Reinforcement learning controller of bearingless permanent-magnet synchronous motor and construction method of reinforcement learning controller

Abstract

The invention discloses a reinforcement learning controller of a bearingless permanent-magnet synchronous motor and a construction method of the reinforcement learning controller. Input ends of a suspension winding current control module and a torque winding current control module are connected with an actor-critic module, the actor-critic module comprises an actor neural network, a critic neural network, a reinforcement signal module and an instantaneous difference module, output ends of the critic neural network and the reinforcement signal module are connected with an input end of the instantaneous difference module, an output end of the instantaneous difference module is connected with input ends of the actor neural network and the critic neural network, a displacement error and a rotational speed error are common inputs of the actor neural network, the reinforcement signal module and the critic neural network, outputs of the critic neural network are a suspension winding given current and a torque winding given current under a d-q coordinate system, rotational speed and displacement feedback information is evaluated by an actor, the actor is guided to control the suspension winding current and the torque winding current, and stable tracking control of a bearingless permanent-magnet synchronous motor system is achieved.

Description

technical field [0001] The invention relates to a bearingless permanent magnet synchronous motor, in particular to a controller and a construction method of a bearingless permanent magnet synchronous motor, which are applied to machine tool electric spindles, turbomolecular pumps, centrifuges, compressors, electromechanical energy storage, aerospace, etc. The field of special electric transmission belongs to the field of electric transmission control equipment. Background technique [0002] The controller design of bearingless permanent magnet synchronous motor often needs accurate motor parameter values ​​to assist in the design, such as speed sensorless control, vector control and so on. However, with the change of temperature, load and magnetic saturation, the parameter values ​​of the motor's stator inductance, winding resistance and rotor permanent magnet flux amplitude will change accordingly. When the actual parameter value of the motor changes greatly relative to th...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art mentioned above, to provide an adaptive reinforcement learning controller that can make the bearingless permanent magnet synchronous motor have excellent dynamic and static control performance and the construction method of the controller. Combining the actuator-evaluator algorithm in reinforcement learning in the field of intelligence with traditional vector control, it can update the controller parameters online without an accurate model of the motor, maintain the stable operation of the bearingless permanent magnet synchronous motor, and resist changes in motor parameters and Strong ability to resist load disturbance

Images