A Comparative Analysis of Multi-Parameter Regression and Fuzzy Logic Models for Short-Term Load Forecasting in the Libyan Power Grid

Abstract

Short-term load forecasting (STLF) is serious to power system reliability, assistant economic operation, generation arrangement, and maintenance planning through precise demand estimates. This paper compares Multi-Parameter Regression (MPR) and Fuzzy Logic (FL) for guessing load on the Libyan grid. Using temperature, humidity, and old peak loads as input variables, the MPR model applies statistical regression to create a linear relationship, while the FL model uses fuzzy rules to report nonlinearities and uncertainty. Both models are trained on local data, with the FL system simulated via MATLAB/Simulink. The predicting performance of both techniques was evaluated and compared using standard error indices. The results validate the efficiency of each method in short-term load forecast and offer insights into their relative accuracy, flexibility, and suitability for use in the Libyan electric network.