Optimization method for a moving-coil electromagnetic linear actuator using an improved Macaca thibetana behavior-based hierarchical GA–PSO algorithm

• Published in: Structural and multidisciplinary optimization Vol. 65; no. 9
• Main Authors: Wang, Geng, Lyu, Zhiqiang, Gao, Renjing, Tan, Cao
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.09.2022; Springer Nature B.V
• Subjects: Actuators; Algorithms; Computational Mathematics and Numerical Analysis; Consumption; Convergence; Design optimization; Engineering; Engineering Design; Exact solutions; Genetic algorithms; Industrial Application Paper; Linearity; Magnetic circuits; Mathematical models; Moving coil; Optimization; Parameters; Particle swarm optimization; Performance evaluation; Permanent magnets; Searching; Structural design; Theoretical and Applied Mechanics; Thrust; Analytical model; Genetic algorithm; Particle swarm optimization; Hybrid algorithm; Optimization method; Electromagnetic linear actuator
• ISSN: 1615-147X; 1615-1488
• DOI: 10.1007/s00158-022-03335-5

This study proposes an optimization method for the structural design of a moving-coil electromagnetic linear actuator (MCELA) with a Halbach permanent magnet array to minimize the volume of the structure and permanent magnet (PM) consumption while maintaining the output performance. To solve the fitness function of each parameter combination accurately, a novel compensatory analytical solution model is developed. The model is built based on the magnetic potential vector model, and the equivalent magnetic circuit model is used to consider the nonlinear material characteristics. To determine the optimal structural parameters of the MCELA with high efficiency, an improved Macaca thibetana behavior-based hierarchical genetic algorithm (GA)–particle swarm optimization (PSO) algorithm (referred to as MTA) is presented. The MTA adopts GA as the exploration phase for a broad search around the entire area and PSO as the exploitation phase for a local search around the promising regions identified in the exploration phase. The average position information for individuals and time-varying acceleration coefficients is introduced into the PSO to control the search, reduce premature convergence in the early optimization stage, and improve convergence in the later optimization stage. Furthermore, 23 benchmark functions are employed to evaluate the performance of the MTA; the results show that the MTA is superior to other related algorithms. A typical numerical example of an MCELA for pump applications is implemented to verify the effectiveness of the optimization method. The results show that the optimized MCELA can effectively reduce the volume and PM consumption by 2.99 × 10 5 mm 3 and 754.72 g, respectively, while ensuring the MCELA has good output performance of a thrust force of 450.1 N and thrust force linearity of 97.03%. The experimental results further validate the proposed method.