Short-Term Wind Power Forecasting Using Structured Neural Network

The effect of global warming are being experienced more and more by people now a days. Leaders are trying to change our dependency on energy into renewable energies. Wind energy is one of the choices since it has the advantage of being clean and efficient but the downside is that the power being dependent on the wind speed is unfavourable to the power system. Having an information ahead of time could somehow stabilize the output of the wind energy source before injecting to the grid. It is therefore the aim of this study to investigate the applicability of structured neural network in a power forecasting. This study was done by first assembling a small-scale 200 watt wind turbine and then recording the power it has produced over a period of time. Using these recorded data, the structured neural network was modelled, trained, and validated using the Neural Network Toolbox in MATLAB. The results showed that the network was able to forecast the power with a mean squared error of 9.7495.