Short-term electrical loads forecasting for new Baron Technopark hybrid renewable energy power plant Yogyakarta using artificial neural network

Human activities today are very dependent on the use of electricity ranging from simple activities to large activities. It has an impact on increasing electricity demand. At PLT EBT Baron Technopark Yogyakarta, electricity needs are generated by themselves. However, PLT EBT itself has a fluctuating level of electricity production depending on the natural phenomena around it. Therefore, it is necessary to plan for the supply of electricity to balance the electrical power generated, prevent blackouts, and better energy sources management to deal with fluctuating electrical energy production. One way to balance electric power with power demand in conditions of fluctuating electricity production is to perform accurate load forecasting in the future. This study uses the Artificial Neural Network (ANN) method using data on the electricity load of PLT EBT Baron Technopark Yogyakarta from October to November 2021. In the training process, the best network model was obtained with input_width = 48, Train_size = 0.8, and build model = 64, 32, 32 where the MSE value is 0.001 and MAPE is 8.03%.