A Study on Improved Algorithms for Mid-to Long-Term Power Forecasting of New Energy Photovoltaics

Mid-to long-term photovoltaic power forecasting is of great significance for reducing grid dispatching difficulty, increasing the stability of new energy grid integration, and achieving optimal operation of power systems. In this regard, a Multiple Mixed Strategy Improved Whale Optimization Algorithm (MMSWOA) is proposed. This algorithm uses a novel Chebyshev chaotic map to initialize the population settings, increasing the diversity and randomness of individuals in the population to better simulate actual conditions. Additionally, the convergence factor of the algorithm is improved by introducing adaptive weights, optimal neighborhood perturbation, and a spiral position curve, which helps the algorithm to more accurately adjust the search space, improve convergence precision, and prevent it from falling into local optima. The algorithm is further optimized with a CNN-BiLSTM model. Matlab simulations and experimental results show that the proposed method can be effectively applied to actual photovoltaic power forecasting.