Electrical Load Forecasting on A 500/150 kV Substation Based on Historical Daily Records Using Multivariate LSTM Network

A smart grid concept with a prediction system provides accurate information as an early warning in the process of generation and distribution of electrical energy. The complexity of the power distribution system that involves complex parameter settings is a major challenge that is difficult to predict. Although the data in the smart grid system can be acquired centrally, however, the prediction system faces a lack of accuracy due to incomplete records and low data quantity. In this paper, a machine learning (ML) model is developed based on multivariate long short-term memory (MV-LSTM) to extract long short-term patterns inside an electrical load dataset. This dataset is recorded from daily measurements in six months at Cawang Baru Substation, Indonesia. The proposed model adopts the basic concept of multi-layer perceptron to record temporal patterns in several stages, thereby producing more accurate results. The proposed architecture supports multivariate feature extraction so as to capture important correlations between multi-dimensional features. This study also uses the basic Univariate LSTM (UV-LSTM) model and naive ML models including Linear Regression (LR), Random Forest (RF), and Support Vector Regression (SVR) as benchmarking methods for the proposed model. In the validation stage, MV-LSTM achieves higher accuracy than UV-LSTM, SVR, LR, and RF with scores of 0.3688, 0.3645, 0.1332, 0.1438, and 0.1234, respectively, evaluated using R-squared. Finally, experimental results support the view that using multivariate data and sequential ML models is superior for time series prediction tasks rather than those using univariate data.