Study On The Heuristic Random Key Genetic Optimization Method Of Packing Payloads For Multi-module Satellite

Constrained layout problems originate from many practical engineering problems,such as packing payloads for multi-module satellite,transportation route design,industrial plant and equipment layout design,advanced equipment manufacturing,and steel company steel plate design.This paper discusses the optimization of multi-cabin satellite Payloads packing with performance constraints.Specifically,the problem is how to place orthogonal satellite Payloads on four bearing plates,so that the rotational inertia of the entire packing scheme is as small as possible and meets the given constraints.This problem belongs to the NP-hard problem.The current solution strategy mainly includes an evolutionary algorithm with random initial points,and a combination of sequence optimization of the Payload placement sequence through the evolutionary algorithm and a heuristic placement rule.The former has the problem of huge search space interference judgment and extremely time-consuming.It is easy to fall into local optimum by solving Payload allocation and packing subsections.Neither of them organically combines Payload allocation and packing optimization.Inspired by the free lunch theorem and previous work,this paper studies the random key heuristic construction and genetic and particle swarm collaborative optimization methods for the multi-cabin satellite load allocation and filling problem.Its main innovations include the following two aspects.(1)A heuristic random key genetic algorithm is proposed to solve the Payloads allocation and packing optimization of multi-cabin satellite packing.Acquire relevant knowledge of balance mechanics and inertia definition,construct corresponding knowledge individuals for algorithm initialization,and integrate knowledge into the initial population of evolutionary algorithm;propose a two-part encoding and decoding rule of random key genetic algorithm to realize Payloads packing Collaborative optimization of allocation and single-cabin packing to expand the global search space;and a new objective function is constructed,and the optimization index of the envelope radius in the objective function is automatically adjusted during the iteration process to quickly find the balance point of the moment of inertia and envelope radius optimization.Experimental data shows that the algorithm is higher than existing methods in terms of accuracy and efficiency.(2)Propose an improved regional positioning heuristic algorithm and combine it with the proposed random key genetic algorithm to solve the optimization of Payloads allocation and packing of multi-cabin satellites with fixed position Payloads.The improved area positioning is to prioritize the placement of fixed Payloads,and then place non-fixed Payloads in the area positioning rules,and at the same time enable the Payloads to move between compartments.The population is initialized based on the balance mechanics and the relevant knowledge of the model,and the proposed random key genetic algorithm is used to realize the coordinated optimization of the allocation and packing of the Payloads set with a fixed position.Experimental data shows that the algorithm is higher than existing methods in overall performance optimization,stability and computational efficiency.In this paper,the international commercial communication satellite is mainly used to study the model of satellite module,and at the same time,the model of the return satellite module with fixed Payload is also studied.The core idea of this paper is to carry out collaborative optimization of Payload allocation and constrained packing of multi-cabin satellites combined with regional positioning strategy to solve the packing scheme to improve the calculation accuracy and efficiency.It is hoped that the algorithms and strategies in this paper can inspire more complex layout optimization problems.