Differential Evolution (DE) algorithm is a type of global optimization algorithm based on population searching for the optimal solution. It has characteristics such as fast convergence speed, simple and understandable algorithm, few parameters, and high stability. To further enhance its exhibition, we propose a differential evolution algorithm based on subpopulation adaptive scale and multi-adjustment strategy (ASMSDE). The algorithm divides population into three sub-populations based on the fitness values, the different operation strategies are adopted according to their characteristics. The superior population and inferior population adopt respectively Gaussian disturbance and Levy flights. The task of the intermediate population is to maintain the overall diversity of the population. The sizes of the three sub-populations are adaptively adjusted based on the evolutionary results to accommodate changes in individual differences during the evolution process. With the number of iterations increases and the differences between individuals decrease, adopt single population model instead of multi-population model in the later stage of evolution. To evaluate the performance of the ASMSDE algorithm, it is compared with other advanced algorithms using benchmark function optimizations. Experimental results demonstrate that the ASMSDE algorithm outperforms the comparison algorithms in most cases, validating its effectiveness and ability to handle local optima situations.