Servo motor control method and system based on multi-dimensional real-time monitoring

Abstract

The invention provides a servo motor control method and system based on multi-dimensional real-time monitoring, and relates to the field of servo motor control, and the method comprises the steps: determining vibration monitoring positions and temperature monitoring positions of a plurality of sample servo motors through finite element analysis; a plurality of vibration sensors and a plurality of temperature sensors are arranged on the servo motor; carrying out an output torque compensation experiment on the servo motor through a plurality of vibration sensors and a plurality of temperature sensors, and establishing an individual deviation compensation model of the servo motor; in the operation process of the servo motor, real-time load characteristics and real-time motor temperature characteristics of the servo motor are determined based on the operation state data sets collected by the multiple vibration sensors and the multiple temperature sensors; based on the real-time load characteristics, the real-time motor temperature characteristics and the individual deviation compensation model of the servo motor, output torque compensation is carried out on the servo motor, and the method has the advantage that the accuracy and the real-time performance of output torque compensation of the servo motor are improved.

Description

Technical Field

[0001] The present invention relates to the field of servo motor control, and in particular to a servo motor control method and system based on multi-dimensional real-time monitoring. Background Art

[0002] In the field of modern industrial automation and precision manufacturing, precise control of servo motors is the key to achieving high-quality processing. As the core component of driving various mechanical equipment, the performance of servo motors is directly related to the operating efficiency and processing accuracy of the equipment. Especially in situations where high-precision torque control is required, such as CNC machine tools, precision assembly robots, power tools, etc., it is crucial to ensure the consistency between the output torque of the servo motor and the command torque. In actual application, the output torque of the servo motor is often affected by a variety of complex factors, resulting in deviations from the preset torque command. Among them, the temperature and load condition of the servo motor are the two most significant influencing factors. When the servo motor runs for a long time or is under high load, a large amount of heat will be generated inside due to friction, electromagnetic effects, etc., causing the motor temperature to rise. The increase in temperature will not only affect the physical properties of the internal materials of the motor, such as resistivity, magnetic permeability, etc., but also change the thermal expansion coefficient of the motor, thereby affecting the geometric dimensions and assembly accuracy of the motor. These changes act together on the electromagnetic conversion process of the motor, ultimately leading to fluctuations in the output torque. On the other hand, changes in load conditions will also affect the output torque of the servo motor. The size of the load, the difference in inertia, and the dynamic changes of the load will directly affect the dynamic response characteristics and steady-state output characteristics of the motor. Especially under sudden load changes or complex working conditions, the deviation between the motor's output torque and the command torque may increase significantly, seriously affecting the processing accuracy and stability of the equipment.

[0003] To address this problem, existing output torque compensation technologies monitor the motor's operating status in real time, including parameters such as temperature, current, and speed, and combine advanced control algorithms to adjust the motor's output torque in real time to reduce the deviation between it and the command torque. However, existing output torque compensation technologies still have certain limitations. For example, some technologies are only optimized for specific working conditions or load conditions and lack wide applicability; some technologies are insufficient in real time due to high algorithm complexity and large amount of calculation, making it difficult to achieve accurate compensation under rapidly changing working conditions.

[0004] Therefore, it is necessary to provide a servo motor control method and system based on multi-dimensional real-time monitoring to improve the accuracy and real-time performance of the output torque compensation of the servo motor. Summary of the invention

[0005] The present invention provides a servo motor control method based on multi-dimensional real-time monitoring, comprising: determining vibration monitoring positions and temperature monitoring positions of multiple sample servo motors through finite element analysis; arranging multiple vibration sensors and multiple temperature sensors on the servo motor based on the vibration monitoring positions and temperature monitoring positions of the multiple sample servo motors; performing an output torque compensation experiment on the servo motor through the multiple vibration sensors and the multiple temperature sensors, and establishing an individual deviation compensation model of the servo motor; during the operation of the servo motor, determining the real-time load characteristics and real-time motor temperature characteristics of the servo motor based on an operation status data set collected by the multiple vibration sensors and the multiple temperature sensors; and performing output torque compensation on the servo motor based on the real-time load characteristics, the real-time motor temperature characteristics and the individual deviation compensation model of the servo motor.

[0006] Furthermore, vibration monitoring positions and temperature monitoring positions of multiple sample servo motors are determined through finite element analysis, including: for each sample servo motor, the vibration monitoring position of the sample servo motor is determined through finite element analysis; for each sample servo motor, the multiple initial temperature monitoring positions of the sample servo motor are determined through finite element analysis, based on the multiple initial temperature monitoring positions of the sample servo motor, multiple temperature sensors are installed on the sample servo motor, the sample servo motor is tested, experimental temperature data and motor torque deviation data of the multiple initial temperature monitoring positions under different loads are obtained, and the temperature monitoring positions are screened from the multiple initial temperature monitoring positions.

[0007] Furthermore, the vibration monitoring position of the sample servo motor is determined through finite element analysis, including: constructing a geometric model corresponding to the sample servo motor; meshing the geometric model corresponding to the sample servo motor to generate a finite element model corresponding to the sample servo motor; determining multiple initial vibration monitoring positions on the finite element model corresponding to the sample servo motor; obtaining simulated vibration data of multiple initial vibration monitoring positions of the finite element model corresponding to the sample servo motor under different loads; for each initial vibration monitoring position, based on the simulated vibration data of the initial vibration monitoring position under different loads, calculating a first correlation coefficient between the initial vibration monitoring position and the load; based on the first correlation coefficient between each initial vibration monitoring position and the load, screening multiple candidate vibration monitoring positions from the multiple initial vibration monitoring positions; testing the sample servo motor to obtain experimental vibration data of multiple candidate vibration monitoring positions under different loads; for each candidate vibration monitoring position, based on the experimental vibration data of the candidate vibration monitoring position under different loads, calculating a second correlation coefficient between the candidate vibration monitoring position and the load; based on the second correlation coefficient between each candidate vibration monitoring position and the load, screening the vibration monitoring position from the multiple candidate vibration monitoring positions.

[0008] Further, the temperature monitoring positions are screened from a plurality of initial temperature monitoring positions, including: for each initial temperature monitoring position, based on the experimental temperature data of the initial temperature monitoring position under different loads, calculating the third correlation coefficient between the initial temperature monitoring position and the load, and based on the experimental temperature data of the initial temperature monitoring position under different loads and the motor torque deviation data, calculating the fourth correlation coefficient between the initial temperature monitoring position and the motor torque deviation; based on the third correlation coefficient between each initial temperature monitoring position and the load and the fourth correlation coefficient between each initial temperature monitoring position and the motor torque deviation, screening a plurality of candidate temperature monitoring positions from a plurality of initial temperature monitoring positions; based on the motor torque deviation data under different loads, calculating the fifth correlation coefficient between the load and the motor torque deviation; for each candidate temperature monitoring position, based on the third correlation coefficient between each initial temperature monitoring position and the load, the fourth correlation coefficient between each initial temperature monitoring position and the motor torque deviation, and the fifth correlation coefficient between the load and the motor torque deviation, calculating the sixth correlation coefficient between the candidate temperature monitoring position and the motor torque deviation; based on the sixth correlation coefficient between each candidate temperature monitoring position and the motor torque deviation, screening the temperature monitoring position from a plurality of candidate temperature monitoring positions.

[0009] Furthermore, based on the vibration monitoring positions and temperature monitoring positions of multiple sample servo motors, multiple vibration sensors and multiple temperature sensors are set on the servo motor, including: acquiring structural information of the servo motor; determining similar sample servo motors from multiple sample servo motors based on the structural information of the servo motor; and setting multiple vibration sensors and multiple temperature sensors on the servo motor based on the vibration monitoring positions and temperature monitoring positions of the similar sample servo motors.

[0010] Furthermore, an output torque compensation experiment is performed on the servo motor through multiple vibration sensors and multiple temperature sensors, and an individual deviation compensation model of the servo motor is established, including: for each sample servo motor, establishing and training a torque compensation model corresponding to the sample servo motor; determining multiple target experimental scenarios, wherein the target experimental scenarios include target loads and target temperatures; for each target experimental scenario, determining a standard output torque compensation based on the target load and target temperature through the torque compensation models corresponding to similar sample servo motors, obtaining the actual output torque of the servo motor based on the standard output torque compensation, and determining the output torque deviation of the servo motor corresponding to the target experimental scenario based on the actual output torque of the servo motor; and establishing an individual deviation compensation model of the servo motor based on the output torque deviation of the servo motor corresponding to each target experimental scenario.

[0011] Furthermore, multiple target experimental scenarios are determined, including: determining multiple candidate experimental scenarios, wherein the candidate experimental scenarios include candidate loads and candidate temperatures; obtaining output torque deviations of multiple test servo motors of similar sample servo motors in multiple candidate experimental scenarios, wherein the structure of the test servo motor is consistent with the structure of the similar sample servo motor; for each candidate experimental scenario, based on the output torque deviations of multiple test servo motors of similar sample servo motors in the candidate experimental scenario, calculating the individual deviation difference coefficient corresponding to the candidate experimental scenario; based on the individual deviation difference coefficient corresponding to each candidate experimental scenario, screening the target experimental scenario from the multiple candidate experimental scenarios.

[0012] Furthermore, based on the operating status data set collected by multiple vibration sensors and multiple temperature sensors, the real-time load characteristics and real-time motor temperature characteristics of the servo motor are determined, including: establishing a feature determination model corresponding to each sample servo motor; determining the real-time load characteristics and real-time motor temperature characteristics of the servo motor based on the operating status data set collected by multiple vibration sensors and multiple temperature sensors through the feature determination model corresponding to similar sample servo motors.

[0013] Furthermore, based on the real-time load characteristics, real-time motor temperature characteristics and individual deviation compensation model of the servo motor, output torque compensation is performed on the servo motor, including: determining the real-time standard output torque compensation based on the real-time load characteristics and real-time motor temperature characteristics of the servo motor through the torque compensation model corresponding to the similar sample servo motor; determining the real-time individual deviation compensation based on the real-time load characteristics and real-time motor temperature characteristics of the servo motor through the individual deviation compensation model of the servo motor; and performing output torque compensation on the servo motor based on the real-time standard output torque compensation and the real-time individual deviation compensation.

[0014] The present invention provides a servo motor control system based on multi-dimensional real-time monitoring, which applies the above-mentioned servo motor control method based on multi-dimensional real-time monitoring, including: a sample analysis module, used to determine the vibration monitoring positions and temperature monitoring positions of multiple sample servo motors through finite element analysis; a data acquisition module, used to set multiple vibration sensors and multiple temperature sensors on the servo motor based on the vibration monitoring positions and temperature monitoring positions of multiple sample servo motors; an individual analysis module, used to perform an output torque compensation experiment on the servo motor through multiple vibration sensors and multiple temperature sensors, and determine the individual deviation compensation model of the servo motor; an operation monitoring module, used to determine the real-time load characteristics and real-time motor temperature characteristics of the servo motor based on the operation status data set collected by multiple vibration sensors and multiple temperature sensors during the operation of the servo motor; a torque compensation module, used to perform output torque compensation on the servo motor based on the real-time load characteristics, real-time motor temperature characteristics and individual deviation compensation model of the servo motor.

[0015] Compared with the prior art, the servo motor control method and system based on multi-dimensional real-time monitoring provided by the present invention have at least the following beneficial effects: 1. By determining the vibration and temperature monitoring positions through finite element analysis and setting up multiple sensors, the operating status of the servo motor can be fully and accurately monitored. This multi-dimensional monitoring provides rich real-time data and provides a basis for precise control. By conducting output torque compensation experiments on servo motors and establishing individual deviation compensation models, accurate compensation can be made for the characteristics of each motor. This helps to eliminate individual differences between motors and improve the accuracy and stability of overall control. Due to factors such as manufacturing process, material differences, and use environment, the performance of each servo motor will have certain individual deviations. By establishing an individual deviation compensation model, accurate torque compensation can be made for the characteristics of each servo motor, thereby eliminating the influence of individual deviations on torque control. Based on the real-time load characteristics, real-time motor temperature characteristics, and individual deviation compensation models of the servo motor, output torque compensation is performed on the servo motor to compensate for the deviation between the output torque and the torque command. Reducing the deviation helps to improve the processing quality of the equipment.

[0016] 2. By calculating the correlation coefficient between the initial vibration monitoring position and the load, the vibration monitoring position that is most sensitive to load changes is selected from multiple candidate positions, thereby improving the accuracy and efficiency of monitoring. On this basis, a second round of screening of vibration monitoring positions is achieved by conducting real tests on sample servo motors, further ensuring the effectiveness of the selected vibration monitoring positions.

[0017] 3. By calculating the third correlation coefficient between the initial temperature monitoring position and the load, the temperature change of the temperature monitoring position under different loads can be accurately reflected. This helps to more accurately understand the temperature characteristics of the motor under different working conditions. By comprehensively considering the correlation coefficients between the initial temperature monitoring position and the load and the motor torque deviation, representative temperature monitoring positions can be targetedly screened out. These positions can more accurately reflect the temperature state of the motor and provide strong support for subsequent temperature monitoring and fault diagnosis. On this basis, based on the third correlation coefficient between the candidate temperature monitoring position and the load, the fourth correlation coefficient between the candidate temperature monitoring position and the motor torque deviation, and the fifth correlation coefficient between the load and the motor torque deviation, the sixth correlation coefficient between the candidate temperature monitoring position and the motor torque deviation is calculated. It excludes the influence of the correlation between the load and the motor torque deviation, and can more accurately reflect the correlation between the candidate temperature monitoring position and the motor torque deviation, that is, it can more accurately reflect which positions' temperature changes will affect the motor torque deviation, thereby ensuring the effectiveness of the screened temperature monitoring positions. 4. By predetermining multiple candidate experimental scenarios and calculating the individual deviation difference coefficient based on the test data of similar sample servo motors, the experimental scenarios that have the greatest impact on the performance of the servo motor can be screened out as target experimental scenarios. This avoids redundant experiments on unimportant scenarios, thereby improving experimental efficiency. By calculating the individual deviation difference coefficient of the candidate experimental scenarios, it is possible to identify which load and temperature combinations have the greatest impact on the output torque deviation of the servo motor. By selecting candidate experimental scenarios with large individual deviation difference coefficients as target experimental scenarios, it can be ensured that the data collected in these scenarios are more representative and reliable. These data can more accurately reflect the performance of the servo motor under different working conditions. If the selected experimental scenario is too single or not comprehensive enough, it may cause deviations or errors in the experimental results. The target experimental scenario determined through a meticulous screening process can more comprehensively cover the various working conditions that the servo motor may face, thereby reducing experimental errors. By screening the target experimental scenarios, experiments can be avoided on unnecessary scenarios, thereby reducing the resource consumption required for the experiment (such as time, manpower, equipment, etc.). This helps to reduce experimental costs and improve the economy of the experiment. Focusing limited experimental resources on the most representative experimental scenarios can improve the utilization of experimental resources and ensure the accuracy and reliability of the individual bias compensation model. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein: Figure 1 is a flow chart of a servo motor control method based on multi-dimensional real-time monitoring according to some embodiments of this specification; Figure 2 is a schematic diagram of a process for determining a vibration monitoring position of a sample servo motor according to some embodiments of this specification; Figure 3 It is a module schematic diagram of a servo motor control system based on multi-dimensional real-time monitoring according to some embodiments of this specification. DETAILED DESCRIPTION

[0019] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, this specification can also be applied to other similar scenarios based on these drawings without creative work. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

[0020] Figure 1 is a flow chart of a servo motor control method based on multi-dimensional real-time monitoring according to some embodiments of this specification, such as Figure 1 As shown, the servo motor control method based on multi-dimensional real-time monitoring may include the following steps.

[0021] S101. Determine vibration monitoring positions and temperature monitoring positions of multiple sample servo motors through finite element analysis.

[0022] In some embodiments, S101 specifically includes: For each sample servo motor, determine the vibration monitoring position of the sample servo motor through finite element analysis; For each sample servo motor, multiple initial temperature monitoring positions of the sample servo motor are determined through finite element analysis. Based on the multiple initial temperature monitoring positions of the sample servo motor, multiple temperature sensors are installed on the sample servo motor, and the sample servo motor is tested to obtain experimental temperature data and motor torque deviation data of multiple initial temperature monitoring positions under different loads, and the temperature monitoring positions are screened from the multiple initial temperature monitoring positions.

[0023] It is understandable that through finite element analysis, it is possible to avoid installing sensors in unnecessary locations, thereby reducing the number of sensors. This not only reduces hardware costs, but also simplifies the complexity of the monitoring system. The optimized monitoring location can more directly reflect the load and temperature conditions of the motor, thereby reducing the collection and processing of redundant data. This helps to improve monitoring efficiency and reduce data processing costs.

[0024] Figure 2 is a schematic diagram of a process for determining a vibration monitoring position of a sample servo motor according to some embodiments of this specification, such as Figure 2 As shown, in some embodiments, the vibration monitoring position of the sample servo motor is determined by finite element analysis, including: Construct a geometric model corresponding to the sample servo motor. Specifically, when constructing the geometric model of the sample servo motor, it is necessary to accurately model the sample servo motor based on its actual size and structural characteristics. For example, the main components of the sample servo motor, such as the stator, rotor, bearing, housing, etc., are determined, and detailed size information of these components is collected. The geometric models of the various components of the sample servo motor are gradually constructed based on the collected size information using CAD (computer-aided design) software or professional finite element analysis software (such as ANSYS) modeling tools. The geometric models of the various components are assembled to form a complete geometric model corresponding to the sample servo motor. Mesh the geometric model corresponding to the sample servo motor to generate a finite element model corresponding to the sample servo motor. Specifically, set mesh parameters, including the size, shape, density and other parameters of the mesh. The setting of these parameters will directly affect the quality of the mesh and the accuracy of the finite element analysis. Use the meshing tool of the finite element analysis software to mesh the geometric model of the servo motor according to the set parameters. After the meshing is completed, a finite element model containing a large number of nodes and units will be generated. Determine a plurality of initial vibration monitoring positions on the finite element model corresponding to the sample servo motor, for example, evenly set a plurality of initial vibration monitoring positions on the finite element model corresponding to the sample servo motor according to a preset spacing; Acquire simulated vibration data of multiple initial vibration monitoring positions of the finite element model corresponding to the sample servo motor under different loads. Specifically, set the load conditions, set different load sizes according to the actual working conditions of the sample servo motor, use finite element analysis software to simulate and analyze the vibration state of the sample servo motor under different load sizes, and after the simulation analysis is completed, extract the vibration data of the initial vibration monitoring position under different loads from the finite element analysis results. These data should include key vibration characteristics such as vibration amplitude, vibration frequency, and vibration direction; For each initial vibration monitoring position, based on the simulated vibration data of the initial vibration monitoring position under different loads, a first correlation coefficient between the initial vibration monitoring position and the load is calculated. Specifically, for each key vibration feature, the Pearson correlation coefficient between the key vibration feature and the load can be calculated based on the characteristic value of the key vibration feature of the initial vibration monitoring position under different loads according to the Pearson correlation calculation formula, and the Pearson correlation coefficient of each key vibration feature and the load is weighted and summed to obtain the first correlation coefficient between the initial vibration monitoring position and the load; Based on a first correlation coefficient between each initial vibration monitoring position and the load, a plurality of candidate vibration monitoring positions are screened from the plurality of initial vibration monitoring positions. For example, the initial vibration monitoring positions whose first correlation coefficients are greater than a first correlation coefficient threshold value may be selected as candidate vibration monitoring positions. Testing the sample servo motor to obtain experimental vibration data of multiple candidate vibration monitoring positions under different loads. Specifically, applying loads of different sizes to the sample servo motor to obtain experimental vibration data of multiple candidate vibration monitoring positions under different loads. For each candidate vibration monitoring position, based on the experimental vibration data of the candidate vibration monitoring position under different loads, a second correlation coefficient between the candidate vibration monitoring position and the load is calculated, wherein the calculation method of the second correlation coefficient is consistent with the calculation method of the first correlation coefficient, which will not be repeated here; Based on the second correlation coefficient between each candidate vibration monitoring position and the load, a vibration monitoring position is screened from multiple candidate vibration monitoring positions. For example, a candidate vibration monitoring position whose second correlation coefficient is greater than a second correlation coefficient threshold may be selected as a vibration monitoring position.

[0025] It can be understood that by calculating the correlation coefficient between the initial vibration monitoring position and the load, the vibration monitoring position that is most sensitive to load changes is selected from multiple candidate positions, thereby improving the accuracy and efficiency of monitoring. On this basis, by conducting real tests on sample servo motors, a second round of screening of vibration monitoring positions is achieved, further ensuring the effectiveness of the selected vibration monitoring positions.

[0026] In some embodiments, determining a plurality of initial temperature monitoring positions of a sample servo motor through finite element analysis specifically includes: In the finite element model corresponding to the sample servo motor, thermal boundary conditions and heat sources are set, wherein the thermal boundary conditions include the external ambient temperature and heat dissipation conditions of the motor, and the heat sources include the heat generated by the resistance loss and core loss inside the motor. Use finite element analysis software to perform thermal analysis on the servo motor. During the thermal analysis process, the finite element analysis software will calculate the temperature distribution and temperature gradient of each finite element; Based on the temperature distribution and temperature gradient of each finite element, the areas with higher temperature and potential hot spots inside the sample servo motor are determined, and according to the results of finite element thermal analysis, the areas with larger temperature changes are selected as the initial temperature monitoring positions.

[0027] As you can understand, finite element analysis can simulate the heat conduction and heat dissipation process inside the motor, so as to accurately calculate the temperature distribution and temperature gradient of each finite element. This helps to accurately identify the areas with higher temperatures and potential hot spots inside the motor, and provides a scientific basis for the selection of temperature monitoring locations. Based on the results of finite element analysis, areas with large temperature changes can be selected as the initial temperature monitoring locations. This targeted layout can ensure the effectiveness of temperature monitoring.

[0028] In some embodiments, selecting a temperature monitoring location from a plurality of initial temperature monitoring locations includes: For each initial temperature monitoring position, based on the experimental temperature data of the initial temperature monitoring position under different loads, the third correlation coefficient between the initial temperature monitoring position and the load is calculated; based on the experimental temperature data of the initial temperature monitoring position under different loads and the motor torque deviation data, the fourth correlation coefficient between the initial temperature monitoring position and the motor torque deviation is calculated; specifically, the experimental temperature data may include key temperature characteristics such as temperature mean and temperature variance; the Pearson correlation coefficient between the key temperature characteristics and the load may be calculated based on the characteristic values ​​of the key temperature characteristics of the initial vibration monitoring position under different loads according to the Pearson correlation calculation formula; the Pearson correlation coefficient between each key temperature characteristic and the load is weighted and summed to obtain the third correlation coefficient between the initial temperature monitoring position and the load; the motor torque deviation data is the deviation between the actual torque and the expected torque of the sample servo motor; the Pearson correlation coefficient between the key temperature characteristics and the motor torque deviation may be calculated based on the characteristic values ​​of the key temperature characteristics of the initial vibration monitoring position under different loads according to the Pearson correlation calculation formula; the Pearson correlation coefficient between each key temperature characteristic and the motor torque deviation is weighted and summed to obtain the fourth correlation coefficient between the initial temperature monitoring position and the motor torque deviation; Based on the third correlation coefficient between each initial temperature monitoring position and the load and the fourth correlation coefficient between each initial temperature monitoring position and the motor torque deviation, multiple candidate temperature monitoring positions are screened from the multiple initial temperature monitoring positions. For example, the third correlation coefficient between the initial temperature monitoring position and the load and the fourth correlation coefficient between each initial temperature monitoring position and the motor torque deviation can be weighted and summed to obtain a comprehensive correlation coefficient of the initial temperature monitoring position, and the initial temperature monitoring position whose comprehensive correlation coefficient is greater than a comprehensive correlation coefficient threshold is used as a candidate temperature monitoring position; Based on the motor torque deviation data under different loads, a fifth correlation coefficient between the load and the motor torque deviation is calculated. For example, the fifth correlation coefficient between the load and the motor torque deviation can be calculated based on the motor torque deviation data under different loads according to a Pearson correlation calculation formula; For each candidate temperature monitoring position, based on the third correlation coefficient between each initial temperature monitoring position and the load, the fourth correlation coefficient between each initial temperature monitoring position and the motor torque deviation, and the fifth correlation coefficient between the load and the motor torque deviation, calculate a sixth correlation coefficient between the candidate temperature monitoring position and the motor torque deviation; Based on the sixth correlation coefficient between each candidate temperature monitoring position and the motor torque deviation, the temperature monitoring position is screened from multiple candidate temperature monitoring positions. For example, the candidate temperature monitoring position whose sixth correlation coefficient is greater than the sixth correlation coefficient threshold can be used as the temperature monitoring position.

[0029] For example, the sixth correlation coefficient between the candidate temperature monitoring position and the motor torque deviation can be calculated according to the following formula: , in, is the sixth correlation coefficient between the ith candidate temperature monitoring position and the motor torque deviation, is the fourth correlation coefficient between the initial temperature monitoring position corresponding to the i-th candidate temperature monitoring position and the motor torque deviation, is the third correlation coefficient between the initial temperature monitoring position corresponding to the i-th candidate temperature monitoring position and the load, It is the fifth correlation coefficient of the load and motor torque deviation.

[0030] It can be understood that by calculating the third correlation coefficient between the initial temperature monitoring position and the load, the temperature change of the temperature monitoring position under different loads can be accurately reflected. This helps to more accurately understand the temperature characteristics of the motor under different working conditions. By comprehensively considering the correlation coefficients between the initial temperature monitoring position and the load and the motor torque deviation, representative temperature monitoring positions can be targetedly screened out. These positions can more accurately reflect the temperature state of the motor and provide strong support for subsequent temperature monitoring and fault diagnosis. On this basis, based on the third correlation coefficient between the candidate temperature monitoring position and the load, the fourth correlation coefficient between the candidate temperature monitoring position and the motor torque deviation, and the fifth correlation coefficient between the load and the motor torque deviation, the calculated sixth correlation coefficient between the candidate temperature monitoring position and the motor torque deviation excludes the influence of the correlation between the load and the motor torque deviation, and can more accurately reflect the correlation between the candidate temperature monitoring position and the motor torque deviation, that is, it can more accurately reflect which positions The temperature changes will affect the motor torque deviation, thereby ensuring the effectiveness of the screened temperature monitoring position.

[0031] S102 : Based on the vibration monitoring positions and temperature monitoring positions of the plurality of sample servo motors, a plurality of vibration sensors and a plurality of temperature sensors are arranged on the servo motor.

[0032] In some embodiments, S102 specifically includes: Get the structural information of the servo motor; Determine similar sample servo motors from a plurality of sample servo motors based on structural information of the servo motor; Based on the vibration monitoring positions and temperature monitoring positions of similar sample servo motors, multiple vibration sensors and multiple temperature sensors are set on the servo motor.

[0033] Specifically, clarify which structural information will be used as the similarity comparison dimension of the motor structure. The structure of the servo motor is complex and contains multiple components, such as stator, rotor, winding, encoder, bearing, etc. In order to simplify the comparison process, the following key structural information can be selected as the comparison dimension: Stator structure: including the material of the stator core, lamination method, winding arrangement, etc.

[0034] Rotor structure: including rotor material, pole arrangement, rotor resistance, etc.

[0035] Winding type: such as single-phase winding, multi-phase winding, and winding connection method, etc.

[0036] Encoder type: The encoder is used to provide real-time feedback of the motor's position and speed information. Its type, accuracy, and installation method will affect the motor's control performance.

[0037] Bearing type: The type and performance of the bearing will affect the motor's rotational inertia, noise, life, etc.

[0038] For each comparison dimension, develop quantitative standards or indicators to facilitate numerical comparison. For example: Stator structure: can be quantified as the thickness of the stator core, stacking coefficient, number of winding turns, etc.

[0039] Rotor structure: can be quantified as the density of the rotor material, the number of magnetic poles, the rotor resistance value, etc.

[0040] Winding type: It can be classified according to the number of phases and connection method of the winding, and assigned corresponding numerical labels.

[0041] Bearing type: Bearings can be classified according to their type (such as rolling bearings, sliding bearings), size and performance, and assigned corresponding numerical identification.

[0042] After quantifying the comparison dimensions, the similarity scores of the servo motor and the sample servo motor in each comparison dimension can be weighted and summed to obtain the structural similarity between the servo motor and the sample servo motor. The sample servo motor with a structural similarity greater than a structural similarity threshold and the largest structural similarity can be taken as a similar sample servo motor.

[0043] It can be understood that by obtaining the structural information of the servo motor and determining similar sample servo motors based on this information, vibration sensors and temperature sensors can be set more specifically. This setting method can ensure that the sensors are placed at the locations where the vibration and temperature changes inside the motor are most significant, thereby improving the accuracy of monitoring. Targeted sensor settings can reduce false alarms and missed alarms. Because the sensors are placed in key positions, they can capture changes in load and temperature more promptly, thereby avoiding monitoring errors caused by improper sensor placement. By determining similar sample servo motors, their successful sensor setting experience can be used as a reference to avoid setting sensors in unnecessary locations. This can not only reduce the number of sensors and reduce monitoring costs, but also simplify the structure of the monitoring system and improve monitoring efficiency.

[0044] S103, performing an output torque compensation experiment on the servo motor through multiple vibration sensors and multiple temperature sensors, and establishing an individual deviation compensation model of the servo motor.

[0045] In some embodiments, S103 specifically includes: For each sample servo motor, a torque compensation model corresponding to the sample servo motor is established and trained, wherein the torque compensation model may be a convolutional neural network model; Determine a plurality of target experimental scenarios, wherein the target experimental scenarios include a target load and a target temperature; For each target experimental scenario, a torque compensation model corresponding to a similar sample servo motor is used to determine a standard output torque compensation based on a target load and a target temperature, and the actual output torque of the servo motor is obtained based on the standard output torque compensation. Based on the actual output torque of the servo motor, an output torque deviation of the servo motor corresponding to the target experimental scenario is determined; Based on the output torque deviation of the servo motor corresponding to each target experimental scenario, an individual deviation compensation model of the servo motor is established, wherein the individual deviation compensation model can be a convolutional neural network model.

[0046] It is understandable that each servo motor will have certain individual deviations in performance due to factors such as manufacturing process, material differences, and operating environment. By establishing an individual deviation compensation model, accurate torque compensation can be performed based on the characteristics of each servo motor, thereby eliminating the impact of individual deviations on torque control.

[0047] In some embodiments, multiple target experimental scenarios are determined, including: determining a plurality of candidate experimental scenarios, wherein the candidate experimental scenarios include candidate loads and candidate temperatures; Obtaining output torque deviations of multiple test servo motors similar to the sample servo motors in multiple candidate experimental scenarios, wherein the structure of the test servo motor is consistent with the structure of the similar sample servo motor; For each candidate experimental scenario, based on the output torque deviations of multiple test servo motors of similar sample servo motors in the candidate experimental scenario, the individual deviation difference coefficient corresponding to the candidate experimental scenario is calculated. For example, the variance of the output torque deviations of multiple test servo motors of similar sample servo motors in the candidate experimental scenario can be calculated as the individual deviation difference coefficient corresponding to the candidate experimental scenario. Based on the individual deviation difference coefficient corresponding to each candidate experimental scenario, a target experimental scenario is screened from multiple candidate experimental scenarios. For example, a candidate experimental scenario whose individual deviation difference coefficient is greater than an individual deviation difference coefficient threshold can be used as a target experimental scenario.

[0048] It can be understood that by predetermining multiple candidate experimental scenarios and calculating the individual deviation difference coefficient based on the test data of similar sample servo motors, those experimental scenarios that have the greatest impact on the performance of the servo motor can be screened out as target experimental scenarios. This avoids redundant experiments on unimportant scenarios, thereby improving experimental efficiency. By calculating the individual deviation difference coefficient of the candidate experimental scenarios, it is possible to identify which load and temperature combinations have the greatest impact on the output torque deviation of the servo motor. By selecting candidate experimental scenarios with large individual deviation difference coefficients as target experimental scenarios, it can be ensured that the data collected in these scenarios are more representative and reliable. These data can more accurately reflect the performance of the servo motor under different working conditions. If the selected experimental scenario is too single or not comprehensive enough, it may cause deviations or errors in the experimental results. The target experimental scenario determined through a meticulous screening process can more comprehensively cover the various working conditions that the servo motor may face, thereby reducing experimental errors. By screening the target experimental scenarios, experiments can be avoided on unnecessary scenarios, thereby reducing the resource consumption required for the experiment (such as time, manpower, equipment, etc.). This helps to reduce experimental costs and improve the economy of the experiment. Focusing limited experimental resources on the most representative experimental scenarios can improve the utilization of experimental resources and ensure the accuracy and reliability of the individual bias compensation model.

[0049] S104 . During the operation of the servo motor, determine the real-time load characteristics and the real-time motor temperature characteristics of the servo motor based on the operation status data sets collected by the multiple vibration sensors and the multiple temperature sensors.

[0050] In some embodiments, S104 specifically includes: Establishing a feature determination model corresponding to each sample servo motor, wherein the feature determination model may be a long short-term memory network model; A model is determined by using the characteristics corresponding to similar sample servo motors, and based on an operating status data set collected by multiple vibration sensors and multiple temperature sensors, the real-time load characteristics and real-time motor temperature characteristics of the servo motor are determined, wherein the real-time load characteristics of the servo motor may include the real-time load size, and the real-time motor temperature characteristics may include the average temperature of the servo motor, temperature fluctuation characteristics, etc.

[0051] S105 , based on the real-time load characteristics of the servo motor, the real-time motor temperature characteristics and the individual deviation compensation model, the output torque compensation of the servo motor is performed.

[0052] In some embodiments, S105 specifically includes: By using the torque compensation model corresponding to the similar sample servo motor, based on the real-time load characteristics and real-time motor temperature characteristics of the servo motor, the real-time standard output torque compensation is determined; Determine the real-time individual deviation compensation through the individual deviation compensation model of the servo motor based on the real-time load characteristics and the real-time motor temperature characteristics of the servo motor; Based on real-time standard output torque compensation and real-time individual deviation compensation, the output torque of the servo motor is compensated.

[0053] Specifically, the current of the servo motor can be adjusted by the servo motor controller according to the calculated real-time standard output torque compensation and real-time individual deviation compensation. The controller will calculate the target current value according to the real-time standard output torque compensation and real-time individual deviation compensation, and adjust the input current of the servo motor to approach or reach the target current value. The current adjustment can be achieved by changing the duty cycle of the voltage or PWM (pulse width modulation) signal.

[0054] Figure 3 is a module schematic diagram of a servo motor control system based on multi-dimensional real-time monitoring according to some embodiments of this specification, such as Figure 3 As shown, the servo motor control system based on multi-dimensional real-time monitoring may include a sample analysis module, a data acquisition module, an individual analysis module, an operation monitoring module and a torque compensation module.

[0055] A sample analysis module, used for determining vibration monitoring positions and temperature monitoring positions of a plurality of sample servo motors through finite element analysis; A data acquisition module, for setting a plurality of vibration sensors and a plurality of temperature sensors on the servo motor based on the vibration monitoring positions and temperature monitoring positions of a plurality of sample servo motors; An individual analysis module is used to perform an output torque compensation experiment on the servo motor through multiple vibration sensors and multiple temperature sensors to determine an individual deviation compensation model of the servo motor; An operation monitoring module is used to determine the real-time load characteristics and real-time motor temperature characteristics of the servo motor based on the operation status data set collected by multiple vibration sensors and multiple temperature sensors during the operation of the servo motor; The torque compensation module is used to compensate the output torque of the servo motor based on the real-time load characteristics of the servo motor, the real-time motor temperature characteristics and the individual deviation compensation model.

[0056] The servo motor control system based on multi-dimensional real-time monitoring can be used to perform Figure 1 The servo motor control method based on multi-dimensional real-time monitoring shown will not be repeated here.

[0057] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A servo motor control method based on multi-dimensional real-time monitoring, characterized in that: include: Through finite element analysis, the vibration monitoring positions and temperature monitoring positions of multiple sample servo motors are determined; Based on the vibration monitoring positions and temperature monitoring positions of the plurality of sample servo motors, a plurality of vibration sensors and a plurality of temperature sensors are arranged on the servo motor; Through multiple vibration sensors and multiple temperature sensors, the output torque compensation experiment of the servo motor is carried out to establish the individual deviation compensation model of the servo motor; During the operation of the servo motor, based on the operation status data set collected by multiple vibration sensors and multiple temperature sensors, the real-time load characteristics and real-time motor temperature characteristics of the servo motor are determined; Based on the real-time load characteristics, real-time motor temperature characteristics and individual deviation compensation model of the servo motor, the output torque of the servo motor is compensated.

2. The servo motor control method based on multi-dimensional real-time monitoring according to claim 1 is characterized in that: Through finite element analysis, the vibration monitoring positions and temperature monitoring positions of multiple sample servo motors are determined, including: For each sample servo motor, determine the vibration monitoring position of the sample servo motor through finite element analysis; For each sample servo motor, multiple initial temperature monitoring positions of the sample servo motor are determined through finite element analysis. Based on the multiple initial temperature monitoring positions of the sample servo motor, multiple temperature sensors are installed on the sample servo motor, and the sample servo motor is tested to obtain experimental temperature data and motor torque deviation data of multiple initial temperature monitoring positions under different loads, and the temperature monitoring positions are screened from the multiple initial temperature monitoring positions.

3. The servo motor control method based on multi-dimensional real-time monitoring according to claim 2 is characterized in that: Through finite element analysis, the vibration monitoring positions of the sample servo motor are determined, including: Construct the geometric model corresponding to the sample servo motor; Meshing the geometric model corresponding to the sample servo motor to generate a finite element model corresponding to the sample servo motor; determining a plurality of initial vibration monitoring positions on a finite element model corresponding to a sample servo motor; Acquire simulated vibration data of multiple initial vibration monitoring positions of a finite element model corresponding to a sample servo motor under different loads; For each initial vibration monitoring position, calculating a first correlation coefficient between the initial vibration monitoring position and the load based on simulated vibration data of the initial vibration monitoring position under different loads; screening a plurality of candidate vibration monitoring positions from a plurality of initial vibration monitoring positions based on a first correlation coefficient between each initial vibration monitoring position and the load; Test the sample servo motor to obtain experimental vibration data of multiple candidate vibration monitoring positions under different loads; For each candidate vibration monitoring position, based on experimental vibration data of the candidate vibration monitoring position under different loads, a second correlation coefficient between the candidate vibration monitoring position and the load is calculated; Based on the second correlation coefficient between each candidate vibration monitoring position and the load, a vibration monitoring position is screened from the plurality of candidate vibration monitoring positions.

4. The servo motor control method based on multi-dimensional real-time monitoring according to claim 3 is characterized in that: Select temperature monitoring locations from multiple initial temperature monitoring locations, including: For each initial temperature monitoring position, based on the experimental temperature data of the initial temperature monitoring position under different loads, a third correlation coefficient between the initial temperature monitoring position and the load is calculated; based on the experimental temperature data of the initial temperature monitoring position under different loads and the motor torque deviation data, a fourth correlation coefficient between the initial temperature monitoring position and the motor torque deviation is calculated; Based on a third correlation coefficient between each initial temperature monitoring position and the load and a fourth correlation coefficient between each initial temperature monitoring position and the motor torque deviation, screening a plurality of candidate temperature monitoring positions from the plurality of initial temperature monitoring positions; Calculating a fifth correlation coefficient between the load and the motor torque deviation based on the motor torque deviation data under different loads; For each candidate temperature monitoring position, based on the third correlation coefficient between each initial temperature monitoring position and the load, the fourth correlation coefficient between each initial temperature monitoring position and the motor torque deviation, and the fifth correlation coefficient between the load and the motor torque deviation, calculate a sixth correlation coefficient between the candidate temperature monitoring position and the motor torque deviation; A temperature monitoring position is screened from the plurality of candidate temperature monitoring positions based on a sixth correlation coefficient between each candidate temperature monitoring position and the motor torque deviation.

5. The servo motor control method based on multi-dimensional real-time monitoring according to any one of claims 1 to 4, characterized in that: Based on the vibration monitoring positions and temperature monitoring positions of multiple sample servo motors, multiple vibration sensors and multiple temperature sensors are set on the servo motor, including: Get the structural information of the servo motor; Determine similar sample servo motors from a plurality of sample servo motors based on structural information of the servo motor; Based on the vibration monitoring positions and temperature monitoring positions of similar sample servo motors, multiple vibration sensors and multiple temperature sensors are set on the servo motor.

6. The servo motor control method based on multi-dimensional real-time monitoring according to claim 5 is characterized in that: Through multiple vibration sensors and multiple temperature sensors, the output torque compensation experiment of the servo motor is carried out to establish the individual deviation compensation model of the servo motor, including: For each sample servo motor, a torque compensation model corresponding to the sample servo motor is established and trained; Determining a plurality of target experimental scenarios, wherein the target experimental scenarios include a target load and a target temperature; For each target experimental scenario, a torque compensation model corresponding to a similar sample servo motor is used to determine a standard output torque compensation based on a target load and a target temperature, and the actual output torque of the servo motor is obtained based on the standard output torque compensation. Based on the actual output torque of the servo motor, an output torque deviation of the servo motor corresponding to the target experimental scenario is determined; Based on the output torque deviation of the servo motor corresponding to each target experimental scenario, an individual deviation compensation model of the servo motor is established.

7. The servo motor control method based on multi-dimensional real-time monitoring according to claim 6 is characterized in that: Identify multiple target experiment scenarios, including: Determine a plurality of candidate experimental scenarios, wherein the candidate experimental scenarios include candidate loads and candidate temperatures; Obtaining output torque deviations of multiple test servo motors similar to the sample servo motors in multiple candidate experimental scenarios, wherein the structure of the test servo motor is consistent with the structure of the similar sample servo motor; For each candidate experimental scenario, based on output torque deviations of multiple test servo motors of similar sample servo motors in the candidate experimental scenario, the individual deviation difference coefficient corresponding to the candidate experimental scenario is calculated; Based on the individual deviation difference coefficient corresponding to each candidate experimental scenario, the target experimental scenario is screened from multiple candidate experimental scenarios.

8. The servo motor control method based on multi-dimensional real-time monitoring according to claim 5, characterized in that: Based on the operating status data set collected by multiple vibration sensors and multiple temperature sensors, the real-time load characteristics and real-time motor temperature characteristics of the servo motor are determined, including: Establish a feature determination model corresponding to each sample servo motor; A feature determination model corresponding to similar sample servo motors is used to determine the real-time load characteristics and real-time motor temperature characteristics of the servo motor based on the operating status data set collected by multiple vibration sensors and multiple temperature sensors.

9. The servo motor control method based on multi-dimensional real-time monitoring according to claim 6, characterized in that: Based on the real-time load characteristics, real-time motor temperature characteristics and individual deviation compensation model of the servo motor, the output torque compensation of the servo motor is performed, including: By using the torque compensation model corresponding to the similar sample servo motor, based on the real-time load characteristics and real-time motor temperature characteristics of the servo motor, the real-time standard output torque compensation is determined; Determine the real-time individual deviation compensation through the individual deviation compensation model of the servo motor based on the real-time load characteristics and the real-time motor temperature characteristics of the servo motor; Based on real-time standard output torque compensation and real-time individual deviation compensation, the output torque of the servo motor is compensated.

10. A servo motor control system based on multi-dimensional real-time monitoring, characterized in that: The servo motor control method based on multi-dimensional real-time monitoring according to any one of claims 1 to 9 comprises: A sample analysis module, used for determining vibration monitoring positions and temperature monitoring positions of a plurality of sample servo motors through finite element analysis; A data acquisition module, for setting a plurality of vibration sensors and a plurality of temperature sensors on the servo motor based on the vibration monitoring positions and temperature monitoring positions of a plurality of sample servo motors; An individual analysis module is used to perform an output torque compensation experiment on the servo motor through multiple vibration sensors and multiple temperature sensors to determine an individual deviation compensation model of the servo motor; An operation monitoring module is used to determine the real-time load characteristics and real-time motor temperature characteristics of the servo motor based on the operation status data set collected by multiple vibration sensors and multiple temperature sensors during the operation of the servo motor; The torque compensation module is used to compensate the output torque of the servo motor based on the real-time load characteristics of the servo motor, the real-time motor temperature characteristics and the individual deviation compensation model.

Images