Wind power has been occupying the substantial proportion of energy structure since it occurs to people that dependence on fossil fuels with the aims of worldwide energy supply would provoke an avalanche of challenges concerning on environmental degradation like the increase of greenhouse gas emissions. Nevertheless, most of wind power resource cannot be utilized completely as wind power generation have high volatility, which also possess partly repercussion with aims of contributing to the operation of the power system. As the technology has burgeoned, numerous scientists have made great strides in the wind power prediction, which can facilitate the consistent functioning of the electrical grid. However, there are massive difference in the wind power predication effect due to the respective character of algorithm. This essay will introduce the category of multiple relative models and compares the consequence of the improved algorithm based in Grey Wolf Optimizer (GWO), Grey Model (GM) and Social Spider Algorithm (SSA) model by analyzing the relevant table Comparison results that GM model is capable of significantly enhancing the predictive performance to achieve heightened authenticity and decreased presage error in contrast to the previous algorithm.