Load forecasting for power system planning and operation using artificial neural network (a case study on larona hydro power in the nickel smelting plant)

The research describes in this paper aims to help improve the performance of a Hydro Power plant operation and planning by analyzing the performance of its power loads. The amount of electrical energy used at certain times cannot be calculated exactly. This can lead to a lack of electricity supply on the consumer side if the generated power is less than the need electrical energy required. Increased electricity needs can also be cause problems to the quality of electric power that is delivered respectively. To overcome things, it is necessary to have a proper electric power system operation plan reliable through forecasting the electrical load in the future. In this study, carried out load forecasting for power system planning and operation of Larona Hydro Power Plant, in smelter plant Sorowako propose an Artificial Neural Network (ANN) method. The ANN was implemented using tools of MATLAB. The structure of the ANN is MLP (Multi-layer Perceptron). The load forecasting conducted in simulation, proceed the data by constructing and train the neural network with this data. After the validation of neural network error rate, the network function used to estimate a short-term prediction with determination predictor parameters namely number of learning input, activation function, and learning rate. Error was calculated as MAPE (Mean Absolute Percentage Error), the model selection criteria are based on the best RMSE values with the smallest MAPE value, and with error of about 0.957% this paper was successfully carried out.