Application of Seasonal Trend Decomposition using Loess and Long Short-Term Memory in Peak Load Forecasting Model in Tien Giang

Accurate peak load forecasting is essential for the stable operation and planning of modern electrical power grids. This project presents a methodology-oriented implementation support framework for a hybrid forecasting model combining Seasonal Trend decomposition using Loess (STL) with Long Short-Term Memory (LSTM) networks. The proposed approach decomposes complex, non-linear daily electrical peak load time-series data into trend, seasonal, and residual components to simplify the underlying patterns for the neural network. Using historical demand data, meteorological variables, and holiday indicators, the LSTM network is trained on these decomposed components to predict future peak loads. The performance of the STL-LSTM model is systematically evaluated and compared against alternative architectures, including Convolutional Neural

Network-LSTM (CNN-LSTM), Wavenet, standard Artificial Neural Networks (ANN), and standalone LSTM models. Evaluation metrics such as Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) are utilized to quantify forecasting accuracy. This research direction provides structured guidance for implementing robust forecasting models suitable for regional power systems, facilitating informed decision-making, infrastructure planning, and resource allocation for utility providers.

Message: Hi FE, I need help with “Application of Seasonal Trend Decomposition using Loess and Long Short-Term Memory in Peak Load Forecasting Model in Tien Giang” in “Electrical Engineering”