Speed-sensor-less construction method for bearing-less asynchronous motor

Abstract

The invention discloses a speed-sensor-less construction method for a bearing-less asynchronous motor, which comprises the following steps of: firstly, constructing an embedded rotating speed subsystem for an original bearingless asynchronous motor system; then, establishing a bearingless asynchronous motor rotating speed left-inverse system of the embedded rotating speed subsystem; subsequently, adopting a static neural network and two differentiators S to form a support vector machine inverse, training the support vector machine, and adjusting and determining all weight coefficients of the support vector machine so as to realize the bearing-less asynchronous motor rotating speed left-inverse system; and finally, inversely connecting the support vector machine behind the original bearing-less asynchronous motor system in series so as to finish the speed-sensor-less construction. The speed-sensor-less control strategy of the bearing-less asynchronous motor based on support vector machine left-inverse omits a mechanical speed sensor and an interface circuit thereof of the original control system, so that the cost of the control system is lowered, and rotating speed identification can be quickly and accurately carried out in the full-speed range. The speed-sensor-less construction method for a bearing-less asynchronous motor has strong adaptability, fault tolerance and robustness.

Description

technical field [0001] The invention is a construction method of a bearingless asynchronous motor without a speed sensor based on the left inverse of a support vector machine, which provides a new control strategy for the speedless operation of a bearingless asynchronous motor, and is suitable for high-performance control of a bearingless asynchronous motor. It belongs to the technical field of electric drive control equipment. Background technique [0002] The high-performance control of a bearingless asynchronous motor is inseparable from the closed-loop control of the speed, so it is necessary to detect the speed of the rotor. The traditional motor rotor speed is detected by a mechanical speed sensor, which has problems such as installation, connection and reliability. For bearingless asynchronous motors, the use of mechanical speed sensors has greater limitations, because the sensor itself is mechanically difficult to achieve high-speed and ultra-high-speed operation of ...

AI Technical Summary

Problems solved by technology

These speed identification methods do not consider the influence of factors such as rotor eccentricity, magnetic saturation, temperature rise, and load changes during high-speed operation, and have not really realized accurate online identification of speed parameters.

The speed sensorless method based on the neural network inverse system uses the powerful approximation ability of the static neural network to the nonlinear function to break through the bottleneck of the inverse system in the implementation or application, but the neural network itself has small local poles, over-learning, and a large amount of calculation. Defects, which limit the further application of the neural network inverse method, also affect its control performance

Images