Different Forecasting Horizons Based Performance Analysis of Electricity Load Forecasting Using Multilayer Perceptron Neural Network

With an uninterrupted power supply to the consumer, it is obligatory balance electricity generated by load. The effective planning of economic dispatch, reserve requirements, and quality provision for accurate consumer information concerning load needed. burden on system engineers eased forecasting essential ensure enhanced operation reliable provision. Fickle nature, atmospheric parameters influence makes a very complex challenging task. This paper proposed multilayer perceptron neural network (MLPNN) with association recursive fine-tuning strategy-based different horizons model forecasting. We consider as inputs model, overcoming effect Hidden layers hidden neurons based performance investigation performed. Analyzed other existing models; comparative reveals that performs rigorous minimal evaluation index (mean square error (MSE) 1.1506 × 10−05 Dataset 1 MSE 4.0142 10−07 2 concern single layer 2.9962 1, 1.0425 10−08 two model) compared considered models. possesses good ability because we develop various input variables, which overcomes variance. It has generic capability robust more reliable.