Load forecasting with support vector regression: influence of data normalization on grid search algorithm

<span>In recent years, support vector regression (SVR) models have been widely applied in short-term electricity load forecasting. A critical challenge when applying the SVR model is to determine for optimal hyperparameters, which can be solved using several optimization methods as grid search algorithm. Another that affects response time and precision of normalization process input data. In this paper, algorithm will suggested based on data including Z-score, min-max, max, decimal, sigmoidal, softmax; then utilized evaluate both precision. To verify proposed methods, actual demand two cities, Queensland Australia Ho Chi Minh City Vietnam, were study.</span>