP) structure, is established. RNN models have a particular configuration that makes them efficient for processing ordered data in time series. The results of this work have been obtained through controlled experiments with different configurations of its hyperparameters for hybrid RNN-ANN models. From these, the three models with the best performance are selected, and after a comparative analysis between them, the forecasting of photovoltaic energy production for the next few hours can be determined with a determination coefficient of 0.97 and root mean square error (RMSE) of 0.17. It is concluded that the proposed and implemented models are functional and capable of predicting with a high level of accuracy the photovoltaic energy production of the solar plant, based on historical data on photovoltaic energy production.

1 October 2024 in IEEE Access
Verified
Journal Article
Open access
Wilson Castillo-RojasWilson Castillo-Rojas
Wilson Castillo-Rojas
JSJavier Pastén Salinas
Javier Pastén Salinas
The effective management of renewable energy facilities, particularly those utilizing solar PV and wind technologies, faces a number of challenges such as weather variability, data resolution, and handling large data volumes. The number of publications addressing these issues is increasing, and it is anticipated that this growth will continue in the near term. Implementing data analysis tools for power plant management can enhance our understanding of weather events and improve electric power generation forecasting accuracy. Despite common characteristics between solar PV and wind energy, including their reliance on weather conditions, handling data volumes, and processing time series data, few studies offer a comprehensive view of generalizing forecast models. Our objective is to determine the trends in forecasts models in this area, the most commonly used techniques that show better performance and efficiency, the variables involved in each type of energy, among other aspects. A systematic mapping study of scientific literature was performed, selecting 75 primary studies grouped by type of machine learning technique. Some common aspects were found regardless of the type of energy, solar PV or wind, mainly related to single or hybrid forecasting models. Other common factors found are the types of model optimization algorithms used and feature extraction and decomposition techniques. The study highlights the importance of hybrid models that demonstrate, compared to conventional or individual models, significant improvements in the level of accuracy of their forecasts for both solar PV and wind energy, combining various statistical, machine learning or deep learning techniques.

1 July 2023 in Energies
Verified
Journal Article
Open access
PDF file
Wilson Castillo-RojasWilson Castillo-Rojas
Wilson Castillo-Rojas
FQFernando Medina Quispe
Fernando Medina Quispe
CHCésar Hernández
César Hernández
In this article, forecast models based on a hybrid architecture that combines recurrent neural networks and shallow neural networks are presented. Two types of models were developed to make predictions. The first type consisted of six models that used records of exported active energy and meteorological variables as inputs. The second type consisted of eight models that used meteorological variables. Different metrics were applied to assess the performance of these models. The best model of each type was selected. Finally, a comparison of the performance between the selected models of both types was presented. The models were validated using real data provided by a solar plant, achieving acceptable levels of accuracy. The selected model of the first type had a root mean square error (RMSE) of 0.19, a mean square error (MSE) of 0.03, a mean absolute error (MAE) of 0.09, a correlation coefficient of 0.96, and a determination coefficient of 0.93. The other selected model of the second type showed lower accuracy in the metrics: RMSE = 0.24, MSE = 0.06, MAE = 0.10, correlation coefficient = 0.95, and determination coefficient = 0.90. Both models demonstrated good performance and acceptable accuracy in forecasting the weekly photovoltaic energy generation of the solar plant.

18 March 2018 in RISTI - Revista Ibérica de Sistemas e Tecnologias de Informação
Verified
Journal Article
Open access
PDF file
Wilson Castillo-RojasWilson Castillo-Rojas
Wilson Castillo-Rojas
FQFernando Medina Quispe
Fernando Medina Quispe
FMFrancisco Fariña Molina
Francisco Fariña Molina
El artículo presenta una nueva metodología para procesos data warehousing, que integra diversos enfoques, técnicas y metodologías, tales como: especificación de requisitos de información, modelamiento relacional, modelo de desarrollo combinado a partir de las propuestas de Kimball y Hefesto, un proceso aumentado de extracción-transformación y carga que incorpora explícitamente una fase de validación de indicadores, y finalmente visualizaciones integradas e interactivas para el análisis multidimensional de los indicadores obtenidos. La metodología propuesta no sólo se basa en los aspectos teóricos descritos en el artículo, sino que además en la experiencia lograda por parte del equipo investigador, en el desarrollo de diversos proyectos de data warehousing, principalmente orientados a la generación de indicadores de productividad académica y de gestión de una universidad, lo que corresponde al caso de estudio de aplicación de la metodología que se describe. Los resultados de éxito en diferentes proyectos donde ha sido utilizada esta metodología avalan su eficacia

1 December 2020 in Ingeniare. Revista chilena de ingeniería
Verified
Journal Article
Open access
PDF file
FMFernando Medina-Quispe
Fernando Medina-Quispe
Wilson Castillo-RojasWilson Castillo-Rojas
Wilson Castillo-Rojas
CMClaudio Meneses
Claudio Meneses
La exploración de un modelo de Minería de Datos (MD), mediante el uso de técnicas de representación visual apropiadas y mecanismos de interacción integrados, presentan ventajas para el analista o minero de datos al momento de intentar comprender un modelo datos. Actualmente, existen nuevas propuestas de metodologías y esquemas de visualización para apoyar procesos de MD, que integran características que combinan técnicas de MD y artefactos gráficos ad-hoc con el objetivo de facilitar el análisis y exploración de modelos, mediante el uso de visualización en la entrada (análisis exploratorio de datos) del proceso de MD, luego en el proceso de generación del modelo (visualización y exploración del modelo y de sus componentes internas), y finalmente en la salida de este proceso (visualización de patrones). Sin embargo, esto apunta a un análisis cualitativo y muchas veces subjetivo, que depende directamente de la experiencia y experticia del analista o minero de datos. Para poder complementar este análisis cualitativo, es necesario incorporar en el esquema visual, funciones con métricas que permitan corroborar cuantitativamente. Este trabajo se orienta en esta dirección, y describe la definición, adaptación, e implementación de un conjunto de métricas que permiten validar y complementar el análisis visual de un modelo de MD, mediante el uso de métricas de distancia y similitud, aplicadas sobre las componentes del modelo de MD. Este trabajo utiliza como caso de estudio, un modelo de MD generado a través de la técnica Árbol de Decisión (ÁD), combinada con la técnica mapas de Kohonen o Self-Organizing Map (SOM) aplicada sobre las componentes o nodos del ÁD. Se logra comprobar la validez de las métricas propuestas a partir de su aplicación, sobre un conjunto de datos conocido a partir de una tarea de MD previamente definida.