Abstract: In response to the drawbacks of existing multi-objective evolutionary algorithms in handling complex Pareto frontier issues, such as insufficient selection pressure and uneven distribution, a novel high-dimensional multi-objective evolutionary algorithm, MOEA/D-DAM, is proposed. This algorithm adopts a dynamic association strategy (DAM strategy) and an adaptive weight update strategy based on target space transformation (OTS strategy) to enhance the exploration capability, quality, and diversity of the algorithm. The DAM strategy connects the weight vector with the population individuals through the Chebyshev weighting approach, achieving the dynamic association between the population and the weight vector. The OTS strategy, based on target space transformation and adaptive weight update, estimates the curvature of the Pareto frontier and adjusts the population coordinates via the target space coordinates when the curvature is less than 1. The algorithm effectively ameliorates the distribution of individuals on the Pareto frontier, enabling the population to converge rapidly while maintaining diversity. MOEA/D-DAM is compared with advanced algorithms MOEA/D-UR, MOEA/D-URAW, MOEA/D-VOV, PeEA, and TS-NSGA-Ⅱ on the DLTZ test problem and WFG test problem through simulation experiments. The experimental results indicate that MOEA/D-DAM outperforms other algorithms in the IGD performance indicator on 43, 48, 52, 49, and 40 problems, with a best solution ratio of 54.6%. It shows superior performance in the HV performance indicator on 28, 46, 42, 37, and 43 problems, with a best solution ratio of 42.1%. The results suggest that the algorithm demonstrates strong competitiveness in solving complex Pareto problems.