Abstract: To obtain more uniformly， convergence distributed of Pareto Front （PF） and the diversity of Pareto Sets （PSs）， a crowding measurement strategy based on L_1/2 in the solution space and objective space was proposed. A non‑dominated solution decomposition and merging strategy， which clusters the solutions by using K⁃means， was designed to obtain uniformly PF and diversity PSs. For the sake of demonstrating the performance of the proposed algorithm， the experiments have been conducted on the CEC’2019 benchmark functions with five compared algorithms. Experimental results showed that the proposed algorithm could obtain a better metric value on rPSP， rHV， IGDX， IGDF.