Minutely Active Power Forecasting Models Using Neural Networks

Abstract

Power forecasting is an integral part of the Demand Response design philosophy for power systems, enabling utility companies to understand the electricity consumption patterns of their customers and adjust price signals accordingly, in order to handle load demand more effectively. Since there is an increasing interest in real-time automation and more flexible Demand Response programs that monitor changes in the residential load profiles and reflect them according to changes in energy pricing schemes, high granularity time series forecasting is at the forefront of energy and artificial intelligence research, aimed at developing machine learning models that can produce accurate time series predictions. In this study we compared the baseline performance and structure of different types of neural networks on residential energy data by formulating a suitable supervised learning problem, based on real world data. After training and testing long short-term memory (LSTM) network variants, a convolutional neural network (CNN), and a multi-layer perceptron (MLP), we observed that the latter performed better on the given problem, yielding the lowest mean absolute error and achieving the fastest training time.

Reducing the Cost of Electricity by Optimizing Real-Time Consumer Planning Using a New Genetic Algorithm-Based Strategy

“…These algorithms specialize in solving these types of problems. They have been successfully used in multi-criteria optimization problems in various fields, ranging from finding the optimal configuration (minimum cost, minimum weight) for the realization of composite beams and frames taking into account a set of input criteria (including were related to the climate) [23], to making financial statistics and predictions starting from several economic indicators, data sets, or moods, as presented in reference [24]. In addition to the papers presented above, other studies have addressed the issue of optimizing energy cost (or energy consumption) through genetic algorithms.…”

Power prediction of waste heat recovery system for a cement plant using back propagation neural network and its thermodynamic modeling

“…26 Similarly, results from different neural networks were compared by Kontoggiannis et al for predicting the power consumption from a residential dataset. 27 Multi-layer perceptron was found to be the better option, as it converged the data fast and gave fairly accurate results of prediction.…”

“…For each algorithm, a hyper‐tuning parameter was run for optimizing the results, and a gradient boosting machine regressor showed the most promising result in terms of the least mean absolute percentage error (MAPE) among all techniques followed 26 . Similarly, results from different neural networks were compared by Kontoggiannis et al for predicting the power consumption from a residential dataset 27 . Multi‐layer perceptron was found to be the better option, as it converged the data fast and gave fairly accurate results of prediction.…”

A Meta-Modeling Power Consumption Forecasting Approach Combining Client Similarity and Causality

“…On the other side of the spectrum, advances in artificial intelligence and machine learning led to the development of more robust models, which are capable of discovering complex relationships between input and output features. Many different architectures involving neural networks, such as the multilayer perceptron (MLP) [15] and long shortterm memory (LSTM) network [16], were used successfully in many time series forecasting tasks, achieving impressive performance [17]. These neural network models follow a blackbox approach in the approximation of nonlinear functions.…”