Analysis of the effects of a passing cloud on a grid-interactive photovoltaic system with battery storage using neural networks

In this paper, a combined radial-basis-functions (RBF) and backpropagation network is used to predict the effects of passing clouds on a utility-interactive photovoltaic (PV) system with battery storage. Using the irradiance as input signal, the network models the effects of random cloud movement on the electrical variables of the maximum power point tracker (MPPT) and the variables of the utility-linked inverter over a short period of time. During short time intervals, the irradiance is considered as the only varying input parameter affecting the electrical variables of the system. The advantages of artificial neural network (ANN) simulation over standard linear models is that it does not require the knowledge of internal system parameters, involves less computational effort, and offers a compact solution for multiple-variable problems. The model can easily integrated into a typical utility system and resulting system behavior can be determined. The viability of the battery-supported PV power system as a dispatchable unit is also investigated. The simulated results are compared with the experimental results captured during cloudy days. This model can be a useful tool in solar energy engineering design and in PV-integrated utility operation