Particle swarm optimization method based on sequence approximate optimization

Abstract

The invention belongs to the related technical field of design optimization. The invention discloses a particle swarm optimization method based on sequence approximation optimization, and the method comprises the following steps: (1) determining the design space of a to-be-optimized actual engineering optimization problem according to the to-be-optimized actual engineering optimization problem, and constructing a simulation model between a design variable and a target response value for the actual engineering optimization problem; (2) determining the population number N of the particle swarm optimization algorithm according to the design space, selecting N sample points in the design space, calculating the real response values of the N sample points by adopting the simulation model, and setting the current sample point as the current population of the particle swarm optimization algorithm; (3) updating particles in the current population; (4) judging whether the particle swarm optimization algorithm converges or not, and outputting an optimal solution if the particle swarm optimization algorithm converges; otherwise, going to the step 3 until the particle swarm optimization algorithm converges. According to the method, the optimization efficiency of the particle swarm algorithm is improved, and the time cost is reduced.

Description

technical field [0001] The invention belongs to the technical field related to design optimization, and more specifically relates to a particle swarm optimization method based on sequential approximate optimization. Background technique [0002] In engineering practice, design optimization problems usually involve complex computer simulations, although with the development of computing technology, simulation software on finite element analysis (FEA) and computational fluid dynamics (CFD) can reduce the computational burden in the design process, But it still suffers from high computational cost and cannot even optimize some complex engineering design problems. In addition, the high latitude of design variables also makes it difficult to effectively optimize this expensive design problem. [0003] In the field of design optimization, when the design variable dimension is high, it will face the problems of high calculation cost and long simulation time, and heuristic algorith...

AI Technical Summary

Problems solved by technology

The particle swarm optimization method effectively combines the sequence approximate optimization and the particle swarm optimization algorithm, and uses the space division strategy for the approximate optimization of the constrained particle sequence, which improves the efficiency of the particle swarm optimization algorithm and prevents the algorithm from prematurely converging. On the time-consuming high-dimensional problems in the field of design optimization, the particle swarm optimization method can solve better solutions in less time cost

Images