Aiming at the shortcomings of the original grey Wolf algorithm in solving global optimization problems, such as slow convergence, easy local optimum and low optimization efficiency, a multi-strategy improved grey wolf optimization algorithm (Multi-Strategy Improved Grey Wolf Optimization, MSI_GWO) is proposed. Three improvement strategies are introduced from parameter, search mechanism and optimal solution disturbance: nonlinear adjustment strategy for control parameter to improve the exploration and development capabilities of the algorithm; dynamic weight strategy to update position to improve the convergence of the algorithm; wavelet optimal solution disturbance strategy to improve population diversity and avoid the algorithm from local optimum. The optimal performance of MSI_GWO algorithm is verified, and nine test functions are selected to complete the simulation experiment, comparing it with other improved grey wolf optimization algorithms Particle Swarm Optimization, Aquila Optimizer and Honey Badger Algorithm. The results show that MSI_GWO algorithm is optimal in terms of convergence speed and optimal efficiency.