CN-122001266-A - Motor operation deviation self-adaptive regulation and control system

CN122001266ACN 122001266 ACN122001266 ACN 122001266ACN-122001266-A

Abstract

The invention relates to the technical field of motor control, in particular to a motor operation deviation self-adaptive regulation and control system which comprises a multi-mode state data acquisition unit for generating multi-mode state data, an implicit working condition modeling unit for carrying out normalization and weighting calculation by combining a preset reference value, an upper limit value and a fusion weight to generate an implicit working condition characteristic vector, a deviation trend prediction unit for carrying out calculation by adopting an autoregressive integral model to generate a feedforward deviation index, a risk level judgment unit for generating a risk level judgment result, a dynamic compensation factor generation unit for carrying out calculation by adopting a nonlinear mapping function to generate a dynamic compensation factor, and a final instruction synthesis unit for combining the dynamic compensation factor and a conventional feedback control instruction to generate a corrected control instruction.

Inventors

YU LILI
GUO JIONGXIAN
GUO BINGXUN
YANG YONGYU

Assignees

淮安仲益电机有限公司

Dates

Publication Date: 20260508
Application Date: 20260408

Claims (7)

1. A motor run bias adaptive regulation and control system, comprising: The multi-mode state data acquisition unit is used for synchronously acquiring stator phase current, vibration acceleration and core component temperature to generate multi-mode state data; The implicit working condition modeling unit is used for carrying out normalization and weighting calculation based on the multi-mode state data and combining a preset reference value, an upper limit value and a fusion weight to generate an implicit working condition feature vector; The deviation trend prediction unit is used for calculating by adopting an autoregressive integral model according to the time sequence information of the implicit working condition feature vector to generate a feedforward deviation index; The risk level judging unit is used for comparing the feedforward deviation index with a preset risk threshold value to generate a risk level judging result; the dynamic compensation factor generation unit is used for generating a dynamic compensation factor by calculating through a nonlinear mapping function according to the feedforward deviation index; And the final instruction synthesis unit is used for acquiring a conventional feedback control instruction and generating a corrected control instruction by combining the dynamic compensation factor and the conventional feedback control instruction.
2. The adaptive control system for motor operation deviation according to claim 1, wherein the implicit operating mode modeling unit generates the implicit operating mode feature vector by: acquiring real-time acquisition values of stator phase current, vibration acceleration and core component temperature; the real-time acquisition value of various data is differenced with a preset corresponding reference value, and the difference value is divided by the difference value between the preset corresponding upper limit value and the reference value to obtain a normalized value; multiplying the normalized values of various data with the corresponding preset fusion weights, and summing all the products to generate an implicit working condition feature vector.
3. The adaptive motor operating bias control system according to claim 1, wherein the bias trend prediction unit generates the feedforward bias index as follows: Acquiring a working condition characteristic vector at the current moment and a working condition characteristic vector at the previous moment, and multiplying the difference value of the working condition characteristic vector at the current moment and the working condition characteristic vector at the previous moment by a preset trend sensitivity coefficient to obtain change increment information; multiplying the working condition characteristic vector at the current moment by a preset state sensitivity coefficient to obtain state stock information; and summing the deviation index value, the change increment information and the state stock information at the current moment to generate a feedforward deviation index.
4. A motor run deviation adaptive regulation and control system according to claim 3, wherein the system further comprises a predictive model adaptation unit; The prediction model self-adaptive unit is used for acquiring the real-time rotating speed and the reference rotating speed of the motor, calculating errors of the real-time rotating speed and the reference rotating speed, and carrying out normalization processing to obtain a real deviation index; Comparing the real deviation index with the feedforward deviation index to obtain a prediction error; and according to the prediction error, adopting a least mean square or recursive least square algorithm to iteratively update the trend sensitivity coefficient and the state sensitivity coefficient.
5. The motor running deviation adaptive control system according to claim 1, wherein the risk level determination unit determines the following: Presetting an early warning area threshold value and a dangerous area threshold value; when the feedforward deviation index is smaller than or equal to the early warning area threshold value, determining a risk level judging result as a safety area; When the feedforward deviation index is larger than the early warning area threshold value and smaller than or equal to the dangerous area threshold value, determining a risk level judging result as an early warning area, and starting primary compensation; and when the feedforward deviation index is larger than the dangerous area threshold value, determining a risk level judging result as the dangerous area, and starting secondary reinforcement compensation.
6. The adaptive control system for motor running deviation according to claim 1, wherein the process of generating the corrected control command by the final command synthesizing unit is as follows: Adding one to the dynamic compensation factor to obtain a correction coefficient; and multiplying the correction coefficient by a conventional feedback control command to generate a corrected control command.
7. The motor operation deviation adaptive control system according to claim 2, wherein the system further comprises a fusion model adaptive unit; The fusion model self-adaptive unit is used for analyzing the correlation between various multi-mode state data in the historical data and known deviation events; And according to the correlation, adopting a gradient descent optimization algorithm to carry out online self-adaptive adjustment on the fusion weight.

Description

Motor operation deviation self-adaptive regulation and control system Technical Field The invention relates to the technical field of motor control, in particular to a motor operation deviation self-adaptive regulation and control system. Background The motor is a core power device in industrial production, and the stable and reliable operation of the motor is important; to ensure that the motor accurately operates in accordance with the preset command, a closed loop feedback control system is generally employed, which monitors the actual operating state of the motor by means of a sensor, and when a deviation from the desired state is detected, the controller adjusts the output command to correct the deviation; in the situations that the motor always faces complex working conditions such as dynamically-changed load or potential fault disturbance and the like in practical application, the traditional feedback control exposes inherent limitations, and the core problem is that the control logic is passive, namely, the deviation responds after the deviation occurs; Therefore, when the motor encounters severe load fluctuation or initial failure, the conventional control manner may not effectively inhibit rapid expansion of deviation, which may result in a larger operation deviation peak value and a longer system recovery time, thereby affecting dynamic performance and stability of the system, reducing operation robustness of the motor under complex working conditions, and bringing hidden trouble to equipment safety and production continuity. Disclosure of Invention In order to solve the technical problems, the invention provides a motor operation deviation self-adaptive regulation and control system, and specifically, the technical scheme of the invention is as follows: a motor run bias adaptive regulation and control system, comprising: The multi-mode state data acquisition unit is used for synchronously acquiring stator phase current, vibration acceleration and core component temperature to generate multi-mode state data; The implicit working condition modeling unit is used for carrying out normalization and weighting calculation based on the multi-mode state data and combining a preset reference value, an upper limit value and a fusion weight to generate an implicit working condition feature vector; The deviation trend prediction unit is used for calculating by adopting an autoregressive integral model according to the time sequence information of the implicit working condition feature vector to generate a feedforward deviation index; The risk level judging unit is used for comparing the feedforward deviation index with a preset risk threshold value to generate a risk level judging result; the dynamic compensation factor generation unit is used for generating a dynamic compensation factor by calculating through a nonlinear mapping function according to the feedforward deviation index; And the final instruction synthesis unit is used for acquiring a conventional feedback control instruction and generating a corrected control instruction by combining the dynamic compensation factor and the conventional feedback control instruction. Preferably, the process of generating the implicit working condition feature vector by the implicit working condition modeling unit is as follows: acquiring real-time acquisition values of stator phase current, vibration acceleration and core component temperature; the real-time acquisition value of various data is differenced with a preset corresponding reference value, and the difference value is divided by the difference value between the preset corresponding upper limit value and the reference value to obtain a normalized value; multiplying the normalized values of various data with the corresponding preset fusion weights, and summing all the products to generate an implicit working condition feature vector. Preferably, the process of generating the feedforward bias index by the bias trend prediction unit is as follows: Acquiring a working condition characteristic vector at the current moment and a working condition characteristic vector at the previous moment, and multiplying the difference value of the working condition characteristic vector at the current moment and the working condition characteristic vector at the previous moment by a preset trend sensitivity coefficient to obtain change increment information; multiplying the working condition characteristic vector at the current moment by a preset state sensitivity coefficient to obtain state stock information; and summing the deviation index value, the change increment information and the state stock information at the current moment to generate a feedforward deviation index. Preferably, the system further comprises a prediction model adaptation unit; The prediction model self-adaptive unit is used for acquiring the real-time rotating speed and the reference rotating speed of the motor, calculating errors of the real-time