CN112800644A

CN112800644A - Servo motor high-precision simulation method based on data driving

Info

Publication number: CN112800644A
Application number: CN202110003850.7A
Authority: CN
Inventors: 陈恩涛; 卓亮; 赵飞; 葛发华; 陈强; 葛红岩
Original assignee: Guizhou Aerospace Linquan Motor Co Ltd
Current assignee: Guizhou Aerospace Linquan Motor Co Ltd
Priority date: 2021-01-04
Filing date: 2021-01-04
Publication date: 2021-05-14

Abstract

The invention provides a high-precision simulation method of a servo motor based on data driving, which comprises the following steps: presetting: setting motor characteristic parameters based on simulation; actually measuring: carrying out entity test on the motor to obtain test data; and (6) correcting. The invention can use the torque-current test data of a plurality of product prototypes as samples, adopts the 'big data' prediction idea, performs the mechanical characteristic performance fitting of the actual servo motor based on the test data, and the more sufficient the data in the product characteristic database along with the increase of the number of the samples, the more the fitted torque characteristic curve can truly reflect the torque characteristic change caused by the factors of component processing, material heat treatment, electromechanical material magnetic conductivity deviation, assembly and the like in the manufacturing process, thereby providing the torque characteristic which is more suitable for the actual situation for the design of the servo motor controller, realizing the optimization of the servo motor control strategy and achieving the purpose of high-precision control of the servo motor.

Description

Servo motor high-precision simulation method based on data driving

Technical Field

The invention relates to a high-precision simulation method of a servo motor based on data driving, and belongs to the technical field of electrical engineering.

Background

With the need of modern technology development, the development of sophisticated equipment becomes the key point of the scientific and technological strength of the new and big countries. The servo system is an important component of advanced technologies such as precise action, a follow-up system and the like, and the performance sub-link thereof determines the development of weaponry to the aspects of high intelligence, high automation, high reliability and the like. However, the servo motor system is a strongly coupled, nonlinear and multivariable time-varying parameter system, and is oriented to the environment and working conditions, and the current full-characteristic simulation technology of the servo motor system cannot realize motor and drive integrated modeling, so that characteristic indexes such as speed regulation range, dynamic response, control precision, torque stability and the like are caused, and research and development of advanced equipment in China are directly restricted. The full-characteristic high-precision intelligent simulation technology of the servo motor can greatly promote the development of the optimal design technology of the servo motor and the driving system and reduce the research and development period and cost on the one hand, and can effectively improve the simulation precision of the ground simulation technology of the whole equipment system on the other hand, thereby realizing the perfect matching of the servo system and the equipment. The servo motor simulation technology based on data driving fully considers the factors of the product manufacturing process, obtains the data characteristics closer to the real product, and effectively improves the control precision of the servo motor.

The traditional servo motor simulation technology usually adopts an empirical formula or a finite element simulation calculation method to obtain a current-torque curve (T-I curve) of the motor, and a control strategy is designed based on the T-I curve in a matching way to complete the servo control of the driving motor. However, the T-I curve obtained by empirical formula or finite element simulation is the most ideal motor load characteristic, and the control strategy is most effective only under the condition that the matching degree of the actual motor performance and the ideal T-I curve is high. Under the influence of production factors, characteristic parameters such as the structure size, the resistance parameter, the inductance parameter and the like of an actual motor are different from a theoretical model, so that a certain deviation exists between an actual mechanical characteristic curve and a theoretical mechanical characteristic curve of the motor, and a control strategy designed based on a theoretical T-I curve cannot realize high-precision control of a prototype.

Disclosure of Invention

In order to solve the technical problems, the invention provides a high-precision simulation method of a servo motor based on data driving, which adopts a data fitting method to obtain the working characteristics of a real motor sample, can provide more real motor load characteristics for controller design, and realizes high-precision control of the servo motor.

The invention is realized by the following technical scheme.

The invention provides a high-precision simulation method of a servo motor based on data driving, which comprises the following steps:

presetting: setting motor characteristic parameters based on simulation;

actually measuring: carrying out entity test on the motor to obtain test data;

and (3) correction: and correcting the set characteristic parameters according to the test data to obtain a simulation result.

The simulation is carried out by constructing a finite element analysis model of the motor.

The motor characteristic parameter is a motor torque characteristic curve which takes current input as an independent variable and takes torque output as a dependent variable.

The test data is a motor torque characteristic point with current input as an independent variable and torque output as a dependent variable.

The motor torque characteristic curve is obtained by performing quadratic polynomial fitting on the characteristic points.

In the simulation step, quadratic polynomial fitting is firstly carried out on the test data, and then correction is carried out.

In the simulation step, the motor characteristic parameters are corrected to expand a motor torque characteristic curve into a motor torque characteristic curve range, and the motor torque characteristic curve range is limited by an upper boundary curve and a lower boundary curve.

When the motor is subjected to physical testing, a plurality of motors are tested; and when the second-order polynomial fitting is carried out on the test data, acquiring a curve range which contains all normal test data and is determined by the upper test boundary curve and the lower test boundary curve.

The invention has the beneficial effects that: the torque-current test data of a plurality of product prototypes can be used as samples, a big data prediction idea is adopted, the mechanical characteristic performance fitting of an actual servo motor is carried out based on test data, along with the increase of the number of the samples, the more sufficient the data in a product characteristic database is, the more the fitted torque characteristic curve can truly reflect the torque characteristic change caused by the factors of part processing, material heat treatment, electromechanical material magnetic conductivity deviation, assembly and the like in the manufacturing process, the torque characteristic which is more suitable for the actual situation can be provided for the design of a servo motor controller, the optimization of a servo motor control strategy is realized, and the aim of high-precision control of the servo motor is fulfilled.

Drawings

FIG. 1 is a schematic flow diagram of one embodiment of the present invention;

FIG. 2 is a comparison of simulated operating conditions and a simulated fit curve K0 according to one embodiment of the present invention;

FIG. 3 is a comparison of an actual motor torque curve K1 and a simulated fit curve K0 in accordance with one embodiment of the present invention;

FIG. 4 is a schematic diagram of an extracted boundary curve of an actual characteristic curve according to an embodiment of the present invention;

fig. 5 is a schematic diagram of K20 and K21 curves corrected based on actual characteristic curves in an embodiment of the present invention.

Detailed Description

The technical solution of the present invention is further described below, but the scope of the claimed invention is not limited to the described.

presetting: setting motor characteristic parameters based on simulation;

actually measuring: carrying out entity test on the motor to obtain test data;

The motor characteristic parameter is a motor torque characteristic curve with a current input as an independent variable and a torque output as a dependent variable.

The test data is a motor torque characteristic point with the current input as an independent variable and the torque output as a dependent variable.

In the simulation step, the motor characteristic parameters are corrected to expand the motor torque characteristic curve into a motor torque characteristic curve range, and the motor torque characteristic curve range is limited by an upper boundary curve and a lower boundary curve.

Example 1

The scheme is adopted, and comprises the following steps as shown in figure 1:

step 1, selecting electromagnetic parameters of the motor, and constructing a finite element analysis model of the motor to obtain an ideal mechanical characteristic curve of the motor.

And 2, calculating the torque characteristics of the motor under different currents by adopting a finite element method to obtain the theoretical relationship between the currents and the torques under multiple working conditions, wherein the input is the current I0, the output is the torque Te0, and a plurality of (I0 and Te0) working condition points are formed, as shown in figure 2.

And 3, combining a quadratic spline interpolation fitting method to form a complete simulation fitting curve K0 of the plurality of (I0, Te0) working condition points, wherein the curve is a torque characteristic curve of the motor in an ideal state. Because the curve is a quadratic polynomial, the numerical value of the coefficient of the quadratic polynomial can be obtained through fitting, and the torque output under any current can be calculated by means of a simulation fitting curve K0.

Step 4, due to the difference of the manufacturing process, the output torque characteristics of the plurality of motors do not particularly coincide with the simulation fitting curve K0 of the theoretical analysis result, but have a certain deviation, as shown in fig. 3. By building a characteristic database for actual operation data of a plurality of motors, a variation range of the torque characteristic K1 when the motors actually operate can be obtained, as shown in fig. 3. All the characteristic data in the K1 are extracted to obtain actual characteristic envelope curves meeting the index requirements, namely an upper boundary curve K10 and a lower boundary curve K11, as shown in fig. 4.

Step 5 corrects K0 using the actual characteristic database network curves K10, K11. For step 2, K0 calculated using finite elements is only a determined characteristic curve, and after correction using K10 and K11, K0 is expanded into a characteristic curve range, which is covered by upper boundary K20 curves and lower boundary K21 curves of the characteristic curve boundary, as shown in fig. 5.

Step 6, making a motor control strategy based on the torque characteristic K1, wherein the made control strategy is only suitable for a part of motors with actual output torque matched with a calculation result in a particularly accurate manner; by using the corrected torque characteristic envelope curves K20 and K21 to make a motor control strategy, a control strategy applicable to both of the corresponding torque characteristics can be obtained, instead of the conventional control strategy for only K1.

And 7, correcting K0 by adopting K10 and K11 to form characteristic curve boundaries K20 and K21 which comprise curves, fully considering the discreteness among motor individuals, ensuring that a control strategy suitable for all motors can be worked out even if the motors have individual differences, and ensuring that all motors in a certain difference range can work with high precision.

Claims

1. A high-precision simulation method of a servo motor based on data driving is characterized by comprising the following steps: the method comprises the following steps:

presetting: setting motor characteristic parameters based on simulation;

actually measuring: carrying out entity test on the motor to obtain test data;

correcting: and correcting the set characteristic parameters according to the test data to obtain a simulation result.

2. The high-precision simulation method of the servo motor based on the data driving as claimed in claim 1, wherein: the simulation is carried out by constructing a finite element analysis model of the motor.

3. The high-precision simulation method of the servo motor based on the data driving as claimed in claim 1, wherein: the motor characteristic parameter is a motor torque characteristic curve which takes current input as an independent variable and takes torque output as a dependent variable.

4. The high-precision simulation method of the servo motor based on the data driving as claimed in claim 1, wherein: the test data is a motor torque characteristic point with current input as an independent variable and torque output as a dependent variable.

5. The high-precision simulation method of the servo motor based on the data driving as claimed in claim 3, wherein: the motor torque characteristic curve is obtained by performing quadratic polynomial fitting on the characteristic points.

6. The high-precision simulation method of the servo motor based on the data driving as claimed in claim 1, wherein: in the simulation step, quadratic polynomial fitting is firstly carried out on the test data, and then correction is carried out.

7. A high-precision simulation method of a servo motor based on data driving according to claim 1 or 3, wherein: in the simulation step, the motor characteristic parameters are corrected to expand a motor torque characteristic curve into a motor torque characteristic curve range, and the motor torque characteristic curve range is limited by an upper boundary curve and a lower boundary curve.

8. The high-precision simulation method of the servo motor based on the data driving as claimed in claim 1 or 6, wherein: when the motor is subjected to physical testing, a plurality of motors are tested; and when the second-order polynomial fitting is carried out on the test data, acquiring a curve range which contains all normal test data and is determined by the upper test boundary curve and the lower test boundary curve.