Taking the evaluation of water resources use efficiency in Yunnan Province from 2006 to 2015 and 2020 as an example,15 indicators were selected from aspects of comprehensiveness, industry, agriculture, life and ecology to construct the evaluation index system and grading standard of water resources utilization efficiency, and the dynamic evaluation of regional water resources utilization efficiency was conducted based on maximum entropy projection pursuit (MEPP). The MEPP objective function was constructed by randomly generating the samples between the thresholds of the target classification criteria, and the optimal projection direction of the MEPP was optimized by using the water cycle algorithm (WCA). The WCA-MEPP water resource utilization efficiency evaluation model was proposed, and the biogeographic optimization (BBO) Algorithm, differential evolution (DE) algorithm and particle swarm optimization (PSO) algorithm - MEPP water use efficiency evaluation model were established. The results show that: (1)the optimal value, the worst value , the average value and the standard deviation of WCA-based optimal MEPP are better than those of BBO, DE and PSO, and have the best global extreme value optimization ability y. (2)WCA-MEPP model evaluates the water use efficiency of Yunnan Province as "low level" from 2006 to 2007 and "medium level" from 2008 to 2015, and "high level" in 2020. From 2006 to 2015, the water use efficiency of Yunnan Province increased with the time, and the trend of upgrading was remarkable. (3)the WCA-MEPP model has the same results as the BBO-MEPP model, but the rankings are different in the WCA-MEPP model. There are differences in the evaluation results and ranking between the DE-MEPP model and the PSO-MEPP model.