A novel Optimized initial condition and Seasonal division based Grey Seasonal Variation Index model for hydropower generation

For making earlier realization on peak carbon dioxide emissions and carbon neutrality, hydropower develop-ment in countries all over the world can effectively reduce the Greenhouse Gas (GHG) emissions and solve the problem of global climate change. This paper proposes a novel Optimized initial condition and Seasonal division based Grey Seasonal Variation Index (OSGSVI) model to accurately predict the hydropower generation in some countries. Firstly, for enhancing the fitting accuracy, the initial condition is optimized based on the weighted average methods and the data grouping with OSVI is utilized by seasonal divisions. Secondly, an OSGSVI model is established coupled optimization on both optimized initial conditions and seasonal divisions. Thirdly, the Whale Optimized Algorithm (WOA) is employed to determine estimated parameters to further enhance the fitting accuracy for the hydropower generation. Finally, the experimental results of the prediction study show that three error measure values are all the smallest in all the fitting results and the MAPE values are converged before 30 iterations by utilizing our proposed model when compared with a set of baseline prediction models. It demonstrates the superiority of our proposed model over the others on the fitting accuracy with fast-convergence for the hydropower generation in these selected countries.