Prediction of Electricity Consumption Profiles Using Potential Polynomials of Degree One and Artificial Neural Networks in Smart Metering Infrastructure
This work analyzes methods and algorithms for predicting the behavior of electricity consumption based on neural networks using data obtained from the Advanced Measurement Infrastructure (AMI) of an educational institution. Also, a contrast between the use of conventional neural networks (ANN), wave...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Online |
| Language: | eng |
| Published: |
Universidad Pedagógica y Tecnológica de Colombia
2021
|
| Subjects: | |
| Online Access: | https://revistas.uptc.edu.co/index.php/ingenieria/article/view/12772 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1801706092189712384 |
|---|---|
| author | Urgilés, Pablo Inga-Ortega, Juan Peralta, Arturo Ortega, Andrés |
| author_facet | Urgilés, Pablo Inga-Ortega, Juan Peralta, Arturo Ortega, Andrés |
| author_sort | Urgilés, Pablo |
| collection | OJS |
| description | This work analyzes methods and algorithms for predicting the behavior of electricity consumption based on neural networks using data obtained from the Advanced Measurement Infrastructure (AMI) of an educational institution. Also, a contrast between the use of conventional neural networks (ANN), wavelet-based neural networks (WNN) and potential polynomials of degree one (P1P) has been performed. The correlation of each prediction method is analyzed, as well as the behavior of the Mean Square Error (MSE), to finally establish if there is an imbalance in the computational cost through the Big-O analysis and the executing time. The quantitative results of the MSE are below 0.05% for ANN predictions and they use a high computational cost. For P1P, errors around 1.2% are presented, showing as a low computational consumption prediction method but mainly applicable for a short-term analysis. This work is given in response to the need to establish a platform to take advantage of the smart metering structure through the prediction of electricity consumption profile, with the objective of developing a plan for maintenance and management of electricity demand to reduce operating costs from the final consumer to the distribution network operator. For the analysis of projections on the electrical load profile, the statistical characteristics of the consumption are considered to select the prediction algorithms according to the number of days to be projected using data from any of the smart meters, which can be monitored in an electrical network oriented to Smart Grids. |
| format | Online |
| id | oai:oai.revistas.uptc.edu.co:article-12772 |
| institution | Revista Facultad de Ingeniería |
| language | eng |
| publishDate | 2021 |
| publisher | Universidad Pedagógica y Tecnológica de Colombia |
| record_format | ojs |
| spelling | oai:oai.revistas.uptc.edu.co:article-127722022-06-15T15:52:54Z Prediction of Electricity Consumption Profiles Using Potential Polynomials of Degree One and Artificial Neural Networks in Smart Metering Infrastructure Predicción de perfiles de consumo eléctrico usando polinomios potenciales de grado uno y redes neuronales artificiales en la infraestructura de medición inteligente Urgilés, Pablo Inga-Ortega, Juan Peralta, Arturo Ortega, Andrés AMI medición inteligente P1P predicción de consumo eléctrico WNN AMI electricity consumption prediction P1P smart metering WNN This work analyzes methods and algorithms for predicting the behavior of electricity consumption based on neural networks using data obtained from the Advanced Measurement Infrastructure (AMI) of an educational institution. Also, a contrast between the use of conventional neural networks (ANN), wavelet-based neural networks (WNN) and potential polynomials of degree one (P1P) has been performed. The correlation of each prediction method is analyzed, as well as the behavior of the Mean Square Error (MSE), to finally establish if there is an imbalance in the computational cost through the Big-O analysis and the executing time. The quantitative results of the MSE are below 0.05% for ANN predictions and they use a high computational cost. For P1P, errors around 1.2% are presented, showing as a low computational consumption prediction method but mainly applicable for a short-term analysis. This work is given in response to the need to establish a platform to take advantage of the smart metering structure through the prediction of electricity consumption profile, with the objective of developing a plan for maintenance and management of electricity demand to reduce operating costs from the final consumer to the distribution network operator. For the analysis of projections on the electrical load profile, the statistical characteristics of the consumption are considered to select the prediction algorithms according to the number of days to be projected using data from any of the smart meters, which can be monitored in an electrical network oriented to Smart Grids. Este trabajo analiza métodos y algoritmos de predicción del comportamiento de consumo eléctrico basados en redes neuronales, usando datos obtenidos de la infraestructura de medición avanzada (AMI) de una institución educativa. También, se ha realizado un contraste entre el uso de redes neuronales convencionales (ANN), redes neuronales basadas en wavelets (WNN) y los polinomios potenciales de grado uno (P1P). Se analiza la correlación de cada método de predicción, así como el comportamiento del error cuadrático medio (MSE) para finalmente establecer si existe un desbalance en el coste computacional a través del análisis de Big-O y el tiempo de ejecución. Los resultados cuantitativos del error MSE están por debajo del 0,05% para predicciones con ANN y usan un alto costo computacional. Para P1P se presentan errores alrededor del 1,2% mostrando como método de predicción de bajo consumo computacional pero aplicable de forma principal para un análisis a corto plazo. Este trabajo se da en respuesta a la necesidad de establecer una plataforma que permita aprovechar la estructura de medición inteligente, a través de la predicción de perfil de consumo eléctrico con el objetivo de elaborar una planificación de mantenimiento y gestión de la demanda eléctrica para reducir costos de operación desde el consumidor final hasta el gestor de la distribución de energía eléctrica. Para el análisis de las proyecciones sobre el perfil de carga eléctrica se consideran las características estadísticas del consumo para seleccionar los algoritmos de predicción según la cantidad de días a proyectar, usando los datos de cualquiera de los medidores inteligentes, que pueden ser monitoreados en una red eléctrica orientada a las Smart Grids. Universidad Pedagógica y Tecnológica de Colombia 2021-06-02 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion application/pdf text/xml https://revistas.uptc.edu.co/index.php/ingenieria/article/view/12772 10.19053/01211129.v30.n56.2021.12772 Revista Facultad de Ingeniería; Vol. 30 No. 56 (2021): April-June 2021 (Continuous Publication); e12772 Revista Facultad de Ingeniería; Vol. 30 Núm. 56 (2021): Abril-Junio 2021 (Publicación Continua); e12772 2357-5328 0121-1129 eng https://revistas.uptc.edu.co/index.php/ingenieria/article/view/12772/10919 https://revistas.uptc.edu.co/index.php/ingenieria/article/view/12772/10945 Copyright (c) 2021 Pablo Urgilés, Juan Inga-Ortega, Arturo Peralta, Andrés Ortega http://creativecommons.org/licenses/by/4.0 |
| spellingShingle | AMI medición inteligente P1P predicción de consumo eléctrico WNN AMI electricity consumption prediction P1P smart metering WNN Urgilés, Pablo Inga-Ortega, Juan Peralta, Arturo Ortega, Andrés Prediction of Electricity Consumption Profiles Using Potential Polynomials of Degree One and Artificial Neural Networks in Smart Metering Infrastructure |
| title | Prediction of Electricity Consumption Profiles Using Potential Polynomials of Degree One and Artificial Neural Networks in Smart Metering Infrastructure |
| title_alt | Predicción de perfiles de consumo eléctrico usando polinomios potenciales de grado uno y redes neuronales artificiales en la infraestructura de medición inteligente |
| title_full | Prediction of Electricity Consumption Profiles Using Potential Polynomials of Degree One and Artificial Neural Networks in Smart Metering Infrastructure |
| title_fullStr | Prediction of Electricity Consumption Profiles Using Potential Polynomials of Degree One and Artificial Neural Networks in Smart Metering Infrastructure |
| title_full_unstemmed | Prediction of Electricity Consumption Profiles Using Potential Polynomials of Degree One and Artificial Neural Networks in Smart Metering Infrastructure |
| title_short | Prediction of Electricity Consumption Profiles Using Potential Polynomials of Degree One and Artificial Neural Networks in Smart Metering Infrastructure |
| title_sort | prediction of electricity consumption profiles using potential polynomials of degree one and artificial neural networks in smart metering infrastructure |
| topic | AMI medición inteligente P1P predicción de consumo eléctrico WNN AMI electricity consumption prediction P1P smart metering WNN |
| topic_facet | AMI medición inteligente P1P predicción de consumo eléctrico WNN AMI electricity consumption prediction P1P smart metering WNN |
| url | https://revistas.uptc.edu.co/index.php/ingenieria/article/view/12772 |
| work_keys_str_mv | AT urgilespablo predictionofelectricityconsumptionprofilesusingpotentialpolynomialsofdegreeoneandartificialneuralnetworksinsmartmeteringinfrastructure AT ingaortegajuan predictionofelectricityconsumptionprofilesusingpotentialpolynomialsofdegreeoneandartificialneuralnetworksinsmartmeteringinfrastructure AT peraltaarturo predictionofelectricityconsumptionprofilesusingpotentialpolynomialsofdegreeoneandartificialneuralnetworksinsmartmeteringinfrastructure AT ortegaandres predictionofelectricityconsumptionprofilesusingpotentialpolynomialsofdegreeoneandartificialneuralnetworksinsmartmeteringinfrastructure AT urgilespablo predicciondeperfilesdeconsumoelectricousandopolinomiospotencialesdegradounoyredesneuronalesartificialesenlainfraestructurademedicioninteligente AT ingaortegajuan predicciondeperfilesdeconsumoelectricousandopolinomiospotencialesdegradounoyredesneuronalesartificialesenlainfraestructurademedicioninteligente AT peraltaarturo predicciondeperfilesdeconsumoelectricousandopolinomiospotencialesdegradounoyredesneuronalesartificialesenlainfraestructurademedicioninteligente AT ortegaandres predicciondeperfilesdeconsumoelectricousandopolinomiospotencialesdegradounoyredesneuronalesartificialesenlainfraestructurademedicioninteligente |
