Short-Term Load Forecast for Energy Management Systems Using Time Series Analysis and Neural Network Method with Average True Range

Abstract

The load forecasting is one of the important tools for Energy Management System (EMS). It is used for planning and management power balance. Short term load forecasting (STLF) has a significant impact on the efficiency of operation. The load forecasting model must be able to accurately predict the demand of electrical power. This paper proposes the load forecasting models based on time series analysis and neural network methods. The data is taken from Mae Hong Son (MHS) located in the northern Thailand. Time series analysis utilizes auto regressive integrated moving average (ARIMA) and seasonal auto regressive integrated moving average (SARIMA). In addition, neural networks cover artificial neural network (ANN) and long-short term memory (LSTM) based recurrent neural network (RNN). Additional, the Average True Range (ATR) index is adapted to improve the performance of RNN model. We compare the performance of these models using statistic criteria, namely, root mean square error and mean absolute percent error and choose the best model to implement for micro EMS.