Although the traditional NSGA-II has good convergence performance in multi-objective optimization, it still has problems such as insufficient search ability of the crossover operator, premature convergence, and the polynomial mutation operator relying on random and subjective parameters, as well as a relatively slow convergence speed. To address these shortcomings, this paper improves NSGA-II at the operator level: it adopts a normal distribution-based crossover operator (NDX) to enhance the expansion ability of the search space and improve the problem of limited search range caused by the traditional SBX; at the same time, it designs an adaptive mutation operator, which dynamically adjusts the mutation probability according to the fitness of individuals, thereby enhancing the stability of population evolution, maintaining diversity, and accelerating the convergence speed. To verify the effectiveness of the improved strategy, experiments are conducted on the ZDT2 and ZDT3 benchmark test functions, and multi-objective performance indicators such as GD, SP, and DIV are used for quantitative evaluation. The experimental results show that the improved NSGA-II is significantly superior to the traditional NSGA-II in terms of convergence, uniformity of solution set distribution, and diversity, and can more stably and effectively approach the true Pareto frontier, verifying the rationality and advantages of the proposed algorithm improvement.