Asynchronous motor model type predict flux linkage control method and device

Abstract

The invention discloses an asynchronous motor model type predict flux linkage control method. The asynchronous motor model type predict flux linkage control method includes: according to a mathematical model of a motor, generating the stator flux linkage vector reference value; according to the acquired stator flux linkage vector reference value, calculating the reference voltage vectors; adopting a space vector pulse width modulation method to calculate two nonzero vectors and one zero vector required for synthesizing the reference voltage vectors and action time of the reference voltage vectors; according to size relationship of the action time is summed in pairs, acquiring two sections of basic voltage vectors and action time thereof in one control period. By the arrangement, a stator flux linkage vector is taken as a control target, simpleness and practicability in calculation are achieved, and complicated weight coefficient design and adjust problem in a traditional control scheme is effectively solved; the two voltage vectors are generated, so that control effect is obviously improved by reducing steady-state torque ripple and current harmonics of a motor, and meanwhile, requirements on sampling frequency is lowered.

Description

technical field [0001] The invention relates to the field of speed regulation control of asynchronous motors, in particular to a method and device for model prediction flux linkage control of asynchronous motors. Background technique [0002] Model predictive control is an online optimization control algorithm with a simple concept. It has the advantages of simple principle, easy handling of nonlinear constraints, and easy realization of multi-variable control. In recent years, it has attracted a large number of scholars to study its application in the field of electric power transmission. . However, the traditional scheme takes electromagnetic torque and stator flux linkage amplitude as the control variables. In order to ensure that the system has good dynamic and static performance under various working conditions, repeated simulation and experiments are needed to determine the appropriate weight coefficient. Compared with the most widely used vector control method, the t...

AI Technical Summary

Problems solved by technology

However, the traditional scheme uses the electromagnetic torque and stator flux amplitude with inconsistent dimensions as the control variables. In order to ensure that the system has good dynamic and static performance under various working conditions, repeated simulations and experiments are required to determine the appropriate weight coefficient.

Compared with the most widely used vector control method, the traditional multi-vector model predictive control modulation method is complicated, and the vector selection and duty cycle calculation are cumbersome, which is not conducive to the combination with other control methods, which greatly limits the versatility and practicality of the control algorithm. sex

[0003] In order to solve the complicated design problem of weight coefficients, some scholars have proposed some solutions, but most of these methods are relatively complicated and not very practical

For example, in the document "Using a weighting factor table for FCS-MPC of induction motors with extended prediction horizon", based on the principle of minimum torque ripple, the weight coefficient is optimized online by looking up the table, the calculation expression is complex and the optimization effect is affected by the optimization step size

At the same time, in order to improve the system control performance, some methods apply two effective voltage vectors in one control cycle, such as the literature "Generalized Two-Vector-Based Model-Predictive Torque Control of Induction Motor Drives", and some methods optimize the voltage vector Switching time or multi-step prediction to reduce torque ripple, such as the literature "VariableSwitching Point Predictive Torque Control With Extended Prediction Horizon", but these methods still rely on high real-time computing power of the control system

[0004] At present, there is no better method that can meet the requirements at the same time: no need for complicated weight coefficient design and adjustment; two vectors can be non-zero vectors in one control cycle; the vector selection method is easy to understand and does not require complicated calculations; the modulation method is easy Unified with other control methods, it is easy to realize different control modes under the unified control program framework

Images