Optimal Size Analysis of ANN Forecast Model of Off-Grid AC-DC Micro Grid System

To ensure stable operation with limited energy storage, power grids integrating wind power require improved short-term wind speed forecasting for optimal management of transmission system parameters. The time series analysis method is often used for the study of wind speed variability. A time series can be defined through a series of monitoring events considering the sequential hierarchy. The study of cost and energy efficiency plays an predominant character in numerous applications including cost effectiveness for energy consumers and impact of reduction in climate change, controlling air pollution for better health condition, development of indoor environment, energy security improvement respectively. To conquer the challenges and issues, the sensitivity analysis algorithms and optimization techniques of HOMER pro's is useful for the easier evaluation of different possible system configurations. This paper consists of two sections. Section 1 narrates the wind speed prediction using various historical data. This study investigates the implementation of an Artificial Neural Network (ANN) model for prediction of short-term wind speed at a certain location. The model's performance is evaluated using standard metrics like Mean Squared Error (MSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). Furthermore, the research explores the energy efficiency of two distinct renewable energy systems: one employing an AC natural gas generator and the other utilizing an AC diesel generator. Section 2 delves into a comparison of these systems' efficiency. At this point, the PV, batteries, wind turbine, and converter system model has been considered. This study focuses on the calculation of the most energy-efficient model for Jafrabad, Gujrat, using HOMER pro.