Design of Optimal Power Point Tracking Controller Using Forecasted Photovoltaic Power and Demand

With the advent of grid-connected photovoltaic systems for energy generation, new technologies must be created that maintain a continuous and stable balance between supply and demand of generated electricity. Consequently, accurate prediction of solar energy generation and consumption is required. Solar energy generation and electric power demand are both stochastic and nonstationary in nature and often incongruous. The imbalance between demand and supply can be costly and leads to long-term ineffectiveness of power generation and distribution. The aim of this work is to propose methods for maintaining demand-supply balance in PV power generation and distribution systems. To achieve this, we build and combine three different tools: 1) a predictive model for forecasting solar energy generation, 2) a predictive model for demand prediction, and 3) a real-time control algorithm that uses the outputs of prediction models and adjusts the output voltage of PV system to maintain demand-supply balance. Our prediction models are based on time-series forecasting tools and artificial neural networks. The control algorithm is called optimal power point tracking (OPPT) and is based on the perturb and observe algorithm. We evaluate the performance of the combined prediction-controller system using real-world data.