A Method for Assembly Sequence Planning for Remote Maintenance of Parts in Radiation Environment

Abstract

The invention relates to an assembly sequence planning method based on the particle swarm optimization algorithm, in particular to a method for achieving remote control and maintenance assembly sequence planning of parts in a radiation environment. The method is implemented based on the adaptive mutation particle swarm optimization algorithm and improves the algorithm. The method for achieving assembly sequence planning mainly comprises the steps: (1) establishing an interference matrix and then establishing an enhanced adjacency matrix and an enhanced support matrix of the parts; (2) establishing the assembly sequence of an assembly tool set; (3) determining the feasible assembly direction according to the geometric feasibility of the assembly sequence; (4) establishing an objective function based on evaluation indexes established in the previous three steps; (5) redefining the positions and the speeds of particles and an algorithm for updating operation of the positions and the speeds; (6) carrying out assembly sequence planning. The method overcomes the defects of the particle swarm optimization algorithm in assembly sequence planning, and by redefining the positions and the speeds of the particles according to the dispersion characteristic of assembly sequence planning, the efficiency and quality of assembly sequence planning can be effectively improved.

Description

technical field [0001] The invention relates to an assembly sequence planning method based on a particle swarm algorithm, in particular to an assembly sequence planning method for parts remote maintenance maintenance in a radiation environment based on an adaptive variation particle swarm optimization algorithm and improving the algorithm. Background technique [0002] When radiation equipment such as nuclear power plants or high-energy physics devices are assembled in a specific environment, the specific environment is generally a radioactive radiation environment, and the key components involved in the assembly are likely to fail due to the high difficulty of radiation assembly in the radiation environment Improvement, if a key component fails, it will cause an unplanned shutdown, or cause serious nuclear pollution and nuclear accidents; when operating remote maintenance of parts in a radiation environment, equipment maintenance operators should be in a safe area far away f...

AI Technical Summary

Problems solved by technology

[0003] Generally speaking, using the assembly sequence planning method based on experience requires a lot of manpower and low efficiency for equipment with complex structures, and cannot reflect the feasibility of parts assembly in the design stage. Long; in order to improve the method of assembly sequence planning, in recent years, scientific research has applied virtual maintenance planning to computers, thereby establishing a virtual reality planning environment and providing effective experience for remote maintenance assembly sequence planning. However, the realization of remote maintenance assembly sequence planning is not only It is necessary to meet the requirements of the conditions in the virtual planning environment, and also need to meet the requirements of assembly time, assembly cost, assembly reliability and other conditions, and determine the optimal solution from it

[0004] In recent years, researchers have successively applied intelligent optimization algorithms such as genetic algorithm, ant colony algorithm, simulated annealing algorithm, artificial neural network, and particle swarm algorithm to solve the problem of combinatorial explosion. When the quality is poor, that is, when the proportion of the initial feasible assembly sequence is small, the convergence speed of the genetic algorithm will be very slow, and it is not easy to obtain the ideal assembly sequence; because the simulated annealing algorithm does not expand the solution space well enough, it is not easy to search for the most effective area, so the search efficiency is low, and there are also disadvantages such as poor population diversity and difficulty in obtaining the optimal assembly sequence; the artificial neural network algorithm lacks global search capabilities; the ant colony algorithm has the disadvantage of being prone to stagnation in the later stage of the search; Among other intelligent algorithms, the particle swarm optimization algorithm is a better method to solve the assembly sequence planning, but the particle swarm optimization algorithm has the disadvantage of premature convergence

Images