Axial split-phase magnetic suspension flywheel motor suspension force numerical modeling method

Abstract

The invention discloses an axial split-phase magnetic suspension flywheel motor suspension force numerical modeling method. The method comprises the steps of simulation and experiment design, sample collection and processing, model offline training and model online optimization. On one hand, the extreme learning machine based on principal component analysis is used for improving the adaptability and robustness of the model to parameter changes, rapid and accurate modeling of small sample data is achieved, and the precision and speed of the suspension force model are improved. On the other hand, a differential evolution algorithm is introduced to optimize the network structure, so that the built model can meet the precision requirement and the real-time control requirement, the problems oftoo many hidden layer neurons and huge network structure are avoided, the calculation speed of the model is increased, and the model is more suitable for modeling of such motors.

Description

technical field [0001] The invention discloses a numerical modeling method for the levitation force of an axially separated phase-separated magnetic levitation flywheel motor, which belongs to the technical field of the magnetic levitation flywheel motor. Background technique [0002] The flywheel energy storage system is a physical energy storage device for electromechanical energy conversion. It has the advantages of large specific power, small size, long life, fast charging and discharging, clean and pollution-free, etc. It is a new type of energy storage with high research value and broad application prospects. technology. On the basis of fully retaining the high-speed and excellent characteristics of the switched reluctance motor, the axial split-phase magnetic levitation flywheel motor further improves the high-speed performance and operating efficiency of the motor through the active control of its own levitation force. Introducing it into the flywheel energy storage...

AI Technical Summary

Problems solved by technology

At present, many scholars try to use extreme learning machines in the prediction research of nonlinear models, but there are still the following main problems: First, when the correlation between input variables is strong, unnecessary input dimensions will be added, which will Not only will it lead to a rapid increase in the calculation time of the neural network, but it may also cause a waste of calculations due to the small impact of some variables on the output results; second, the ELM must randomly set the initial weights and thresholds of the network before training the network, while An optimal value cannot be given, and the choice of initial parameters will have a great impact on the final output of the network

Images