Daily Electric Forecast for Various Indian Regions Using ANN

In current scenario, electric profiles have shown complex behavior, forcing power engineers to improve power system planning (PSP) strategies. Day ahead electric forecasting is very important for PSP. Conventional methods are limited to linear loads, hence are inappropriate for accurately forecasting current behavior of electric profiles. In this manuscript, Artificial Neural Network (ANN) models are developed to forecast (i) wind generation, (ii) solar generation and (iii) peak demand of various Indian regions. To see the performance of the proposed model, the forecasting performance of Auto Regressive (AR) and Autoregressive Integrated Moving Average (ARIMA) models are also recorded. ANN model outperforms over classical models. Furthermore, Genetic Algorithm (GA) is employed to optimize the ANN model as it rapidly convergence to better forecasting performance.