In order to obtain the optimal process parameters for 17-4PH precipitation hardened stainless steel powder in the broadband laser cladding process, this study takes laser power, preheating temperature, and scanning speed as controllable input parameters, and takes the aspect ratio, saturation, and wetting angle of the cladding layer as output. Three factors and five levels orthogonal experiments are conducted. The response value nonlinear mathematical model constructed by the Mcquart algorithm is used to explore the influence of optimization variables and their weight ranking on the macroscopic morphology of the coating. The second generation non dominated genetic algorithm is used for optimization, and the output has the Pareto optimal frontier to determine the optimal process parameters. The results show that the optimal combination of process parameters after optimization is laser power of 1 878 W, scanning speed of 14 mm/s, preheating temperature of 200 ℃. At this time, the width to height ratio of the obtained cladding layer is 32.08, saturation is 0.76, and wetting angle is 10.2°. The average error of the optimized output is less than 5%.