Short-Term Electric Load Forecast Model Using the Combination of Ant Lion Optimization with Bi-LSTM Network

The optimization of smart grids heavily relies on accurate electric load prediction, as the demand from consumers can significantly impact a country’s economic condition. Effective management of electric load requires analyzing the patterns of electricity demand and supply. The amount of recorded electric load data can vary in scale, depending on the duration and frequency of data collection. In this study, the authors propose a prediction framework for electric load forecasting using a recurrent neural network (RNN) and a meta-heuristic algorithm. The dataset used for experimentation consists of electric load time series data collected at hourly intervals. The authors have developed a Bidirectional long-short term memory (Bi-LSTM) neural network model using the Ant Lion optimization algorithm, which can forecast electric load data with a root-mean-square error (MSE) of 0. 0032during training and testing.