Short-Term Forecasting of Electric Loads Using Nonlinear Autoregressive Artificial Neural Networks with Exogenous Vector Inputs

Abstract

Short-term load forecasting is crucial for the operations planning of an electrical grid. Forecasting the next 24 h of electrical load in a grid allows operators to plan and optimize their resources. The purpose of this study is to develop a more accurate short-term load forecasting method utilizing non-linear autoregressive artificial neural networks (ANN) with exogenous multi-variable input (NARX). The proposed implementation of the network is new: the neural network is trained in open-loop using actual load and weather data, and then, the network is placed in closed-loop to generate a forecast using the predicted load as the feedback input. Unlike the existing short-term load forecasting methods using ANNs, the proposed method uses its own output as the input in order to improve the accuracy, thus effectively implementing a feedback loop for the load, making it less dependent on external data. Using the proposed framework, mean absolute percent errors in the forecast in the order of 1% have been achieved, which is a 30% improvement on the average error using feedforward ANNs, ARMAX and state space methods, which can result in large savings by avoiding commissioning of unnecessary power plants. The New England electrical load data are used to train and validate the forecast prediction.

A Nonlinear Autoregressive Exogenous (NARX) Neural Network Model for the Prediction of the Daily Direct Solar Radiation

“…The second advantage consists in an architecture of the resulting network which is purely feedforward, the usual training algorithms for Multi-Layer Perceptron (MLP) networks can be used. After the training phase, the NARX neural network is converted to the parallel architecture which is beneficial for multi-step-ahead prediction [31,32].…”

Neural network-based power system dynamic state estimation using hybrid data from SCADA and phasor measurement units

“…One of the most important applications of NARX dynamic neural networks is in intelligent control issues where various artificial intelligence or soft computing approaches like neural networks, Bayesian probability, fuzzy logic, machine learning, evolutionary computation, and genetic algorithms are applied. Some significant features of NARX networks are as follows: (1) learning is more effective in NARX networks in comparison with other neural network (especially the gradient descent is better in NARX), and (2) these networks converge much faster and generalize better than other networks …”

A review of forecasting algorithms and energy management strategies for microgrids

“…Additionally, the Principal Component Analysis (PCA) was implemented for outlier detection. Buitrago and Asfour (2017) integrated the neural network with a classic control theory. On the first stage, the training process was implemented in an open-loop to determine the node weighting.…”