Servo motor temperature correction coefficient optimization and multi-stage early warning system and method

Abstract

The application provides a servo motor temperature correction coefficient optimization and multi-stage early warning system and method. It includes: a data acquisition module for obtaining a first temperature value and a second temperature value; a temperature correction module for calculating a temperature difference value according to a third temperature value and the second temperature value; a correction coefficient optimization module for online optimization of the temperature correction coefficient in the temperature correction model, and iterative updating to obtain a target temperature correction coefficient; an early warning determination module for dividing the current working state of the servo motor into one of a normal state and at least three early warning states, and generating early warning level information corresponding to the early warning state; an early warning output module for outputting corresponding early warning signals and control instructions to the upper control device and / or human-machine interface according to the early warning level information, so as to implement hierarchical early warning control on the operation of the servo motor. The application can improve the temperature correction precision, realize dynamic optimization of the correction coefficient, and improve the servo motor temperature early warning precision.

Description

Technical Field

[0001] This application relates to the fields of automation and intelligent manufacturing technology, and in particular to a servo motor temperature correction coefficient optimization and multi-level early warning system and method. Background Technology

[0002] In the fields of automation and intelligent manufacturing, servo control systems are widely used in CNC machine tools, industrial robots, and automated production lines. As a key actuator, the temperature of the servo motor directly affects the stability of position, speed, and torque control, as well as the reliable operation of the equipment. Because servo motors operate under complex conditions such as variable loads and high-frequency start-stop cycles, accumulated heat can easily accumulate in the windings and housing. Therefore, real-time monitoring and reasonable assessment of the servo motor's temperature during operation are crucial for preventing overheating damage and developing protection strategies.

[0003] Existing servo motor temperature monitoring largely relies on temperature sensors deployed on or near the motor surface, combined with simple temperature correction models to estimate the actual motor temperature. These temperature correction models typically use fixed correction coefficients, and the model parameters are usually set once during the design or calibration phase, remaining largely unchanged during operation regardless of ambient temperature, installation conditions, or load variations. Furthermore, existing early warning mechanisms are mostly based on a single or a few fixed thresholds, providing simple alarms or shutdowns when temperatures exceed limits, lacking coordination with temperature assessment models.

[0004] Under the above technical solutions, the existing technology has at least the following problems: First, the temperature correction model uses a fixed correction coefficient, which makes it difficult to reflect changes in ambient temperature, individual differences in motors, and the degree of aging in a timely manner, resulting in insufficient temperature estimation accuracy; Second, there is a lack of a dynamic optimization mechanism for the temperature correction coefficient, making it difficult to continuously correct model parameters using long-term collected operating data; Third, the early warning strategies are mostly single-level or coarse-level, and cannot implement refined, multi-level early warning control based on the size of the temperature difference and the trend of thermal state changes, making it difficult to balance motor safety and operating efficiency. Summary of the Invention

[0005] In view of this, embodiments of this application provide a servo motor temperature correction coefficient optimization and multi-level early warning system and method to solve the problems of insufficient temperature correction accuracy, lack of dynamic optimization of correction coefficient, and coarse temperature early warning classification in the prior art.

[0006] The first aspect of this application provides a servo motor temperature correction coefficient optimization and multi-level early warning system, including: a data acquisition module, used to acquire in real time a first temperature value of the servo motor surface temperature and a second temperature value of the ambient temperature around the servo motor from a temperature sensor connected to the servo motor; a temperature correction module, used to correct the first temperature value according to a preset temperature correction model under the control of a current temperature correction coefficient to obtain a third temperature value of the actual temperature of the servo motor, and calculate a temperature difference value based on the third temperature value and the second temperature value; and a correction coefficient optimization module, used to optimize the temperature correction coefficient in the temperature correction model online within a preset coefficient value range using a differential evolution algorithm, wherein the correction coefficient optimization module optimizes the temperature correction coefficient according to... The first, second, and third temperature values ​​are used to construct a fitness function. Differential mutation, crossover, and selection operations are performed on the candidate set of temperature correction coefficients to iteratively update the target temperature correction coefficient. The target temperature correction coefficient is then provided to the temperature correction module as the current temperature correction coefficient. The early warning determination module is used to compare the temperature difference value with at least three sequentially increasing preset thresholds to classify the current working state of the servo motor into a normal state or one of at least three early warning states, and to generate early warning level information corresponding to the early warning state. The early warning output module is used to output corresponding early warning signals and control commands to the upper-level control device and / or human-machine interface according to the early warning level information, so as to implement graded early warning control for the operation of the servo motor.

[0007] The second aspect of this application provides a method for optimizing the temperature correction coefficient of a servo motor and providing multi-level early warning based on the system of the first aspect. The method includes: acquiring a first temperature value (the surface temperature of the servo motor) and a second temperature value (the ambient temperature around the servo motor) in real time from a temperature sensor connected to the servo motor; correcting the first temperature value according to a preset temperature correction model under the control of the current temperature correction coefficient to obtain a third temperature value (the actual temperature of the servo motor); and calculating the temperature difference between the third temperature value and the second temperature value; and optimizing the temperature correction coefficient in the temperature correction model online within a preset coefficient range using a differential evolution algorithm, including optimizing the coefficient based on the first temperature value, the second temperature value, and the ambient temperature. The fitness function is constructed using the temperature value and the third temperature value. Differential mutation, crossover, and selection operations are performed on the candidate set of temperature correction coefficients to iteratively update the target temperature correction coefficient. The target temperature correction coefficient is then fed back to the temperature correction module as the current temperature correction coefficient used in subsequent corrections. Based on the temperature difference value, it is compared with at least three sequentially increasing preset thresholds to classify the current working state of the servo motor into a normal state and at least one of three warning states, and warning level information corresponding to the warning state is generated. Based on the warning level information, corresponding warning signals and control commands are output to the upper-level control device and / or human-machine interface, and graded warning control is implemented for the operation of the servo motor.

[0008] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects:

[0009] The data acquisition module is used to acquire, in real time, the first temperature value of the servo motor surface temperature and the second temperature value of the ambient temperature surrounding the servo motor from a temperature sensor connected to the servo motor. The temperature correction module is used to correct the first temperature value according to a preset temperature correction model and under the control of the current temperature correction coefficient to obtain a third temperature value representing the actual temperature of the servo motor, and calculates the temperature difference between the third and second temperature values. The correction coefficient optimization module is used to optimize the temperature correction coefficient in the temperature correction model online within a preset coefficient value range using a differential evolution algorithm. The correction coefficient optimization module constructs a model based on the first, second, and third temperature values. The fitness function performs differential mutation, crossover, and selection operations on the candidate set of temperature correction coefficients, iteratively updating to obtain the target temperature correction coefficient, and providing the target temperature correction coefficient to the temperature correction module as the current temperature correction coefficient. The early warning determination module compares the temperature difference value with at least three sequentially increasing preset thresholds to classify the current operating state of the servo motor into a normal state or one of at least three early warning states, and generates early warning level information corresponding to the early warning state. The early warning output module outputs corresponding early warning signals and control commands to the upper-level control device and / or human-machine interface based on the early warning level information, thereby implementing graded early warning control for the operation of the servo motor. This application can improve temperature correction accuracy, achieve dynamic optimization of the correction coefficient, and enhance the temperature early warning accuracy of the servo motor. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A schematic diagram of the structural composition of the servo motor temperature correction coefficient optimization and multi-level early warning system provided in the application embodiment;

[0012] Figure 2 This is a flowchart illustrating the servo motor temperature correction coefficient optimization and multi-level early warning method provided in the embodiments of this application. Detailed Implementation

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0014] In today's field of automation and intelligence, servo control systems, as the core system for achieving high-precision position, speed, and torque control, are widely used in numerous production scenarios such as CNC machine tools, robots, and automated production lines. As the core actuator in this system, the stable operation of the servo motor directly affects the efficiency of the entire production process, product quality, and equipment safety. However, in actual operating environments, servo motors face many complex challenges, among which motor overheating is particularly prominent.

[0015] Existing temperature monitoring systems typically rely on fixed temperature correction models, which cannot adapt to environmental changes in real time, leading to inaccurate early warnings. Furthermore, most current research treats temperature correction and early warning mechanisms separately, lacking intelligent optimization algorithms for coordinated optimization of both. Therefore, an intelligent temperature monitoring and early warning system capable of processing data in real time and dynamically adjusting the temperature correction coefficient is needed.

[0016] Existing servo motor temperature monitoring technologies mainly monitor motor temperature through fixed temperature correction models, but these methods have the following problems:

[0017] (1) Fixed correction model: Existing temperature correction models are usually based on fixed correction coefficients, which cannot adapt to environmental changes and result in inaccurate temperature correction.

[0018] (2) Lack of dynamic optimization: The existing mechanism for dynamic optimization of the correction coefficient is lacking.

[0019] (3) Lack of multi-level early warning: Existing early warning systems are usually based on single-level early warning and cannot provide multi-level early warning based on real-time data.

[0020] This application addresses the problems of inaccurate temperature estimation and coarse early warning for servo motors operating under complex conditions by proposing a servo motor temperature monitoring system and method that integrates temperature correction coefficient optimization and multi-level early warning. In some examples, the servo motor surface temperature and ambient temperature are acquired in real time through a data acquisition module. A temperature correction model incorporating temperature correction coefficients is introduced in the temperature correction module to calculate the actual temperature and temperature difference. In the correction coefficient optimization module, a fitness function based on the error between the actual temperature and the reference temperature is constructed using a differential evolution algorithm within a preset coefficient value range. Differential mutation, crossover, and selection operations are performed on the candidate set of temperature correction coefficients to obtain the target temperature correction coefficient online and feed it back to the temperature correction module. In the early warning determination module, the state is determined based on the temperature difference and multi-level thresholds, and corresponding graded early warning signals and control commands are output to the upper-level control device and human-machine interface through the early warning output module.

[0021] Through the above technical solution, this application can dynamically optimize the temperature correction coefficient considering ambient temperature, individual differences in motors, and changes in operating conditions, thereby improving the accuracy of servo motor actual temperature estimation. By setting multi-level early warning thresholds based on the corrected temperature difference and triggering corresponding control commands, it can achieve refined hierarchical management and automatic intervention of the servo motor's thermal state, thereby improving the accuracy and sensitivity of temperature early warning, reducing the risk of overheating damage, and taking into account the continuity and safety of equipment operation.

[0022] The following will describe in detail the specific composition and function of the servo motor temperature correction coefficient optimization and multi-level early warning system provided in this application, with reference to the accompanying drawings and specific embodiments. Figure 1 A schematic diagram of the structural composition of the servo motor temperature correction coefficient optimization and multi-level early warning system provided in the application embodiment is shown below. Figure 1 As shown, the system may specifically include the following components:

[0023] The data acquisition module 101 is used to acquire, in real time, a first temperature value of the surface temperature of the servo motor and a second temperature value of the ambient temperature around the servo motor from a temperature sensor connected to the servo motor.

[0024] The temperature correction module 102 is used to correct the first temperature value according to the preset temperature correction model under the control of the current temperature correction coefficient to obtain the third temperature value of the actual temperature of the servo motor, and calculate the temperature difference value based on the third temperature value and the second temperature value.

[0025] The correction coefficient optimization module 103 is used to optimize the temperature correction coefficient in the temperature correction model online using the differential evolution algorithm within the preset coefficient value range. The correction coefficient optimization module constructs a fitness function based on the first temperature value, the second temperature value and the third temperature value, performs differential mutation, crossover and selection operations on the candidate set of temperature correction coefficients, iteratively updates the target temperature correction coefficient, and provides the target temperature correction coefficient to the temperature correction module as the current temperature correction coefficient.

[0026] The early warning determination module 104 is used to compare the temperature difference value with at least three sequentially increasing preset thresholds, classify the current working state of the servo motor into one of the normal state and at least three levels of early warning state, and generate early warning level information corresponding to the early warning state.

[0027] The early warning output module 105 is used to output corresponding early warning signals and control commands to the upper control device and / or human-machine interface according to the early warning level information, so as to implement graded early warning control for the operation of the servo motor.

[0028] In some embodiments, the data acquisition module includes a surface temperature sensor disposed on the housing of the servo motor and an ambient temperature sensor disposed in the space surrounding the servo motor. The data acquisition module acquires a first temperature value and a second temperature value from the surface temperature sensor and the ambient temperature sensor, respectively, and adds a time stamp to each temperature value.

[0029] Specifically, the data acquisition module includes a surface temperature sensor mounted on the servo motor housing and an ambient temperature sensor positioned in the space surrounding the servo motor. The surface temperature sensor is fixedly installed on the servo motor housing near the stator windings, using a thermally conductive mounting structure to ensure close contact with the motor housing and reflect changes in the motor surface temperature. The ambient temperature sensor is installed inside a cabinet near the servo motor or on a bracket near the motor, maintaining a preset distance from the servo motor housing to avoid direct exposure to radiant heat from the motor housing, thus characterizing the temperature of the air environment surrounding the servo motor. Both the surface temperature sensor and the ambient temperature sensor are connected to the data acquisition module via independent signal acquisition channels. The signal conditioning unit filters, amplifies, and isolates the sensor output signals before inputting them to the analog-to-digital converter.

[0030] In the data acquisition process, the acquisition control logic within the AI ​​machine or data acquisition and monitoring system sends sampling commands to the data acquisition module at a preset sampling period. Upon receiving the sampling command, the data acquisition module synchronously reads the output signals from the surface temperature sensor and the ambient temperature sensor at a unified sampling time. After analog-to-digital conversion, it obtains the first temperature value of the servo motor surface and the second temperature value of the ambient temperature surrounding the servo motor, respectively. Simultaneously, the time stamp generation unit generates a corresponding time stamp based on the system clock and binds this time stamp to the currently acquired first and second temperature values, forming a temperature data record containing "time stamp, first temperature value, and second temperature value." The data acquisition module caches or sends the temperature data record in chronological order to the temperature correction module and correction coefficient optimization module within the AI ​​machine. This allows the temperature correction module to calculate the third temperature value and temperature difference value based on the first and second temperature values ​​corresponding to the same sampling time, and allows the correction coefficient optimization module to construct a sample set based on the temperature data arranged by time.

[0031] For example, in a specific application scenario, a servo motor is used to drive the joints of an industrial robot with a rated power of several kilowatts. The data acquisition module sets the sampling period to 100ms. Within each sampling period, the signal output by the surface temperature sensor is converted into a first temperature value at the current moment, for example, a first temperature value of 65℃ at a certain moment. The signal output by the ambient temperature sensor is converted into a second temperature value at that moment, for example, 32℃. Both are appended with the same time stamp and sent together to the AI ​​machine. The temperature correction module calculates a third temperature value and a temperature difference value of the actual temperature of the servo motor based on the first and second temperature values ​​corresponding to the time stamp. The correction coefficient optimization module uses the first and second temperature values ​​corresponding to multiple sampling moments and the calibration reference temperature data to form samples for constructing a fitness function and performing differential evolution optimization.

[0032] Through the above implementation method, the data acquisition module can provide a first temperature value and a second temperature value with a unified time identifier for the surface temperature and the ambient temperature, respectively, while ensuring sampling synchronization. This provides a reliable data foundation for the dynamic optimization of the temperature correction coefficient and the multi-level early warning judgment based on the temperature difference.

[0033] In some embodiments, the data acquisition module is also used to acquire one or more of the operating parameters of the servo motor, such as voltage, current, and speed, and send the operating parameters together with the first temperature value and the second temperature value to the correction coefficient optimization module. When constructing the fitness function, the correction coefficient optimization module uses the operating parameters as auxiliary variables in the calculation.

[0034] Specifically, the data acquisition module includes a voltage acquisition channel for acquiring the line voltage or DC bus voltage at the output of the servo driver, a current acquisition channel for acquiring the phase current or equivalent current of the servo motor, and a speed acquisition channel for acquiring servo motor speed information. The voltage acquisition channel can connect to the analog output terminal or internal detection node provided by the servo driver, and, in conjunction with a voltage sensor and signal conditioning circuit, acquire the voltage of the servo motor during operation and convert it into a digital quantity that can be processed by the AI. The current acquisition channel can obtain the phase current or equivalent current of the servo motor through a current transformer, Hall effect current sensor, or current feedback interface provided by the servo driver, and after filtering, amplification, isolation, and analog-to-digital conversion, form the current parameters for the current sampling period. The speed acquisition channel calculates the current speed of the servo motor based on the number and frequency of pulses output by the encoder, and sends this speed value as a speed operating parameter to the data acquisition module.

[0035] In terms of the data acquisition process, within each preset sampling period, the data acquisition module simultaneously acquires the outputs of the surface temperature sensor and the ambient temperature sensor to obtain the first and second temperature values. Simultaneously, it triggers synchronous sampling of the voltage, current, and speed acquisition channels to obtain the voltage, current, and speed parameter values ​​at the corresponding sampling moments. The time stamp generation unit generates a unified time stamp for this sampling period based on the system clock and appends this time stamp to the first temperature value, the second temperature value, and the voltage, current, and speed parameters, forming an operational data record containing the time stamp, temperature data, and operating parameters. The data acquisition module transmits the chronologically ordered operational data record in real-time via industrial Ethernet or fieldbus to the correction coefficient optimization module running in the AI ​​system, which is used to construct a sample set considering the operating conditions.

[0036] In the correction coefficient optimization module, based on the multi-dimensional operating data records provided by the data acquisition module, the first temperature value, the second temperature value, and the reference temperature data corresponding to the servo motor calibration test are selected as the basic temperature samples according to the time window or operating condition interval. At the same time, the voltage parameters, current parameters, and speed parameters corresponding to these samples are included as auxiliary variables in the sample set. When constructing the fitness function, in addition to calculating the error between the third temperature value obtained by the temperature correction model under each candidate temperature correction coefficient and the reference temperature data, the correction coefficient optimization module can also infer the load level or loss degree of the servo motor based on the voltage and current parameters, and distinguish the heating characteristics under different speed conditions based on the speed parameters. These operating parameters are used as weighting coefficients or classification criteria to weight or group the errors under different operating conditions, thereby forming an error evaluation index that reflects the comprehensive relationship between temperature error and operating conditions. When performing differential mutation, crossover, and selection operations, the differential evolution algorithm uses this error evaluation index as the basis for determining the fitness function. It prioritizes retaining the temperature correction coefficient that performs well and has stronger adaptability under various voltage, current, and speed conditions. After iterative convergence, the target temperature correction coefficient is obtained and fed back to the temperature correction module for subsequent temperature correction.

[0037] Through the above implementation method, the data acquisition module collects and correlates the operating parameters of the servo motor, such as voltage, current, and speed, with the first and second temperature values ​​in a unified manner over time. The correction coefficient optimization module uses the operating parameters as auxiliary variables in the calculation when constructing the fitness function, so that the online optimization of the temperature correction coefficient is not only based on the temperature error itself, but also comprehensively considers the heating characteristics of the servo motor under different load and speed conditions. This makes the optimized temperature correction coefficient more applicable and stable under various operating conditions, further improving the accuracy of the actual temperature estimation of the servo motor and providing more reliable temperature basis data for subsequent multi-level early warning judgment based on temperature difference.

[0038] In some embodiments, the temperature correction module is specifically used for:

[0039] In the temperature correction module, a standard ambient temperature parameter is preset. Based on the difference between the second temperature value and the standard ambient temperature parameter and the current temperature correction coefficient, the correction amount corresponding to the first temperature value is determined. The correction amount and the first temperature value are calculated to obtain the third temperature value. The temperature difference value is determined based on the difference between the third temperature value and the second temperature value.

[0040] Specifically, the temperature correction module is located within the software functional unit of the AI ​​machine and interacts with the aforementioned data acquisition module and correction coefficient optimization module. The data acquisition module provides the temperature correction module with a first temperature value and a second temperature value with a time stamp according to a preset sampling period. The correction coefficient optimization module updates the current temperature correction coefficient online based on historical samples and a differential evolution algorithm. After receiving the first temperature value, the second temperature value, and the current temperature correction coefficient with the corresponding time stamp, the temperature correction module performs temperature correction calculation and temperature difference calculation, and outputs the third temperature value and temperature difference value of the actual temperature of the servo motor to the early warning judgment module and the visualization module.

[0041] The temperature correction module has a preset standard ambient temperature parameter, which can be experimentally determined during the servo motor's factory calibration or on-site commissioning. This parameter typically corresponds to the ambient temperature during a temperature rise test of the servo motor in a standard laboratory environment. After receiving a second temperature value at a certain sampling moment, the temperature correction module first calculates the environmental deviation at that moment. This is achieved by subtracting the standard ambient temperature parameter from the second temperature value. The environmental temperature deviation is then multiplied by the current temperature correction coefficient to obtain the correction amount corresponding to the first temperature value at that sampling moment. The current temperature correction coefficient is obtained by the correction coefficient optimization module through iterative optimization within a preset coefficient range using a differential evolution algorithm. It is used to characterize the degree of influence of ambient temperature deviating from standard operating conditions on the actual temperature of the servo motor.

[0042] Furthermore, after obtaining the correction amount, the temperature correction module performs calculations with the first temperature value. For example, by adding the correction amount to the first temperature value, it obtains a third temperature value representing the actual temperature of the servo motor at that sampling moment. This third temperature value can be understood as the equivalent actual temperature after considering the deviation of the ambient temperature from the standard ambient temperature and the influence of the current temperature correction coefficient, used to more accurately reflect the thermal state of the servo motor under the current operating conditions. Subsequently, the temperature correction module subtracts the second temperature value from the third temperature value to calculate the temperature difference value at that sampling moment. This temperature difference value characterizes the degree to which the actual temperature of the servo motor exceeds the ambient temperature, serving as the direct basis for subsequent multi-level early warning judgments.

[0043] For example, in a specific application scenario, a servo motor is installed on an automated production line, and the standard ambient temperature parameter is set to 25℃. At a certain sampling moment, the data acquisition module provides a first temperature value of 70℃ and a second temperature value of 35℃. The correction coefficient optimization module outputs a current temperature correction coefficient of 0.1 based on the previous stage's operating data. The temperature correction module first calculates the ambient temperature deviation as 35 minus 25, obtaining 10℃. It then multiplies 10 by the current temperature correction coefficient of 0.1 to obtain a correction amount of 1℃. This correction amount is then added to the first temperature value to obtain a third temperature value of 71℃. Subsequently, the temperature correction module subtracts 35 from 71 to obtain a temperature difference value of 36℃. Both the third temperature value of 71℃ and the temperature difference value of 36℃ are sent to the early warning judgment module. The early warning judgment module determines the current early warning state of the servo motor based on a preset temperature difference threshold. Simultaneously, the visualization module updates the temperature curve and early warning status indicator based on the third temperature value and the temperature difference value.

[0044] Through the above implementation method, the temperature correction module, based on the preset standard ambient temperature parameters, uses the difference between the second temperature value and the standard ambient temperature parameters, as well as the current temperature correction coefficient, to determine the correction amount corresponding to the first temperature value. Then, it calculates the third temperature value using the correction amount and the first temperature value, and obtains the temperature difference between the third temperature value and the second temperature value. Thus, under the premise of considering changes in environmental conditions and online optimization of the correction coefficient, the actual temperature of the servo motor is uniformly corrected, providing temperature and temperature difference data that are closer to the actual physical process for multi-level early warning judgment, thereby improving the accuracy of temperature assessment and the reliability of multi-level early warning judgment.

[0045] In some embodiments, the correction coefficient optimization module is specifically used for:

[0046] Multiple sets of temperature correction coefficients are generated within the preset coefficient value range to form a candidate set of temperature correction coefficients. Based on the sample set consisting of the first temperature value, the second temperature value, and the reference temperature data corresponding to the servo motor calibration test, the third temperature value obtained under each temperature correction coefficient is calculated using the temperature correction model. A fitness function is constructed based on the error between the third temperature value and the reference temperature data.

[0047] During the differential evolution iteration process, differential mutation, crossover and selection operations are performed on the candidate set of temperature correction coefficients. Each candidate temperature correction coefficient is updated according to the value of the fitness function until the preset convergence condition is met. The temperature correction coefficient with the best fitness function is selected from the updated candidate set of temperature correction coefficients as the target temperature correction coefficient, and the target temperature correction coefficient is provided to the temperature correction module as the current temperature correction coefficient.

[0048] Specifically, the correction coefficient optimization module pre-sets the lower and upper limits of the temperature correction coefficient, for example, setting the value range to 0.05 to 0.18, and randomly generates multiple sets of temperature correction coefficients within this continuous range, forming a candidate set of temperature correction coefficients. Each sample in the sample set includes a first temperature value, a second temperature value, and reference temperature data corresponding to a certain operating condition, all collected under that condition. For each temperature correction coefficient in the candidate set, the correction coefficient optimization module calls a temperature correction model consistent with the temperature correction module, substitutes the first and second temperature values ​​from the sample, calculates the corresponding third temperature value under that candidate temperature correction coefficient, and compares the third temperature value with the reference temperature data under the same operating condition.

[0049] When constructing the fitness function, the correction coefficient optimization module statistically analyzes the error between the third temperature value corresponding to each operating condition in the sample set and the reference temperature data. For example, it can take the absolute value or square of the temperature error for each operating condition and then sum or average them to obtain an error evaluation index that reflects the overall fitting ability of the candidate temperature correction coefficient. The smaller the error evaluation index, the higher the degree of fit of the temperature correction model to the reference temperature data under the candidate temperature correction coefficient, and the higher the fitness of the candidate temperature correction coefficient. The correction coefficient optimization module uses the error evaluation index as the basis for the value of the fitness function, enabling the differential evolution algorithm to select the best candidates based on this fitness function.

[0050] During the differential evolution iteration process, the correction coefficient optimization module performs differential mutation, crossover, and selection operations on the candidate set of temperature correction coefficients according to the preset population size and maximum number of iterations. During differential mutation, several combinations of temperature correction coefficients are selected from the current candidate set to generate mutated individuals. During crossover, the mutated individuals and the original candidate individuals are replaced with parameters according to a certain probability to obtain experimental individuals. During selection, the experimental individuals and the original candidate individuals are compared in terms of fitness function, and individuals with higher fitness are retained to enter the next generation of candidate sets. As the number of iterations increases, different candidate temperature correction coefficients gradually converge towards a smaller error evaluation index under the drive of the fitness function. The differential evolution process terminates when the preset maximum number of iterations is reached or the fitness improvement is less than a preset threshold for several consecutive generations.

[0051] Furthermore, at the end of the iteration, the correction coefficient optimization module selects the temperature correction coefficient with the optimal fitness function value from the updated candidate set of temperature correction coefficients, and outputs it as the target temperature correction coefficient for the current stage. This target temperature correction coefficient is stored internally by the AI ​​for continued application in subsequent operation phases, and is also provided to the temperature correction module through an internal interface, allowing the temperature correction module to use this target temperature correction coefficient as the current temperature correction coefficient in subsequent sampling periods.

[0052] For example, in a specific example, for a certain type of servo motor, several sets of reference temperature data under different ambient temperatures and different loads are obtained during the calibration phase. The correction coefficient optimization module sets the initial size of the candidate set to 20 and the maximum number of iterations to 100. After the iteration is completed, the target temperature correction coefficient obtained can reduce the average temperature difference error between the corrected third temperature value and the reference temperature data by about 0.065℃ compared with the fixed coefficient before optimization. This target temperature correction coefficient is used by the temperature correction module for real-time temperature correction during the actual operation phase.

[0053] Through the above implementation method, the correction coefficient optimization module generates a candidate set of temperature correction coefficients within a preset coefficient value range. Based on the sample set consisting of the first temperature value, the second temperature value, and the calibration reference temperature data, it constructs a fitness function and iteratively updates each candidate temperature correction coefficient through differential mutation, crossover, and selection. After meeting the convergence condition, it selects the temperature correction coefficient with the optimal fitness function as the target temperature correction coefficient and provides it to the temperature correction module as the current temperature correction coefficient. In this way, the temperature correction coefficient is globally optimized and online adaptively adjusted using calibration data and operating data, so that the temperature correction model has a higher fitting accuracy for the actual temperature of the servo motor under different ambient temperatures and load conditions, providing more accurate and reliable basic parameters for subsequent temperature difference calculation and multi-level early warning judgment.

[0054] In some embodiments, a fitness function is constructed based on the error between the third temperature value and the reference temperature data, including:

[0055] For each group of temperature samples in the sample set, the corresponding third temperature value is calculated using the temperature correction model under each candidate temperature correction coefficient, and then compared with the reference temperature data of the same group to obtain the error value. The error values ​​are accumulated and / or averaged to obtain the error evaluation index, and the error evaluation index is used as the value of the fitness function under the candidate temperature correction coefficient.

[0056] Specifically, before executing the differential evolution algorithm, the correction coefficient optimization module first constructs a sample set based on the calibration test results of the servo motor and the data collected during actual operation. Each temperature sample in the sample set includes a first temperature value, a second temperature value, and reference temperature data corresponding to that operating condition obtained by the data acquisition module at a certain moment or under a certain steady-state condition, obtained through calibration tests. The reference temperature data can be the temperature measured by a dedicated temperature sensor placed at the winding or internal hot spot location of the servo motor, or the target temperature determined through fine thermal simulation and calibration tests, used to characterize the accurate value of the actual temperature of the servo motor under that operating condition. When constructing the sample set, the system preferably selects multiple operating conditions including different ambient temperatures, different loads, and different speeds, so that the sample set can cover the typical operating range of the servo motor.

[0057] For example, in a certain generation of differential evolution iteration, the correction coefficient optimization module iterates through each group of temperature samples in the sample set for each candidate temperature correction coefficient in the candidate set of temperature correction coefficients. For each group of temperature samples, the correction coefficient optimization module calls the temperature correction model consistent with the temperature correction module, substitutes the first temperature value, the second temperature value and the current candidate temperature correction coefficient in the group of samples into the temperature correction model, and calculates the third temperature value corresponding to the group of samples.

[0058] Subsequently, the third temperature value is compared with the reference temperature data in the same group one by one, and the error value between the two is calculated. The error value can be the difference between the third temperature value and the reference temperature data, or the absolute value or square of the difference, to reflect the fitting deviation of the temperature correction model under the current sample working conditions under the candidate temperature correction coefficient.

[0059] After calculating the third temperature value and obtaining the error value for all samples in the sample set, the correction coefficient optimization module accumulates and / or averages the error values ​​corresponding to each sample to obtain the error evaluation index corresponding to the candidate temperature correction coefficient.

[0060] For example, the total error index can be obtained by summing the absolute values ​​of the temperature errors of all samples, or by summing the squares of the temperature errors of each sample and then averaging them to obtain the mean square error index. Alternatively, weights can be introduced among different samples to perform a weighted summation of the error values ​​of samples under different operating conditions, thereby constructing an error assessment index that better reflects the importance of different operating conditions. The correction coefficient optimization module uses the above error assessment index as the value of the fitness function under the candidate temperature correction coefficient, so that when the differential evolution algorithm performs mutation, crossover, and selection operations, the magnitude of the error assessment index can be used as a unified standard to measure the quality of the candidate temperature correction coefficient.

[0061] In some specific examples, the servo motor collected several sets of typical operating condition data during the calibration phase. These included the first temperature value, the second temperature value, and the corresponding reference temperature data when operating at rated torque, under light load, and under heavy load at different ambient temperatures. When constructing the fitness function, the correction coefficient optimization module calculated the third temperature value for each candidate temperature correction coefficient under these typical operating conditions and compared it with the corresponding reference temperature data. The error evaluation index was obtained by summing the squared temperature errors of each operating condition and taking the average.

[0062] In this way, the fitting effect of a single candidate temperature correction coefficient under multiple operating conditions is integrated into an error evaluation index. In subsequent iterations, the differential evolution algorithm prioritizes retaining candidate coefficients with smaller error evaluation indices and gradually eliminates candidate coefficients with larger error evaluation indices, so that the temperature correction coefficient has good adaptability under multiple operating conditions.

[0063] Through the above implementation method, when constructing the fitness function, the correction coefficient optimization module calculates the corresponding third temperature value for each group of temperature samples in the sample set using the temperature correction model under each candidate temperature correction coefficient, and compares it with the reference temperature data one by one to obtain the error value. Then, the error values ​​are accumulated and / or averaged to obtain the error evaluation index, which is used as the value of the fitness function. This allows the differential evolution algorithm to judge the quality of the temperature correction coefficient based on the overall fitting effect under multiple operating conditions. This implementation method can ensure the convergence of the algorithm while making the optimized target temperature correction coefficient have higher robustness and fitting accuracy under different ambient temperatures and load conditions, further improving the accuracy of servo motor actual temperature estimation and providing a reliable parameter basis for subsequent temperature difference calculation and multi-level early warning judgment.

[0064] In some embodiments, the early warning determination module is preset with a first temperature difference threshold, a second temperature difference threshold and a third temperature difference threshold, wherein the first temperature difference threshold is less than the second temperature difference threshold and the second temperature difference threshold is less than the third temperature difference threshold.

[0065] When the temperature difference is less than or equal to the first temperature difference threshold, it is determined to be in a normal state. When the temperature difference is greater than the first temperature difference threshold and less than or equal to the second temperature difference threshold, it is determined to be in a third-level warning state. When the temperature difference is greater than the second temperature difference threshold and less than or equal to the third temperature difference threshold, it is determined to be in a second-level warning state. When the temperature difference is greater than the third temperature difference threshold, it is determined to be in a first-level warning state.

[0066] Specifically, during the system configuration phase, the early warning judgment module is used by maintenance personnel or debugging engineers to analyze the safety margin corresponding to different temperature difference ranges based on the servo motor's rated temperature rise capability, insulation class, and thermal balance test results. Combined with experimental data, the values ​​of the first temperature difference threshold, the second temperature difference threshold, and the third temperature difference threshold are determined.

[0067] For example, in a certain type of servo motor application, the first temperature difference threshold can be set to 30℃, the second temperature difference threshold to 40℃, and the third temperature difference threshold to 50℃, while always satisfying the relationship that the first temperature difference threshold is less than the second temperature difference threshold, and the second temperature difference threshold is less than the third temperature difference threshold. These thresholds are stored in the system parameters in a configurable manner, and can either use the factory default values ​​or be adjusted appropriately on-site according to actual working conditions.

[0068] During operation, the early warning determination module receives temperature difference values ​​output from the temperature correction module according to the sampling period and performs hierarchical comparisons based on preset temperature difference thresholds. For the temperature difference value at each sampling moment, the early warning determination module first determines whether the temperature difference value is less than or equal to the first temperature difference threshold. If this condition is met, the current working state of the servo motor is determined to be normal, and an early warning level information representing the normal state is generated for that sampling moment. If the temperature difference value is greater than the first temperature difference threshold, it continues to determine whether the temperature difference value is less than or equal to the second temperature difference threshold. If this condition is met, the current working state is determined to be a third-level early warning state. If the temperature difference value is greater than the second temperature difference threshold, it further determines whether the temperature difference value is less than or equal to the third temperature difference threshold. If this condition is met, the current working state is determined to be a second-level early warning state. When the temperature difference value is greater than the third temperature difference threshold, the current working state is determined to be a first-level early warning state, and the corresponding highest-level early warning level information is generated. The early warning determination module associates the above early warning level information with a time stamp and sends it to the early warning output module and the visualization module for subsequent hierarchical control and graphical display.

[0069] For example, in a specific application scenario, a servo motor is used to drive a conveyor mechanism on an automated production line. The standard ambient temperature parameter is 25℃, and the preset first temperature difference threshold is 30℃, the second temperature difference threshold is 40℃, and the third temperature difference threshold is 50℃. At a certain sampling moment, the temperature correction module calculates the third temperature value to be 71℃, the second temperature value to be 35℃, and the corresponding temperature difference value to be 36℃. After receiving this temperature difference value, the warning judgment module determines that the temperature difference value is greater than the first temperature difference threshold of 30℃ and less than or equal to the second temperature difference threshold of 40℃. Therefore, it determines the current working state to be a third-level warning state, generates third-level warning level information, and sends it to the warning output module to trigger an alert.

[0070] As the load on the servo motor increases further, the temperature difference gradually increases in subsequent sampling periods. When the temperature difference crosses the second temperature difference threshold of 40℃ but does not exceed the third temperature difference threshold of 50℃, the warning judgment module updates the current state to the second-level warning state. When the temperature difference exceeds 50℃, the current state is immediately judged as the first-level warning state, and the corresponding highest-level warning information is provided to the warning output module.

[0071] Through the above implementation method, the early warning judgment module, based on preset first, second, and third temperature difference thresholds, performs a graded comparison of the temperature difference values ​​output by the temperature correction module, classifying the current operating state of the servo motor into different levels such as normal state and third-level early warning state, second-level early warning state, and first-level early warning state. This achieves multi-level early warning judgment with temperature difference as a unified quantitative indicator. This embodiment enables the system to finely classify thermal risks based on the degree to which the actual temperature of the servo motor exceeds the ambient temperature, providing a clear judgment basis for the subsequent early warning output module to implement different levels of prompts, load reduction, or shutdown control. This improves the resolution and flexibility of temperature early warning while ensuring the safe operation of the servo motor.

[0072] In some embodiments, when the warning output module receives warning level information, it outputs a prompt signal to remind the operator to pay attention to the temperature change of the servo motor in the third warning state; outputs a control command to the servo motor drive device to reduce the load of the servo motor and / or adjust the heat dissipation device in the second warning state; and outputs a shutdown control command to the servo motor drive device to stop the operation of the servo motor in the first warning state.

[0073] Specifically, when the warning level information output by the warning judgment module indicates that the current warning state is at level three, the warning output module does not directly intervene in the operation control of the servo motor. Instead, it focuses on reminding the operator to pay attention to the temperature change trend. At this time, the warning output module sends a level three warning prompt signal to the human-machine interface. For example, it highlights the current temperature curve and temperature difference curve of the servo motor in a prominent color on the monitoring screen and pops up a "Caution: Temperature Too High" information prompt box. This can be accompanied by a slight beeping sound or a flashing icon to prompt the operator to pay close attention to the motor, but without changing the output parameters of the servo driver, so that the system remains in normal production status.

[0074] Furthermore, when the warning level information output by the warning judgment module indicates that the current warning state is at level two, the warning output module considers that the thermal state of the servo motor is approaching the risk range and requires certain automatic intervention measures to reduce the rate of temperature rise. In this state, the warning output module continues to output more obvious warning signals to the human-machine interface, such as marking the relevant servo motor status as "warning," increasing the intensity of audible and visual alarms, and prompting maintenance personnel to check the load and heat dissipation as soon as possible. Simultaneously, it sends control commands to the servo motor drive device to reduce the servo motor load and / or adjust the heat dissipation device.

[0075] For example, the early warning output module can send derating operation commands through the control interface with the drive device to limit the maximum output torque of the servo motor or reduce its operating speed, or send start / stop or speed-up commands to external cooling fans, cooling pumps and other heat dissipation devices to enhance heat dissipation capacity, thereby suppressing further temperature rise without interrupting production.

[0076] Furthermore, when the warning level information output by the warning judgment module indicates that the current state is at the first-level warning state, the warning output module considers this state as a serious overheating risk, requiring priority to ensure the safety of the equipment and system. In this state, the warning output module sends the highest-level warning prompt signal to the human-machine interface, for example, marking the servo motor in a flashing red manner on the interface, accompanied by a continuous audible and visual alarm, and simultaneously generating an emergency warning log in the event log.

[0077] More importantly, the early warning output module sends a stop control command to the servo driver through the control channel between the module and the servo motor drive. This command may include control logic that rapidly decelerates to zero speed and cuts off the servo enable, allowing the servo motor to stop safely during controllable deceleration and preventing winding insulation aging, demagnetization, or other irreversible damage caused by continuous high-temperature operation. In some embodiments, to prevent mechanical shocks caused by frequent start-stops, the early warning output module may also combine the trend or duration of temperature difference changes, issuing a stop control command only after the temperature difference has continuously exceeded a third temperature difference threshold for a preset period of time, thus avoiding false stops caused by instantaneous disturbances.

[0078] For example, in a specific application scenario, a servo motor is used to drive a feeding mechanism on an automated assembly line. The early warning judgment module determines the third-level, second-level, and first-level early warning states based on the temperature difference value. In the third-level early warning state, the early warning output module simply highlights the servo motor in yellow on the upper-level monitoring screen and prompts the operator to pay attention to temperature changes. When the temperature difference value further increases and enters the second-level early warning range, the early warning output module issues a derating operation command to the servo drive device, limiting the maximum output torque of the servo motor to a certain percentage of its rated value and activating an additional cooling fan. If the temperature difference value continues to rise and exceeds the third temperature difference threshold, the early warning output module immediately sends a shutdown control command to the servo drive device, safely stopping the servo motor. Simultaneously, the device is marked with a flashing red indicator on the monitoring interface, prompting the operator to promptly investigate the cause of overheating.

[0079] Through the above implementation method, after receiving the warning level information, the warning output module executes different levels of prompting and control strategies for the third-level, second-level, and first-level warning states: from simply reminding users to pay attention, to automatically reducing load and enhancing heat dissipation, and finally to forced shutdown, forming a hierarchical control link that matches the temperature difference classification judgment. This embodiment enables the system to take progressive intervention measures according to different stages of the servo motor's thermal state, ensuring equipment and system safety while minimizing the impact on production cycle time, and achieving effective coordination between temperature warning and operation control.

[0080] In some embodiments, a visualization module is also included. The visualization module is connected to the temperature correction module and the early warning determination module. It is used to generate a temperature curve, a temperature difference curve and an early warning status identifier based on the third temperature value, the temperature difference value and the early warning level information, and to display the temperature change and early warning status change of the servo motor within a preset time range in graphical or textual form on the display interface.

[0081] Specifically, the visualization module includes a data buffer unit, a curve generation unit, and an interface display unit. The data buffer unit is responsible for archiving the continuously received third temperature value, temperature difference value, and warning level information by time, and storing the third temperature value, temperature difference value, and warning level information corresponding to each sampling period as time series data.

[0082] Based on this, the curve generation unit uses the time series data in the data buffer unit to generate the actual temperature curve of the servo motor with a preset time range as the horizontal axis, such as the most recent 10 minutes, 1 hour, or a custom time period, and the third temperature value as the vertical axis. It also generates a temperature difference curve with the temperature difference value as the vertical axis. In conjunction with the warning level information, different colors or symbols of warning status indicators are superimposed on the curve. For example, a specific symbol is marked above the time point when the third-level warning occurs, and the curve interval is highlighted during the time period when the second-level or first-level warning occurs, so that operators can clearly understand the correspondence between temperature changes and warning events when viewing the curve.

[0083] Furthermore, the interface display unit presents the temperature curve, temperature difference curve, and warning status indicators output by the curve generation unit in a combination of graphics and text on the display interface. Operators can select the target servo motor and a preset time range in the interface to view the actual temperature change trend, temperature difference change trend, and the warning status corresponding to each time point within that time range. Simultaneously, the visualization module can also display the third temperature value, temperature difference value, and corresponding warning level information for key time points in a list format below or to the side of the curve. For example, it can list the occurrence time and temperature difference value of several recent Level 2 or Level 1 warnings. To facilitate analysis by maintenance personnel, the visualization module supports operations such as zooming the time axis, dragging the time window, and viewing historical records, allowing users to review the temperature curve and warning records of the previous shift or day to help determine whether a servo motor is experiencing long-term high-temperature operation or periodic abnormal temperature rises.

[0084] For example, in some specific application scenarios, an automated production line is equipped with multiple servo motors. The visualization module on the AI ​​machine lists the real-time operating status of each servo motor on the main interface. After the operator clicks on a servo motor, the visualization module displays the third temperature value curve and temperature difference value curve of that servo motor in a new interface over the past hour. The curves indicate a third-level warning state in yellow when the temperature difference value crosses the first temperature difference threshold to the second temperature difference threshold; a second-level warning state in orange when the temperature difference value crosses the second temperature difference threshold to the third temperature difference threshold; and a first-level warning state in red when the temperature difference value exceeds the third temperature difference threshold. By observing the curves, the operator can clearly see the trend of temperature changes with load, as well as the frequency and duration of warning states. Furthermore, they can combine this information with the drive operation logs and production line cycle data to analyze the causes of overheating.

[0085] Through the above implementation method, the visualization module works closely with the temperature correction module and the early warning judgment module to convert the third temperature value, temperature difference value and early warning level information into intuitive temperature curves, temperature difference curves and early warning status indicators. Within a preset time range, the changes in servo motor temperature and early warning status are displayed in graphical or textual form, thereby providing operators with clear and intuitive monitoring methods, facilitating the timely detection of abnormal temperature trends and high-risk operating conditions, and improving the efficiency and accuracy of operation and maintenance analysis and fault diagnosis.

[0086] The following examples illustrate the components and functions of the system in this application. This application optimizes the temperature correction coefficient using a differential evolution algorithm, corrects the surface temperature in real time, and provides three levels of early warning based on the temperature difference. Verification has shown that, compared to the unoptimized version, this application can dynamically adjust the temperature correction coefficient to achieve multi-level early warning. The system mainly includes the following key components:

[0087] 1. Real-time data reading module: Obtain the surface temperature and ambient temperature of the servo motor in real time through sensors or data interfaces; the data format includes surface temperature and ambient temperature.

[0088] 2. Temperature correction module: Dynamically optimize the temperature correction coefficient k using the differential evolution algorithm (considering the complexity of the servo motor operation, its upper and lower threshold ranges are 0.05 - 0.18), and correct the surface temperature in real time. Correction formula: Tactual = Tsurface + k × (Tambient - Tstandard). The code for this part is as follows.

[0089] Optimize the temperature correction coefficient k using the differential evolution algorithm: bounds = [0.05, 0.18]; options = struct('MaxGenerations', 100, 'PopulationSize', 20, 'Display', true);

[0090] [k_optimal, bestFitness, fitnessHistory] = differential_evolution(@(k) fitness_function(k, surface_temp, ambient_temp, T_standard), bounds, options);

[0091] Calculate the corrected actual temperature: T_actual = surface_temp + k_optimal * (ambient_temp - T_standard);

[0092] Calculate the temperature difference: delta_temp = T_actual - ambient_temp.

[0093] 3. Warning module: Conduct three-level warnings (severe, medium, minor) based on the temperature difference between the corrected temperature and the ambient temperature. The 3-level warning alarm mechanism is a hierarchical warning method based on preset fixed thresholds. By setting 3 different fixed thresholds: T1, T2, T3, and satisfying T1 < T2 < T3, and dividing the value range of the monitoring index into 4 intervals, corresponding to different risk levels respectively, so as to achieve hierarchical warning and alarm for the monitoring object. The code for this part is as follows.

[0094] During the actual monitoring process, based on the surface temperature T of the servo motor monitored in real time, the early warning level is judged according to the above threshold range, and the corresponding early warning and alarm measures are triggered: when T ≤ T1, the system determines that it is in a normal state, no early warning is processed, and the real-time monitoring of the motor temperature continues; when T1 < T ≤ T2, an attention-level early warning is triggered, and the system issues a third-level early warning signal to remind the operator to pay attention to the change trend of the motor temperature and make further observations; when T2 < T ≤ T3, a warning-level early warning is triggered, the system issues a second-level early warning signal, and at the same time sends an early warning notice to the operator, and it is recommended to take cooling measures, such as adjusting the motor load, optimizing the heat dissipation system, etc.; when T > T3, a danger-level early warning is triggered, the system issues a first-level early warning signal, such as reminding with a red mark on the monitoring interface, and at the same time sends an emergency early warning notice to the operator, and immediately starts the shutdown protection program to avoid damage to the motor due to high temperature. Define the three-level early warning range:

[0095] 1) First-level early warning: temperature difference > 50°C;

[0096] 2) Second-level early warning: 40°C < temperature difference ≤ 50°C;

[0097] 3) Third-level early warning: 30°C < temperature difference ≤ 40°C;

[0098] 4) Normal range: temperature difference ≤ 30°C.

[0099] 4. Visualization module: Real-time display of the corrected temperature, temperature difference, and early warning level, which is convenient for monitoring and analysis.

[0100] During the operation of the servo motor in the intelligent control system, the actual performance parameters mainly include: voltage, current, line resistance, input power, output power, rotational speed, torque, inductance, back electromotive force, torque coefficient, ambient temperature, motor surface temperature, etc.

[0101] The following combines specific experimental results to illustrate the effects achieved by this system. In this verification session, a 400W servo motor is used as an example: A servo system is built with a certain brand of servo industrial computer, servo driver, and servo motor, and a temperature data collector is used to obtain the surface temperature index set under the thermal equilibrium state, and 200 groups of parameter data are selected, as shown in Table 1.

[0102] Table 1 Environmental and surface temperature index sets under thermal equilibrium state

[0103]

[0104] Before optimization k The value is set to a fixed value of 0.18, and the optimal value after optimization k is 0.0962. Compare the optimized optimization result with the temperature difference result before optimization.

[0105] Experiments have verified that this application uses the differential evolution algorithm for automatic optimization. k The optimized temperature difference accuracy differs from the unoptimized result by only 0.065℃, improving the accuracy of temperature correction. This means the optimized corrected temperature is closer to the ambient temperature, better reflecting the actual physical process, making the corrected temperature more accurate, and improving the accuracy and reliability of the early warning system. The early warning graph correctly displays the three-level early warning and the normal range early warning for the predicted temperature. This early warning mechanism can normally implement the three-level early warning mechanism.

[0106] In view of the influence of differences in motor performance (such as power and heat dissipation structure) and complex working conditions (such as high temperature environment and long-term load), and the heat generation problem of motor temperature under thermal equilibrium state, this application proposes a servo motor temperature correction coefficient optimization and multi-level intelligent early warning system. After verification by measured servo motor surface test data, this application has at least the following advantages and positive effects: (1) Dynamic optimization: The temperature correction coefficient is dynamically optimized by differential evolution algorithm, which can not only adapt to environmental changes in real time, but also get closer to the ambient temperature and better reflect the actual physical process. (2) Multi-level early warning: A three-level early warning mechanism is designed, which can provide graded early warning according to different temperature difference ranges, providing more detailed monitoring and early warning. (3) Intelligence and automation: The system automatically optimizes the correction coefficient by differential evolution algorithm without manual intervention, which improves the intelligence level of the system.

[0107] The implementation process of the servo motor temperature correction coefficient optimization and multi-level early warning method of this application will be described in detail below with reference to specific embodiments. Figure 2 This is a flowchart illustrating the servo motor temperature correction coefficient optimization and multi-level early warning method provided in this application embodiment, as shown below. Figure 2 As shown, the method may specifically include the following steps:

[0108] S201, acquire in real time a first temperature value of the surface temperature of the servo motor and a second temperature value of the ambient temperature around the servo motor from a temperature sensor connected to the servo motor.

[0109] S202, under the control of the current temperature correction coefficient according to the preset temperature correction model, the first temperature value is corrected to obtain the third temperature value of the actual temperature of the servo motor, and the temperature difference value is calculated based on the third temperature value and the second temperature value.

[0110] S203, within the preset coefficient value range, the differential evolution algorithm is used to optimize the temperature correction coefficient in the temperature correction model online, including constructing a fitness function based on the first temperature value, the second temperature value and the third temperature value, performing differential mutation, crossover and selection operations on the candidate set of temperature correction coefficients, iteratively updating to obtain the target temperature correction coefficient, and feeding the target temperature correction coefficient back to the temperature correction module as the current temperature correction coefficient used in subsequent corrections;

[0111] S204, Based on the temperature difference value, compare it with at least three successively increasing preset thresholds, divide the current working state of the servo motor into one of the normal state and at least three warning states, and generate warning level information corresponding to the warning state.

[0112] S205 outputs corresponding warning signals and control commands to the upper-level control device and / or human-machine interface according to the warning level information, and implements graded warning control for the operation of the servo motor.

[0113] Specifically, in step S201, the data acquisition module in the AI ​​machine acquires a first temperature value of the servo motor surface temperature from a surface temperature sensor installed on the servo motor housing, and a second temperature value at the corresponding moment from an ambient temperature sensor arranged in the space surrounding the servo motor, according to a preset sampling period. The data acquisition module simultaneously reads the output signals from both temperature sensors at the same sampling moment, and after signal conditioning and analog-to-digital conversion, obtains the first and second temperature values ​​respectively. Based on the system clock, a unified time stamp is added to this set of temperature data, forming a temperature data sequence arranged in chronological order. These time-stamped first and second temperature values ​​will subsequently serve as the basic input for temperature correction and correction coefficient optimization.

[0114] In step S202, the temperature correction module receives a first temperature value, a second temperature value, and a current temperature correction coefficient from the data acquisition module. Under the control of an internally preset temperature correction model, it corrects the first temperature value. The temperature correction module has preset standard ambient temperature parameters, such as 25℃ for the servo motor during calibration testing. First, it determines the correction amount corresponding to the first temperature value at that sampling moment based on the difference between the second temperature value and the standard ambient temperature parameter, and the current temperature correction coefficient. Then, it calculates the correction amount with the first temperature value to obtain a third temperature value representing the actual temperature of the servo motor. Subsequently, the temperature correction module subtracts the third temperature value from the second temperature value at the same sampling moment to obtain the temperature difference value at that moment, which reflects the degree to which the actual temperature of the servo motor exceeds the ambient temperature. For example, at a certain sampling moment, the first temperature value is 70℃, the second temperature value is 35℃, the current temperature correction coefficient is 0.1, the temperature correction module calculates an environmental deviation of 10℃, and a correction amount of 1℃, resulting in a third temperature value of 71℃, corresponding to a temperature difference of 36℃.

[0115] In step S203, the correction coefficient optimization module uses a differential evolution algorithm to optimize the temperature correction coefficients in the temperature correction model online within a preset coefficient value range. The module first constructs a sample set based on the reference temperature data obtained from the servo motor calibration test and the first and second temperature values ​​collected during operation. Each sample in the sample set includes the first and second temperature values ​​under a certain operating condition, along with the corresponding reference temperature data. Multiple sets of temperature correction coefficients are generated within the preset coefficient value range (e.g., 0.05 to 0.18), forming a candidate set of temperature correction coefficients. For each candidate temperature correction coefficient, the optimization module uses the same temperature correction model as the temperature correction module, calculates the corresponding third temperature value based on the first and second temperature values ​​in the sample set, and compares it one by one with the reference temperature data of the same sample set to obtain the error value. The errors of each sample are accumulated and / or averaged to obtain the error evaluation index of the candidate temperature correction coefficient. This error evaluation index is used as the fitness function value for the candidate temperature correction coefficient. During the differential evolution iteration process, the correction coefficient optimization module performs differential mutation, crossover, and selection operations on the candidate set, updates the temperature correction coefficients of each candidate based on the fitness function value, and selects the temperature correction coefficient with the optimal fitness function from the candidate set as the target temperature correction coefficient when the iteration reaches the preset number of algebras or the convergence condition is met. This coefficient is then fed back to the temperature correction module to replace the original current temperature correction coefficient for use in the temperature correction calculation of subsequent sampling periods.

[0116] In step S204, the early warning determination module compares the temperature difference value output by the temperature correction module with at least three sequentially increasing preset thresholds to classify the current operating state of the servo motor into different levels. The early warning determination module presets a first temperature difference threshold, a second temperature difference threshold, and a third temperature difference threshold, with the first temperature difference threshold being less than the second temperature difference threshold, and the second temperature difference threshold being less than the third temperature difference threshold. For example, the first temperature difference threshold is set to 30℃, the second temperature difference threshold to 40℃, and the third temperature difference threshold to 50℃. When the temperature difference value is less than or equal to the first temperature difference threshold, the early warning determination module determines the current operating state of the servo motor as normal; when the temperature difference value is greater than the first temperature difference threshold and less than or equal to the second temperature difference threshold, it is determined to be a third-level early warning state; when the temperature difference value is greater than the second temperature difference threshold and less than or equal to the third temperature difference threshold, it is determined to be a second-level early warning state; and when the temperature difference value is greater than the third temperature difference threshold, it is determined to be a first-level early warning state. The early warning determination module encapsulates the above determination results into early warning level information and sends it along with a time stamp to the early warning output module and the visualization module.

[0117] In step S205, the early warning output module outputs corresponding early warning signals and control commands to the upper-level control device and / or human-machine interface according to the early warning level information, implementing graded early warning control for the operation of the servo motor. When the early warning level information indicates the third-level early warning state, the early warning output module mainly outputs prompt-type early warning signals to the human-machine interface, such as highlighting the current motor and its temperature curve on the interface, and reminding the operator to pay attention to temperature changes with a prompt sound, without directly interfering with the operation of the servo motor; when the early warning level information indicates the second-level early warning state, the early warning output module enhances the interface and audible and visual alarm prompts, and sends control commands to the servo motor drive device, instructing the drive device to reduce the load on the servo motor and / or start or enhance the operation of the heat dissipation device to slow down the temperature rise; when the early warning level information indicates the first-level early warning state, the early warning output module issues a stop control command to the servo motor drive device, controlling the servo motor to stop running after controlled deceleration, and at the same time prompting maintenance personnel to handle the situation in a timely manner with the highest level alarm on the monitoring interface.

[0118] Through the above implementation process, the servo motor temperature correction coefficient optimization and multi-level early warning method of this application, based on real-time acquisition of surface temperature and ambient temperature, combines a preset temperature correction model and differential evolution algorithm to achieve online optimization of the temperature correction coefficient, making the third temperature value closer to the actual temperature of the servo motor; at the same time, using the corrected temperature difference value as a unified index, and through multi-level threshold judgment and hierarchical control strategy, it achieves refined management of the thermal state of the servo motor, which is conducive to improving the accuracy of temperature estimation and the sensitivity of early warning judgment, reducing the risk of servo motor overheating damage, and taking into account the safety of equipment operation and production continuity.

[0119] It should be understood that the sequence number of each step in the above method embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0120] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although the technical solutions of this application have been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A servo motor temperature correction coefficient optimization and multi-level early warning system, characterized in that, include: The data acquisition module is used to acquire, in real time, the first temperature value of the surface temperature of the servo motor and the second temperature value of the ambient temperature around the servo motor from the temperature sensor connected to the servo motor. The temperature correction module is used to correct the first temperature value according to the preset temperature correction model and under the control of the current temperature correction coefficient to obtain the third temperature value of the actual temperature of the servo motor, and to calculate the temperature difference value based on the third temperature value and the second temperature value. The correction coefficient optimization module is used to optimize the temperature correction coefficient in the temperature correction model online within a preset coefficient value range using a differential evolution algorithm. The correction coefficient optimization module constructs a fitness function based on the first temperature value, the second temperature value, and the third temperature value, performs differential mutation, crossover, and selection operations on the candidate set of temperature correction coefficients, iteratively updates the target temperature correction coefficient, and provides the target temperature correction coefficient to the temperature correction module as the current temperature correction coefficient. The early warning determination module is used to compare the temperature difference value with at least three sequentially increasing preset thresholds, classify the current working state of the servo motor into one of the normal state and at least three levels of early warning state, and generate early warning level information corresponding to the early warning state. The early warning output module is used to output corresponding early warning signals and control commands to the upper control device and / or human-machine interface according to the early warning level information, so as to implement graded early warning control for the operation of the servo motor; Specifically, the correction coefficient optimization module is used for: Multiple sets of temperature correction coefficients are generated within the preset coefficient value range to form a candidate set of temperature correction coefficients. Based on the sample set consisting of the first temperature value, the second temperature value, and the reference temperature data corresponding to the servo motor calibration test, the third temperature value obtained under each temperature correction coefficient is calculated using the temperature correction model. A fitness function is constructed based on the error between the third temperature value and the reference temperature data. During the differential evolution iteration process, differential mutation, crossover, and selection operations are performed on the candidate set of temperature correction coefficients. Each candidate temperature correction coefficient is updated according to the value of the fitness function until a preset convergence condition is met. The temperature correction coefficient with the best fitness function is selected from the updated candidate set of temperature correction coefficients as the target temperature correction coefficient, and the target temperature correction coefficient is provided to the temperature correction module as the current temperature correction coefficient.

2. The system according to claim 1, characterized in that, The data acquisition module includes a surface temperature sensor disposed on the housing of the servo motor and an ambient temperature sensor disposed in the space surrounding the servo motor. The data acquisition module acquires the first temperature value and the second temperature value from the surface temperature sensor and the ambient temperature sensor, respectively, and adds a time stamp to each temperature value.

3. The system according to claim 2, characterized in that, The data acquisition module is also used to acquire one or more of the operating parameters of the servo motor, such as voltage, current, and speed, and send the operating parameters, along with the first temperature value and the second temperature value, to the correction coefficient optimization module. When constructing the fitness function, the correction coefficient optimization module uses the operating parameters as auxiliary variables in the calculation.

4. The system according to claim 1, characterized in that, The temperature correction module is specifically used for: In the temperature correction module, a standard ambient temperature parameter is preset. Based on the difference between the second temperature value and the standard ambient temperature parameter and the current temperature correction coefficient, a correction amount corresponding to the first temperature value is determined. The correction amount and the first temperature value are calculated to obtain the third temperature value. The temperature difference value is determined based on the difference between the third temperature value and the second temperature value.

5. The system according to claim 1, characterized in that, The process of constructing a fitness function based on the error between the third temperature value and the reference temperature data includes: For each group of temperature samples in the sample set, the corresponding third temperature value is calculated using the temperature correction model under each candidate temperature correction coefficient, and compared with the reference temperature data of the same group to obtain the error value. The error values ​​are accumulated and / or averaged to obtain the error evaluation index, and the error evaluation index is used as the value of the fitness function under the candidate temperature correction coefficient.

6. The system according to claim 1, characterized in that, The early warning judgment module is preset with a first temperature difference threshold, a second temperature difference threshold and a third temperature difference threshold. The first temperature difference threshold is less than the second temperature difference threshold and the second temperature difference threshold is less than the third temperature difference threshold. When the temperature difference value is less than or equal to the first temperature difference threshold, it is determined to be in the normal state. When the temperature difference value is greater than the first temperature difference threshold and less than or equal to the second temperature difference threshold, it is determined to be in the third-level warning state. When the temperature difference value is greater than the second temperature difference threshold and less than or equal to the third temperature difference threshold, it is determined to be in the second-level warning state. When the temperature difference value is greater than the third temperature difference threshold, it is determined to be in the first-level warning state.

7. The system according to claim 6, characterized in that, When the warning output module receives the warning level information, it outputs a prompt signal to remind the operator to pay attention to the temperature change of the servo motor in the third warning state; in the second warning state, it outputs a control command to the servo motor drive device to reduce the load of the servo motor and / or adjust the heat dissipation device; and in the first warning state, it outputs a shutdown control command to the servo motor drive device to stop the operation of the servo motor.

8. The system according to any one of claims 1 to 7, characterized in that, It also includes a visualization module, which is connected to the temperature correction module and the early warning determination module. The visualization module is used to generate a temperature curve, a temperature difference curve and an early warning status identifier based on the third temperature value, the temperature difference value and the early warning level information, and to display the temperature change and early warning status change of the servo motor in a preset time range in the form of graphics or text on the display interface.

9. A method for optimizing the temperature correction coefficient of a servo motor and providing multi-level early warning based on the system described in any one of claims 1 to 8, characterized in that, include: The first temperature value of the servo motor surface temperature and the second temperature value of the ambient temperature around the servo motor are obtained in real time from the temperature sensor connected to the servo motor. According to the preset temperature correction model and under the control of the current temperature correction coefficient, the first temperature value is corrected to obtain the third temperature value of the actual temperature of the servo motor, and the temperature difference value is calculated based on the third temperature value and the second temperature value. Within a preset range of coefficient values, the temperature correction coefficients in the temperature correction model are optimized online using a differential evolution algorithm. This includes constructing a fitness function based on the first temperature value, the second temperature value, and the third temperature value; performing differential mutation, crossover, and selection operations on the candidate set of temperature correction coefficients; iteratively updating the target temperature correction coefficients; and feeding the target temperature correction coefficients back to the temperature correction module as the current temperature correction coefficients used in subsequent corrections. Based on the comparison between the temperature difference value and at least three sequentially increasing preset thresholds, the current working state of the servo motor is divided into one of the normal state and at least three warning states, and warning level information corresponding to the warning state is generated. Based on the aforementioned warning level information, corresponding warning signals and control commands are output to the upper-level control device and / or human-machine interface, and graded warning control is implemented for the operation of the servo motor. In this process, multiple sets of temperature correction coefficients are generated within the preset coefficient value range to form a candidate set of temperature correction coefficients. Based on the sample set consisting of the first temperature value, the second temperature value, and the reference temperature data corresponding to the servo motor calibration test, the third temperature value obtained under each temperature correction coefficient is calculated using the temperature correction model. A fitness function is constructed based on the error between the third temperature value and the reference temperature data. During the differential evolution iteration process, differential mutation, crossover, and selection operations are performed on the candidate set of temperature correction coefficients. Each candidate temperature correction coefficient is updated according to the value of the fitness function until a preset convergence condition is met. The temperature correction coefficient with the best fitness function is selected from the updated candidate set of temperature correction coefficients as the target temperature correction coefficient, and the target temperature correction coefficient is provided to the temperature correction module as the current temperature correction coefficient.

Images