Impact Analysis of Short-term Electrical Demand Factors using Multivariate Forecasting Approach: A case of Kathmandu valley
Authors: Amrit Parajuli, Sumit Shah, Pramish Shrestha, Kamal Chapagain
Abstract—Load forecasting is essential for day ahead market operation as well as system planning activities. The process has been a preliminary work for every power system operating entities and policy makers. Load forecasting is a complex and difficult task that involves the analysis of several factors. The research focuses to give more insights on the development of mathematical models and the factors that influence short-term electricity demand by entailing assessment of different multivariate models on the demand data of Kathmandu valley. The demand is predicted using two time-series models: Multiple Linear Regression (MLR) and Recurrent Neural Network Long Short-Term Memory (RNN-LSTM). The demand of Kathmandu valley is most affected by the minimum daily temperature during the months November-February with a negative correlation. The forecast accuracy is measured in terms of Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). For all three performance indicators, LSTM is found more superior with MAPE of 4.37%, RMSE of 12.91 MW and MAE of 8.58 MW.
Keywords—Short Term Load Forecasting (STLF), Time series, Multiple Linear Regression (MLR), Long Short-Term Memory (LSTM), Mean Absolute Percentage Error (MAPE)
Full Paper: to be purchased for $20 (contact us at [email protected])
Published In: International Conference on Role of Energy for Sustainable Social Development (RESSD-2022)
Date of Conference: 8th-9th August 2022
Conference Location: Bhaktapur, Nepal
Publisher: IEEE Power and Energy Society Nepal Chapter
Cite the paper as:
A. Parajuli, S. Shah, P. Shrestha, K. Chapagain, “Impact Analysis of Short-term Electrical Demand Factors using Multivariate Forecasting Approach: A case of Kathmandu valley ”, International Conference on Role of Energy for Sustainable Social Development, 8th-9th August 2022, Bhaktapur, Nepal