To address the multi-objective coordination problem of water supply, power generation, and ecological targets in the Huokou-Shanzai cascade reservoir system of the Aojiang River Basin, this study proposed a multi-objective non-inferior solution set optimization method based on the improved NSGA-Ⅲ algorithm. By constructing a decision-making strategy that integrates the two-layer reference point method and the boundary weighting approach, this method achieves the scientific selection of the non-inferior solution set. Based on this method, a multi-objective operation model for the cascade reservoir system in the Aojiang River Basin was established and compared with single-objective operation models. The results indicate that the multi-objective operation model effectively addresses the challenge of exponential growth in reference points under high-dimensional objectives. Compared to single-objective (power generation, water supply, or ecology) operation models, the multi-objective operation model maintains a certain level of power generation while significantly reducing the probability of water supply shortages, eliminating ecological water deficits, and substantially decreasing spillage. The multi-objective operation model effectively balances the risks among power generation, water supply, and ecological goals, demonstrating its comprehensive optimization advantages under complex constraints.