Abstract: Aimed at the shortcomings of the standard hippopotamus optimization algorithm, such as the lack of global exploration ability and falling into the local optimization, in the report, an improved hippopotamus optimization algorithm integrating multi-strategy was proposed. During the initialization process, a chaotic mapping was introduced to improve convergence speed, the adaptive weights were introduced to avoid being involved into local optimization, and the reverse learning was used to obtain the reverse solutions to expand the search range of the algorithm. Six benchmark test functions were used to test the performance of the improved hippopotamus optimization algorithm, and which are compared with the multiple optimization algorithms. The results showed that the optimization performance of the improved hippopotamus optimization algorithm is significantly better than that of the other optimization algorithms. The improved hippopotamus optimization algorithm was used for two engineering problems, and the satisfactory results were obtained.