Multi-objective probabilistic optim

Multi-objective probabilistic optimization has been widely used in engineering optimization design in specific fields, mainly focusing on the use of intelligent optimization algorithms in static sampling to solve. For example, Zhang Guoxin et al. [3-7] built a MOPOP model for airborne breakthrough point decision-making, reservoir resource scheduling, and other issues. The weighted objective function conversion model was converted to a single-objective probability optimization model, followed by random simulation or fuzzy simulation, neural networks, and compromise algorithms. Obtained a single-objective hybrid genetic algorithm solution; several scholars [10-12] optimized the MOPOP model for car charging station planning and other projects, using membership functions and weight coefficients to transform the model into a single-objective programming model and design an improved single-objective bat Algorithm, discrete (or continuous) particle swarm algorithm. For the study of the multi-objective intelligent optimization algorithm to solve MOPOP, Virivinti et al. [8-9] used fuzzy simulation or random simulation and Latin hypercube sampling to deal with noise for industrial grinding processing with uncertain parameters and optimal load reduction of the power grid. Then use the fast non-dominated sorting genetic algorithm (NSGA-II) to solve; Yang Xinyu et al. [13-14] coordinated scheduling of MRO service resources, etc., using random simulation, BP neural network to suppress noise, using multi-objective particle swarm algorithm or multi-objective Differential algorithm solving. In the research on the solution of the deterministic transformation model, Li Xiaona et al. [1-2] established a multi-objective opportunity constraint programming model for multi-source joint scheduling and other issues, and transformed the model into an analytical single-objective optimization with complex transformation and linear weighting. model. In addition, for the optimization of wind and fire joint scheduling described as MOPOP, Zhuo Huan et al. [15-16] used dual interior point method, normal boundary crossing method, and hybrid interactive fuzzy programming method to solve the problem, and selected a compromise through the sequential preference approximate ideal solution technology. solution.