Citas bibligráficas
Rojas, J., (2021). Metodología de segmentación de datos para el cálculo de la magnitud reserva secundaria de frecuencia del Sistema Eléctrico Interconectado Nacional [Tesis, Universidad Nacional de Ingeniería]. http://hdl.handle.net/20.500.14076/22509
Rojas, J., Metodología de segmentación de datos para el cálculo de la magnitud reserva secundaria de frecuencia del Sistema Eléctrico Interconectado Nacional [Tesis]. PE: Universidad Nacional de Ingeniería; 2021. http://hdl.handle.net/20.500.14076/22509
@misc{renati/712030,
title = "Metodología de segmentación de datos para el cálculo de la magnitud reserva secundaria de frecuencia del Sistema Eléctrico Interconectado Nacional",
author = "Rojas Hernandez, Joshua Ari",
publisher = "Universidad Nacional de Ingeniería",
year = "2021"
}
Technological advances in the development of Data Science are increasingly required to solve problems involving the processing of large information databases. In particular, the use of machine learning techniques can streamline and make more flexible the processing of data with a specific recurrence or execution with similar characteristics, such as population growth, consumption of goods and services and / or tastes and preferences of users. The execution of the operation of the “Sistema Eléctrico Interconectado Nacional - SEIN“ generates historical records that are quantifiable and qualifiable, and precisely those that define the bases of its physical operation of the same are the demand and generation of electricity, particularly the atter of the type of Renewable Energy Resources (RER), which are executed with a certain predictive particularity. In this sense, because the electricity demand and the RER generation have considerable deviations, the Magnitude of Generation Reserve directed to the Secondary Frequency Regulation of the SEIN will provide the support to address these deviations. Then, using unsupervised machine learning techniques, such as Data Segmentation or “Clustering“ with the application of specialized algorithms for the classification of data groups such as “KMeans“ (K Mean Clusters Classifications) and “DBSCAN“ (Density Based Spatial Clustering of Applications with Noise), this paper establishes a smart and efficient debugging methodology of the set of data that make up the demand and the RER generation executed, and consequently the calculation of the magnitude of secondary frequency reserve is developed based on the provisions of Technical Procedure Nº22 of the Economic Operation Committee of the National lnterconnected Electrical System (COES SINAC). Finally, the conclusions and recommendations derived from the present work are established.
Este ítem está sujeto a una licencia Creative Commons Licencia Creative Commons