Optimization design of Tubular Permanent Magnet Linear Generator based on entropy model for wave energy conversion

-The effective use of wave energy which has huge reserves is one of the ways to solve the climate problem and energy crisis. In this paper, a Tubular Permanent Magnet Linear Generator (TPMLG) is optimized by the com-bination of comprehensive sensitive parameters method and particle swarm optimization based on entropy model (EPSO). By introducing the information entropy model, the aggregation characteristics in the process of particle swarm search are accurately analyzed, which can effectively reduce the invalid iteration of particle swarm. Firstly, the main design parameters of TPMLG are designed using the basic initial principle of permanent magnet linear machines. Then, the two-dimensional parametric finite element analysis model of the TPMLG is established, and a comprehensive sensitivity index Sc is defined to calculate the sensitivity indices according to power density and thrust fluctuation. The strong sensitivity parameters are optimized Particle Swarm Optimi-zation based on entropy model, and the Non-sensitivity parameters maintain the initial design values. Finally, an experimental prototype TPMLG was manufactured according to the optimized results. The speed of TPMLG was simulated by the universal testing machine MTS100kN, the experimental results show that the designed TPMLG can be effectively applied to the direct drive power take-off system.

Automated summary

This paper optimizes the Tubular Permanent Magnet Linear Generator (TPMLG) using a combination of comprehensive sensitive parameters method and particle swarm optimization based on entropy model (EPSO). By introducing the information entropy model, the aggregation characteristics in the process of particle swarm search are accurately analyzed, which can effectively reduce the invalid iteration of particle swarm. The experimental results show that the designed TPMLG can be effectively applied to the direct drive power take-off system.