Optimal design method for block rotor switched reluctance motor

Abstract

The invention discloses an optimal design method for a block rotor switched reluctance motor. Optimization objects are KT, eta, betas, betar, hcs, hcr, Da and N; betas and betar are polar arc of a stator and polar arc of a rotor respectively; hcs and hcr are the thickness of a stator yoke and the radial height of a rotor block respectively; Da is the outside diameter of the rotor; N is the number of turns of windings; an optimization target is to reduce a torque pulse coefficient KT and increase the motor efficiency eta; an optimization object sample data space set is used as input of a fuzzy neural network; the fuzzy neural network is trained; a non-parametric model is output; in combination with a constraint condition and an optimization object, the optimization target is converted into a target function; global optimization of the optimization objects is carried out by adopting a particle swarm optimization algorithm based on self-adaptive weight adjustment; the optimal solution of the target function is output; the optimization targets of reducing the torque pulse coefficient KT and increasing the motor efficiency eta are realized; and the optimization design method disclosed by the invention is easy to implement, high in precision and rapid in convergence.

Description

technical field [0001] The invention belongs to the field of switched reluctance motors, in particular to the optimized design of switched reluctance motors. Background technique [0002] Ordinary switched reluctance motors have been widely used in aviation, electric vehicles, coal mining, wind power and home appliances due to their simple and firm structure, low cost, high reliability, and suitability for high-speed operation and harsh environments. [0003] For the structure of the segmented rotor switched reluctance motor, refer to the motor disclosed in the document with the Chinese patent application number 201520267436.7 and the name "fault-tolerant four-phase switched reluctance motor for driving electric vehicles". Such as figure 1 As shown, the segmented rotor switched reluctance motor is mainly composed of an outer stator 1, an inner rotor 3 and a rotating shaft 4, windings 2 are wound on the stator teeth of the outer stator, and the inner rotor 3 is a fan-shaped ...

AI Technical Summary

Problems solved by technology

[0004] At present, the optimal design of motor parameters is mostly based on analogy, empirical formulas, and finite elements, but this type of optimal design needs to continuously call the parameter model to obtain its output, so the process is cumbersome and the calculation efficiency is low. The serious nonlinearity of the motor makes it difficult to obtain the mathematical analytical formula of the model, so the mathematical relationship between the optimization target and various structural parameters cannot be obtained, so that the optimal proposition cannot be established

The Chinese patent application number is 201410836987.0, and the document titled "A Multi-objective Optimal Design Method for Vehicle Motors Based on Genetic Algorithms" discloses a design method using genetic algorithms for vehicle asynchronous motors, with the goal of improving power density and efficiency However, the genetic algorithm used in this optimization design method needs to carry out the evolutionary procedures of crossover, mutation and selection, which makes the process more cumbersome, and it is easy to converge to the local optimal solution in the optimization process, which is not conducive to global optimization. In addition, the research object of this optimization algorithm is ordinary asynchronous motors, not switched reluctance motors

Images