Industrial multi-axis servo motor discrete temperature health assessment method based on cloud model

Abstract

The invention provides an industrial multi-axis servo motor discrete temperature health assessment method based on a cloud model. The method comprises the following steps: acquiring a temperature discrete sample sequence of each servo motor; performing preprocessing operation on the temperature discrete sample sequence to obtain a preprocessed temperature sample sequence; when the temperature change rate of the pre-processing temperature sample sequence is lower than a set threshold value and lasts for a preset time length, dividing the pre-processing temperature sample sequence into heat balance windows corresponding to the servo motors, and respectively calculating an expected value, an entropy value and a super entropy value by taking the heat balance windows as input and adopting a reverse cloud generator; inputting the expected value, the entropy value and the hyper-entropy value obtained for each servo motor into a forward cloud generator to generate a cloud droplet set corresponding to the servo motor; and calculating a cloud droplet condensation degree based on the cloud droplet set, and generating a temperature health assessment level corresponding to each servo motor according to a preset judgment rule. According to the invention, the uncertainty of the discrete temperature data of the servo motor can be quantified in a unified manner, and the multi-axis temperature health level is generated.

Description

Technical Field

[0001] This application relates to the field of industrial automation technology, and in particular to a method for discrete temperature health assessment of industrial multi-axis servo motors based on a cloud model. Background Technology

[0002] Industrial servo control systems typically employ multi-axis servo motors to collaboratively drive mechanical actuators. During long-term operation, servo motors experience temperature rises due to copper and iron losses, as well as variations in environmental heat dissipation conditions. Their steady-state temperature and temperature rise rate reflect key performance indicators such as winding insulation aging, bearing lubrication status, and the unobstructedness of cooling channels. Online monitoring and health assessment of multi-axis servo motor temperature parameters have become a crucial foundation for ensuring the stable operation of high-precision, high-reliability manufacturing equipment.

[0003] In multi-axis servo scenarios, the load, heat dissipation location, and installation posture of each motor differ, resulting in a discrete temperature distribution. Furthermore, the temperature variation of the same motor under different operating conditions also exhibits randomness and ambiguity. Existing technologies mainly include the following methods: Single-machine thermal equivalent circuit and finite element simulation method: predict the winding temperature rise by establishing a thermal network or three-dimensional simulation model, but it depends on the motor structural parameters, the calculation is complicated, and it is difficult to evaluate the multi-axis in real time on site.

[0004] Fixed threshold or single-point over-temperature alarm method: An absolute temperature upper limit is set inside the driver, and an alarm is triggered when the sampled temperature exceeds the threshold. This method ignores the dispersion of the sampled data within the normal range and cannot detect early anomalies.

[0005] The joint diagnostic method of vibration and current signature: It integrates current harmonics, vibration spectrum and temperature curve to identify faults, focuses on electromagnetic or mechanical fault mechanisms, and is insufficient in handling the fuzzy features of discrete temperature data.

[0006] However, the aforementioned existing technologies are mainly geared towards single motors or continuous temperature rise curves, and cannot adapt to the characteristics of multi-axis servo systems where temperature sampling points are discrete, random, and fuzzy. When using fixed thresholds, it is difficult to reflect the differences in thermal balance ranges of different motors, which can easily lead to false alarms or missed alarms. Methods based on simulation or complex signal analysis have high computational load and implementation costs, and lack a unified evaluation model for rapid field deployment. Summary of the Invention

[0007] In view of this, this application provides a cloud model-based discrete temperature health assessment method for industrial multi-axis servo motors to solve the problem in the prior art that it is difficult to uniformly quantify the randomness and fuzziness of discrete temperature data of multi-axis servo motors and output the health level.

[0008] A first aspect of this application provides a method for discrete temperature health assessment of industrial multi-axis servo motors based on a cloud model, comprising: during the operation of a multi-axis servo control system, collecting discrete temperature sample sequences of each servo motor using a temperature acquisition device and archiving them according to motor identification; performing filtering, de-glitching, and time synchronization processing on the discrete temperature sample sequences to obtain a preprocessed temperature sample sequence; when the temperature change rate of the preprocessed temperature sample sequence is lower than a set threshold and continues for a preset duration, dividing the preprocessed temperature sample sequence into thermal balance windows corresponding to the servo motors, and using the thermal balance windows as input, calculating the expected value, entropy value, and hyper-entropy value respectively using a reverse cloud generator; inputting the expected value, entropy value, and hyper-entropy value obtained for each servo motor into a forward cloud generator to generate a cloud droplet set corresponding to the servo motor; calculating the cloud droplet cohesion based on the cloud droplet set, and generating a temperature health assessment level corresponding to each servo motor according to a preset judgment rule; and outputting the temperature health assessment level to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor.

[0009] A second aspect of this application provides a cloud-based discrete temperature health assessment device for industrial multi-axis servo motors, comprising: a data acquisition module for acquiring discrete temperature sample sequences of each servo motor using a temperature acquisition device during the operation of a multi-axis servo control system, and archiving them according to motor identification; a preprocessing module for performing filtering, de-glitching, and time synchronization processing on the discrete temperature sample sequences to obtain a preprocessed temperature sample sequence; a calculation module for dividing the preprocessed temperature sample sequence into thermal balance windows corresponding to the servo motors when the temperature change rate of the preprocessed temperature sample sequence is lower than a set threshold and continues for a preset duration, and using the thermal balance windows as input, calculating the expected value, entropy value, and hyper-entropy value respectively using a reverse cloud generator; an input module for inputting the expected value, entropy value, and hyper-entropy value obtained for each servo motor into a forward cloud generator to generate a cloud droplet set corresponding to the servo motor; a generation module for calculating the cloud droplet cohesion based on the cloud droplet set, and generating a temperature health assessment level corresponding to each servo motor according to a preset judgment rule; and an output module for outputting the temperature health assessment level to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor.

[0010] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described method.

[0011] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described method.

[0012] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: By collecting discrete temperature sample sequences from each servo motor using a temperature acquisition device during the operation of a multi-axis servo control system, and archiving them according to motor identification, this paper describes a method for processing temperature sample sequences. The discrete temperature sample sequences are then filtered, de-glitched, and time-synchronized to obtain a pre-processed temperature sample sequence. When the temperature change rate of the pre-processed temperature sample sequence is lower than a set threshold and remains below a preset duration, the sequence is divided into thermal equilibrium windows for the corresponding servo motors. Using these windows as input, a reverse cloud generator is used to calculate the expected value, entropy value, and hyper-entropy value. The expected value, entropy value, and hyper-entropy value obtained for each servo motor are then input into a forward cloud generator to generate a cloud droplet set for that servo motor. Based on the cloud droplet set, the cloud droplet cohesion is calculated, and a temperature health assessment level corresponding to each servo motor is generated according to preset judgment rules. The temperature health assessment level is then output to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor. This application can uniformly quantify the uncertainty of discrete temperature data of servo motors and generate multi-axis temperature health levels. Attached Figure Description

[0013] 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.

[0014] Figure 1 This is a flowchart illustrating the discrete temperature health assessment method for industrial multi-axis servo motors based on a cloud model provided in this application embodiment. Figure 2 This is a schematic diagram of the thermal balance cloud model of a two-axis servo motor provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of the cloud-based discrete temperature health assessment device for industrial multi-axis servo motors provided in this application embodiment; Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0015] 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.

[0016] Existing multi-axis servo control systems typically rely on thermal equivalent circuit modeling or finite element simulation to predict motor temperature rise, or set fixed over-temperature thresholds within the driver for single-point alarms. The former depends on detailed structural parameters and high computing power, making rapid field deployment difficult; the latter ignores the dispersion of temperature sampling data for each axis within the normal range, easily leading to false alarms or missed alarms. Furthermore, while some joint diagnostic solutions incorporate signals such as vibration and current, they lack a unified quantification method for the randomness and fuzziness inherent in discrete temperature samples, failing to provide intuitive and usable health grading results.

[0017] Current technology is unable to simultaneously consider random and fuzzy uncertainties and perform unified quantization on discretely acquired temperature samples in multi-axis servo scenarios, nor can it generate health levels for each motor that can be directly used for operation and maintenance management based on the quantization results.

[0018] In view of the problems existing in the prior art, the purpose of this application is to provide a discrete temperature health assessment method for industrial multi-axis servo motors based on a cloud model. This application collects discrete temperature samples of each motor at the same frequency during system operation. After filtering, deburring, and time synchronization, the temperature change rate is detected to determine the thermal equilibrium window. For the data within the thermal equilibrium window, the expected value, entropy value, and hyperentropy value are calculated using a reverse cloud generator, and these three digital features are input into a forward cloud generator to generate a cloud droplet set. Subsequently, the cloud droplet cohesion is calculated and compared with preset judgment rules to obtain the corresponding temperature health assessment level for each motor. This level is then reported to the operation and maintenance decision module to guide operation management.

[0019] By using random-fuzzy unified quantization of cloud models, this application can accurately reflect the uncertainty of discrete parameters of servo motor thermal balance temperature, thereby generating intuitive multi-axis temperature health levels to meet the needs of rapid on-site assessment and operation and maintenance decision-making.

[0020] The technical solution of this application will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0021] Figure 1 This is a flowchart illustrating the discrete temperature health assessment method for industrial multi-axis servo motors based on a cloud model, as provided in an embodiment of this application. Figure 1 As shown, the discrete temperature health assessment method for industrial multi-axis servo motors based on cloud models can specifically include: S101, during the operation of the multi-axis servo control system, uses a temperature acquisition device to collect discrete temperature sample sequences of each servo motor and archives them according to the motor identification. S102, perform filtering, descrambling and time synchronization processing on the discrete temperature sample sequence to obtain a preprocessed temperature sample sequence; S103, when the temperature change rate of the preprocessed temperature sample sequence is lower than the set threshold and continues for a preset duration, the preprocessed temperature sample sequence is divided into the corresponding servo motor thermal balance window, and the expected value, entropy value and hyperentropy value are calculated by using the thermal balance window as input and the reverse cloud generator respectively. S104 inputs the expected value, entropy value and hyperentropy value obtained for each servo motor into the forward cloud generator to generate a set of cloud droplets for the corresponding servo motor. S105 calculates cloud droplet cohesion based on cloud droplet aggregation and generates temperature health assessment levels corresponding to each servo motor according to preset judgment rules; S106 outputs the temperature health assessment level to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor.

[0022] In some embodiments, a temperature acquisition device is used to acquire discrete temperature sample sequences for each servo motor, and these sequences are archived according to motor identification, including: Each servo motor is assigned a corresponding temperature sensing channel. The temperature acquisition device performs synchronous sampling on all channels at a preset sampling period. A timestamp is generated each time a sample is taken, and the motor identifier of the corresponding servo motor is written into the sampled data frame. Temperature sampling data with motor identification and timestamps are written into a circular buffer storage area, and an index is created according to the motor identification. The data is then organized in chronological order to form a discrete temperature sample sequence that corresponds one-to-one with each servo motor.

[0023] Specifically, a multi-axis servo control system typically includes several servo motors and corresponding drivers. In this embodiment, an isolated K-type thermocouple is installed on the outer wall of the stator winding of each servo motor, and connected to an independent channel of the temperature acquisition device via a shielded cable. The acquisition device adopts a modular multi-channel structure, with each module supporting eight differential inputs and providing thermal junction compensation and common-mode rejection at the hardware level. To ensure multi-channel synchronization, the acquisition device has a built-in hardware synchronization pulse generation circuit, which can broadcast the synchronization pulse along the backplane bus to all channels to trigger sampling, thereby avoiding sampling phase offset between channels.

[0024] Furthermore, the sampling period of the temperature acquisition device's control logic unit can be configured by the upper-level system before power failure. In this embodiment, the sampling period is set to 1 second, with allowable adjustments in 0.1-second increments. Whenever a synchronization pulse arrives, all active channels simultaneously perform analog-to-digital conversion on the corresponding thermocouple voltage and immediately perform cold junction compensation. Subsequently, the temperature value, timestamp, and motor identifier are concatenated to generate a data frame. The timestamp resolution is 1ms and can be provided via an internal real-time clock; the motor identifier uses hexadecimal two-byte encoding, with the encoding scheme being issued once by the AI ​​during system initialization and archived on the device. The temperature value is represented using hexadecimal two's complement, with a range of -50℃ to +200℃ and a resolution of 0.1℃.

[0025] Furthermore, the acquisition device incorporates a 32MB dual-port static random access memory (SRAM), divided into several circular buffers, with each channel occupying its own buffer. The buffer write pointer automatically increments with each sample. When the distance between the write and read pointers is less than one frame, an overwrite strategy is triggered, replacing the oldest frame with the newest frame to ensure continuous writability of the circular structure. To reduce the host computer's retrieval latency, the acquisition device updates the index table immediately after writing a new frame based on the motor identifier. The index table is located in the on-chip static storage area, using the motor identifier as the key and the latest write pointer position of the corresponding buffer as the value, supporting rapid location of time-contiguous data blocks by motor identifier.

[0026] Furthermore, the AI-powered machine periodically sends batch read commands to the acquisition device via gigabit Ethernet, with the command header specifying the motor identifier and the number of samples to be collected. Under DMA control, the acquisition device packages and uploads consecutive frames that meet the conditions. The AI-powered machine maintains a sample linked list in its local memory, categorized by motor identifier, and appends newly arrived data to the end of the linked list. Once the linked list reaches a set depth, a time window sliding is triggered, forming a discrete temperature sample sequence that corresponds one-to-one with the corresponding servo motor and has a strictly monotonically increasing time order. This provides the data foundation for subsequent filtering and thermal balance window detection.

[0027] For example, in a specific instance, during the no-load trial run of a dual-axis horizontal truss transport line, two servo motors with a rated power of 400W were selected. A K-type thermocouple, cured by dispensing, was attached to the outer wall of each motor, with the thermocouple junction tightly fitted to the stator housing. The temperature acquisition device was set to a sampling period of 1 second, with a circular buffer capacity of 36,000 frames, capable of storing approximately 10 hours of data. The AI ​​retrieved temperature data every 60 seconds, fetching 60 frames each time, and immediately performed digital low-pass filtering and moving average processing. After 4 hours of trial run, the system formed 14,400 discrete samples in the sample lists of each of the two motors, indexed by motor identifiers, meeting the lower limit of the sample capacity for the subsequent reverse cloud generator.

[0028] The above implementation process completes the collection and archiving of discrete temperature samples for the technical solution of this application, laying a data foundation for subsequent thermal equilibrium window division, cloud model parameter calculation, and health level assessment. Those skilled in the art can make equivalent adjustments to the sampling period, buffer capacity, and transmission method based on equipment scale, temperature fluctuation characteristics, and network bandwidth, without departing from the core idea of ​​this application.

[0029] In some embodiments, the preprocessed temperature sample sequence is divided into thermal balance windows corresponding to the servo motor, including: Calculate the rate of temperature change between adjacent sampling points in the preprocessed temperature sample sequence in chronological order; The temperature change rate is compared with a set threshold. When the temperature change rate is consistently lower than the threshold and the duration is not less than the preset time, the thermal equilibrium start time is determined. Based on the thermal equilibrium start time, continuous temperature data is extracted according to the preset window length to generate the corresponding servo motor thermal equilibrium window. The thermal balance window is associated with the motor identifier and stored, and used as input to the reverse cloud generator.

[0030] Specifically, in this embodiment, the sampling period for the preprocessed temperature sample sequence is 1 second, and each sample contains a timestamp and a temperature value. The system divides the sequence into thermal equilibrium windows according to the following process: Step 1: Calculate the rate of temperature change in chronological order The control algorithm module reads the temperature value T from two adjacent sampling points. k With T k+1 Calculate the rate of temperature change ΔT = (T k+1 -T k The rate of temperature change is calculated as ΔT / Δt, where Δt is 1 second. When the rate of temperature change is measured in °C / min, the system multiplies ΔT by 60 to obtain the rate of change per minute. All rate of change results are written in real time to the buffer corresponding to the motor identifier for threshold comparison.

[0031] Step 2: Compare the rate of temperature change with the set threshold. At the factory, the system has a threshold value of R_th = 0.05℃ / min and a duration of D_th = 300s set in the parameter library. When the rate of change of D_th / Δt = 300 consecutive samples is not higher than R_th, the temperature curve is determined to have entered thermal equilibrium during that period. The control algorithm module records this moment t0 as the start time of thermal equilibrium.

[0032] Step 3: Select the thermal equilibrium window Starting from t0, the system extracts 600 consecutive temperature samples according to a preset window length W_len=600s to form a thermal equilibrium window. If the sampling sequence is less than 600s after t0, the sampling is extended until the window length is reached.

[0033] Step 4: Store and associate the identifier After the capture is complete, the system packages all temperature samples within the window along with the motor identifier and writes them to a persistent file named "HotBalance_Mxxx", where Mxxx represents the motor identifier code. The file header contains the window start and end times, the number of samples, and the sampling period, which are used for subsequent parsing.

[0034] The fifth step serves as the input for the reverse cloud generator. Upon receiving the "HotBalance_Mxxx" file, the operation and maintenance analysis server immediately invokes the reverse cloud generator module, using all temperature values ​​in the window as input samples to calculate the expected value, entropy value, and hyperentropy value, laying the foundation for the forward cloud generator to generate a cloud droplet set.

[0035] Through the steps of the above embodiments, this embodiment achieves automatic identification of the thermal equilibrium window based on the rate of temperature change without adding extra sensors or complex calculations. This ensures that the reverse cloud generator only processes data under steady-state conditions, improving the accuracy and reliability of the evaluation results. Those skilled in the art can substitute R_th, D_th, and W_len with equivalent values ​​based on actual temperature rise characteristics without exceeding the scope of protection of this application.

[0036] In some embodiments, using a thermal equilibrium window as input, a reverse cloud generator is employed to calculate the expected value, entropy value, and hyperentropy value, including: Statistical processing is performed on the temperature sampling data within the thermal equilibrium window to calculate the sample mean and sample variance, and the sample mean is determined as the expected value. The entropy value is calculated based on the sample variance and a preset cloud model formula, and the hyperentropy value is calculated based on the entropy value and a preset cloud model formula. The expected value, entropy value, and hyperentropy value are associated with the motor identifier of the corresponding servo motor and then stored.

[0037] Specifically, after the thermal equilibrium window is captured, this embodiment uses a reverse cloud generator to calculate the expected value, entropy value, and hyperentropy value. The specific operation process is as follows: Step 1: Statistical analysis of sample data After receiving a window file named HotBalance_Mxxx, the operations and maintenance analysis server parses it to obtain 600 consecutive temperature sampling points. The server calls the statistics module to calculate the sample mean μ and sample variance S² using an unbiased estimation method. To reduce calculation errors, the mean and variance are calculated online using a double-loop accumulation algorithm, eliminating the need to load all samples into memory at once. The sample mean μ is directly used as the expected value Ex.

[0038] Step 2: Calculate the entropy value The system pre-defines the cloud model entropy formula En = √π / 2·σ in the model parameter library, where σ is the sample standard deviation. The statistics module immediately takes the square root of S² to obtain σ, and substitutes it into the formula to calculate the entropy value En. If the sample size is less than one hundred, the module automatically triggers an incremental merging strategy, waiting for the next window to be stitched together to ensure statistical stability.

[0039] Step 3: Calculate the hyperentropy value According to the cloud model definition, the hyperentropy value He = En·√1-2 / π. The statistics module directly calculates He by substituting the previously calculated En into the formula, without needing to traverse the samples. The entire calculation process only involves addition, subtraction, multiplication, division, and square root operations, making it suitable for real-time execution on an on-site industrial control computer.

[0040] Step 4: Storage and Identification Association The statistics module packages Ex, En, He, the motor identifier Mxxx, and the window start time t0, and writes them into a parameter file named CloudParam_Mxxx_t0. At the same time, it creates a new record in the relational database with fields including motor identifier, expected value, entropy value, hyperentropy value, number of samples, and window timestamp, which is convenient for subsequent retrieval and calling by the positive cloud generator.

[0041] For example, in a specific scenario, taking the thermal balance window of a 400-watt servo motor on axis one as an example, 600 samples measured μ=67.4℃ and σ=0.91℃, which translates to Ex=67.4, En≈1.13, and He≈0.30. The sample statistics for a motor of the same power on axis two yielded Ex=69.7, En≈0.85, and He≈0.26. It is evident that the entropy and hyperentropy values ​​of axis one are greater than those of axis two, indicating that the temperature distribution on axis one is more dispersed. These parameters will then be input into a forward cloud generator to produce a cloud droplet set, laying the data foundation for determining cloud droplet cohesion.

[0042] Through the above steps, this embodiment achieves real-time on-site extraction of three digital features without requiring code programming or complex mathematical libraries, ensuring the accuracy and interpretability of subsequent temperature health assessments. Those skilled in the art can adjust the window length and sample lower limit according to different motor thermal inertia constants, or calibrate the coefficients in the entropy and hyperentropy formulas to adapt to higher precision scenarios without departing from the core idea of ​​this application.

[0043] In some embodiments, the expected value, entropy value, and hyperentropy value obtained for each servo motor are input into the forward cloud generator to generate a cloud droplet set corresponding to the servo motor, including: Generate a first random number with the entropy value as the mean and the super-entropy value as the variance, and record the first random number as the entropy random number; A second random number is generated with the expected value as the mean and the entropy random number as the variance. The second random number is recorded as the temperature random number. The membership degree is calculated based on the expected value and the temperature random number according to the cloud model membership degree formula. The temperature random number is paired with the corresponding membership degree to generate a single cloud droplet. Repeat the generation operation until the number of cloud droplets reaches the preset value, forming a set of cloud droplets corresponding to the servo motor, and store it with the motor identifier.

[0044] Specifically, after calculating the expected value, entropy value, and hyperentropy value, the system calls the forward cloud generator to generate a set of temperature cloud droplets for each servo motor. The specific implementation process is as follows: Step 1: Set the number of cloud droplets and the random number generator The operation and maintenance analysis server presets 1,000 cloud droplets for each motor in the configuration file, and specifies the seed for the pseudo-random number generator to ensure that the results of multiple evaluations can be compared repeatedly. The random number generator kernel adopts a hybrid structure of hardware noise source and linear congruential algorithm, and outputs a zero-mean, unit-variance normal sequence after whitening.

[0045] Step 2: Generate entropy random numbers The server reads a set of zero-mean, unit-variance normal random numbers within the loop, multiplies the random number by the hyperentropy value and adds the entropy value to obtain the entropy random number; in this embodiment, the entropy random number is limited to a positive value, and if a negative value is encountered, it is immediately discarded and resampled to ensure that the subsequent variance is valid.

[0046] Step 3: Generate random temperature numbers and membership degrees Using the entropy random number as the variance and the expected value as the mean, the random number generator is called again to output a zero-mean, unit-variance random number. This random number is then multiplied by the entropy random number and added to the expected value to obtain the temperature random number. The system calculates the membership degree according to the cloud model membership degree calculation formula μ = exp[-(temperature random number - expected value)² ÷ (double entropy random number²)]. Temperature random numbers and membership degrees are paired one-to-one to form individual cloud droplets, which are then written to the memory buffer in real time.

[0047] Step 4: Building and Persisting Cloud Droplet Collections The second and third steps are executed iteratively until the accumulated cloud droplets reach a preset quantity of one thousand. The system then packages the buffer contents into a cloud droplet collection file, with the filename including the motor identifier and timestamp. The file header records the expected value, entropy value, hyperentropy value, and number of cloud droplets, while the body is arranged in the order of "temperature random number, membership degree". After generation, the system inserts an index record into the relational database, with fields including the motor identifier, cloud droplet file path, and generation time, facilitating rapid retrieval by the subsequent cloud droplet cohesion calculation module.

[0048] Through the above steps, this embodiment can complete the forward cloud generator processing of multi-axis servo motor temperature data without complex programming, forming a set of cloud droplets reflecting the discrete characteristics of temperature, providing high-quality input for cloud droplet cohesion calculation and temperature health level determination. Those skilled in the art can adjust the number of cloud droplets or the implementation method of the random number generator according to the motor power level or maintenance accuracy requirements, all of which are equivalent variations of this application.

[0049] In some embodiments, cloud droplet cohesion is calculated based on the cloud droplet set, and a temperature health assessment level corresponding to each servo motor is generated according to a preset judgment rule, including: The number of cloud droplets with a membership degree not lower than a preset membership threshold in the cloud droplet set is counted, and the ratio of the counted number of cloud droplets to the total number of cloud droplets in the set is determined as the cloud droplet cohesion degree. The cloud droplet cohesion is compared with the evaluation threshold in the preset judgment rule to obtain the corresponding temperature health assessment level identifier, and then the temperature health assessment level identifier is associated with the motor identifier of the servo motor and stored.

[0050] Specifically, after the cloud droplet aggregation is generated by the positive cloud generator, the operation and maintenance analysis server starts the cloud droplet cohesion calculation thread and determines the temperature health assessment level according to the following process: The first step is to read the cloud droplet collection. The server locates the cloud droplet collection file based on the database index, parses the random temperature numbers and membership degrees sequentially, and loads the membership degree values ​​into a memory array. To ensure data consistency, the cloud droplet quantity field in the file header is compared with the actual record entries during parsing; if a discrepancy is found, an exception is immediately reported and processing is aborted.

[0051] The second step is to count the number of highly affiliated cloud droplets. The system pre-sets the membership threshold λ to a value of 0.6 in the parameter library. When traversing the membership array, if the membership degree is not lower than λ, it is considered a high-membership cloud droplet, and the counter k is incremented simultaneously. After the traversal is completed, the total number of cloud droplets n is recorded, and the high-membership ratio η = k ÷ n is calculated. η is the cloud droplet cohesion degree in this embodiment.

[0052] Step 3: Compare and evaluate thresholds The parameter library provides two levels of evaluation thresholds: Level 1 threshold η1 is set to 0.75, and Level 2 threshold η2 is set to 0.5. If η is not lower than η1, the temperature health assessment level is identified as "Level 1"; if η is between η2 and η1, it is identified as "Level 2"; if η is lower than η2, it is identified as "Level 3". These thresholds can be adjusted after calibration to adapt to servo motors of different power levels or insulation levels.

[0053] Step 4: Associate and store the motor identifier. The server packages the temperature health assessment level identifier, motor identifier, cloud droplet cohesion, and window start time into a result structure, writes it to the assessment result table in the relational database, and simultaneously updates the operation and maintenance visualization interface. The primary key of the result table is composed of the motor identifier and the window start time, supporting quick queries by time range or by motor.

[0054] Step 5 triggers subsequent operation and maintenance strategies After receiving the assessment results, the operation and maintenance decision-making module retrieves a preset set of policies based on the level identifier: when the identifier is "Level 1", the current operation is maintained; when the identifier is "Level 2", the acceleration limit is reduced by 5 percentage points and a weekly inspection is scheduled; when the identifier is "Level 3", the on-duty personnel are notified immediately and a shutdown command is issued. The policy execution record is synchronously written to the operation and maintenance log to achieve closed-loop management.

[0055] Through the above steps, this embodiment, without relying on complex external algorithms, utilizes the high membership droplet ratio as a cohesion index to quickly determine the temperature health level of a multi-axis servo motor, providing on-site maintenance personnel with a clear and traceable decision-making basis. Those skilled in the art can make equivalent substitutions for the membership threshold λ, evaluation thresholds η1 and η2, and corresponding maintenance strategies according to actual production needs, without departing from the core technical solution of this application.

[0056] In some embodiments, the temperature health assessment level is output to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor, including: The temperature health assessment level is converted into a digital identification code corresponding to the servo motor identification and encapsulated according to a predetermined data message format; The encapsulated data message is sent to the operation and maintenance decision module via the industrial communication bus, so that the operation and maintenance decision module can parse the data message and obtain the servo motor identifier and the corresponding temperature health assessment level. The operation and maintenance decision module retrieves the preset operation strategy library based on the analysis results, selects the operation management strategy that matches the temperature health assessment level, and sends control commands to the servo drive unit to achieve the corresponding management. The temperature health assessment level and the selected operation and management strategy are written into the operation and maintenance log database for subsequent query and traceability.

[0057] Specifically, after determining the temperature health level of each servo motor, the operation and maintenance analysis server needs to write the assessment results back to the industrial servo control system in real time, so that the operation and maintenance decision-making module can automatically adjust the operating strategy based on the health level. The implementation method in this embodiment is as follows.

[0058] First, the operation and maintenance analysis server converts the temperature health assessment level into a numerical identifier that corresponds one-to-one with the motor identifier. The conversion rules are issued by the configuration file during system initialization: health level 1 is mapped to the number "01", level 2 to the number "02", and level 3 to the number "03". The server builds a message buffer in memory and fills the fields according to the predetermined data message format. The message header is fixed as hexadecimal "AA55", the length field is automatically calculated based on the payload, and the payload part sequentially arranges the motor identifier, the temperature health level numerical identifier, and the window start timestamp. A one-byte XOR checksum is appended to the message tail.

[0059] Next, the server connects to the industrial switch via Gigabit Ethernet and publishes the encapsulated data packets to the control network using the real-time industrial Ethernet protocol. The switch supports priority queuing at the hardware level, marking health assessment packets as the highest priority to ensure bandwidth preemption even under high load scenarios. The operation and maintenance decision module, acting as a network subscriber, immediately verifies the packet header, length, and checksum upon receiving the packet, and parses the motor identifier and its corresponding temperature health assessment level.

[0060] The operation and maintenance decision module pre-stores a library of operating strategies. When it detects a health level indicator "01", it maintains the current operating parameters. When it detects indicator "02", it invokes the "load reduction and maintenance" strategy, automatically reducing the peak acceleration parameter of the servo drive unit by 10% and setting an operation and maintenance task to complete on-site inspection within eight hours. When it detects indicator "03", it triggers the "emergency shutdown" strategy, sending a shutdown command to the corresponding servo drive unit via the fieldbus and simultaneously illuminating a red warning light on the alarm indicator in the control room. All strategy execution commands are directly issued by the operation and maintenance decision module via the fieldbus, with a transmission delay of less than fifty milliseconds.

[0061] Upon completion, the operations and maintenance decision module writes the health level digital identifier, the selected operation management strategy identifier, and the execution status to the operations and maintenance log database. The log database uses a circular write mechanism and supports querying by motor identifier and timestamp index, facilitating operations and maintenance personnel to trace historical health level changes and strategy execution records for each motor. If subsequent analysis requires evaluating the effectiveness of the strategy, continuous data can be extracted from the log database and statistically compared with production cycle time and fault records to further improve the strategy library.

[0062] The following simulation process, along with accompanying drawings, verifies the correctness and effectiveness of the embodiments of this application. Figure 2This is a schematic diagram of the thermal balance cloud model of a two-axis servo motor provided in an embodiment of this application, as shown below. Figure 2 As shown, the simulation process includes the following: In this embodiment, two servo motors with a rated power of 400W (denoted as axis 1 and axis 2) are selected, along with an AI machine, servo driver, and temperature data acquisition instrument of the same brand, to construct a dual-axis servo system. Both servo motors operate continuously under no-load conditions with a rated voltage of 220V and a rated speed of 3000rpm. The sampling period of the temperature acquisition instrument is set to 1s, and it simultaneously samples the thermocouples on the outer walls of both motors under the trigger of a hardware synchronization pulse. This continues until the temperature change rate is stably below 0.05℃ / min for 300s, at which point the system is considered to have entered thermal equilibrium. At this point, 600 consecutive samples are taken from each motor as the temperature index set for the thermal equilibrium window.

[0063] First, reverse cloud generator parameter extraction. Perform statistical processing on the thermal equilibrium window temperature sequences of axis 1 and axis 2: The sample mean for axis 1 is μ1≈67.4℃, and the sample standard deviation is σ1≈0.91℃. The sample mean for axis 2 is μ2≈69.7℃, and the sample standard deviation is σ2≈0.78℃.

[0064] Furthermore, the three digital features are calculated based on the cloud model formula: Axis 1: Expected value 67.4, entropy 1.13, hyperentropy 0.30; Axis 2: Expected value 69.7, entropy 0.97, hyperentropy 0.26.

[0065] It can be seen that the entropy and hyperentropy values ​​of axis 1 are slightly larger, which means that the temperature distribution range is relatively wide; the temperature concentration of axis 2 is higher.

[0066] Next, the forward cloud generator produces cloud droplets. Input the above three digital features into the forward cloud generator and repeat the generation process at a setting of 1000 cloud droplets per axis: First, generate random entropy numbers using the entropy value as the mean and the hyperentropy value as the variance. Then, generate random temperature numbers using the expected value as the mean and the entropy random number as the variance; Calculate membership degrees and pair them into individual cloud droplets until two cloud droplet sets are formed.

[0067] Then, visualize the two-dimensional distribution of cloud droplets. Figure 2 The results of a two-axis servo motor thermal equilibrium cloud model are presented. The horizontal axis represents temperature (°C), and the vertical axis represents membership degree. Cloud droplets are densely distributed in an inverted bell shape, and the centroid of the cloud droplet cluster corresponds to the expected value position on each axis. Comparison shows that: The cloud droplet cluster on axis 1 is slightly wider, with a left-right distribution range of approximately 65°C to 70°C. The cloud droplet clusters of Axis 2 are more slender and mainly concentrated between 69°C and 71°C.

[0068] The statistical results of cloud droplet cohesion η are as follows: Axis 1 is 0.78 and Axis 2 is 0.64. Based on the preset thresholds η1 = 0.75 and η2 = 0.50, Axis 1 is rated as Level 1 healthy and Axis 2 is rated as Level 2 healthy.

[0069] The difference in cloud center of gravity indicates a temperature deviation of approximately 2.3℃ between the two motors at thermal equilibrium, consistent with the actual differences in bearing and heat dissipation environments under no-load conditions. Axis 1 exhibits higher cloud droplet condensation and a thinner cloud layer, reflecting smaller temperature fluctuations and better stability. While Axis 2 is within acceptable limits, its relatively lower condensation suggests the need to monitor its heat dissipation channels or lubrication status. This result was directly adopted by the operation and maintenance decision-making module: Axis 1 will maintain its current parameters; Axis 2's peak acceleration will be reduced by 10%, and routine inspections will be scheduled ahead of schedule.

[0070] This embodiment demonstrates, through real-machine sampling, three-digit feature extraction from a cloud model, and two-dimensional visualization, that the proposed discrete temperature health assessment method can: Consistently handles randomness and fuzziness in a multi-axis environment and provides quantitative levels; Provides intuitive cloud maps to help operations and maintenance personnel quickly locate potential temperature anomalies; Generate health levels that can directly drive adjustments to operational strategies.

[0071] The verification results show that the method in this application is correct and effective, and can provide reliable decision support for the selection and maintenance of servo motors.

[0072] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0073] Figure 3 This is a schematic diagram of the structure of the discrete temperature health assessment device for industrial multi-axis servo motors based on a cloud model provided in this application embodiment. Figure 3 As shown, the cloud-based industrial multi-axis servo motor discrete temperature health assessment device includes: The acquisition module 301 is used to acquire discrete temperature sample sequences of each servo motor using a temperature acquisition device during the operation of the multi-axis servo control system, and archive them according to the motor identification. Preprocessing module 302 is used to perform filtering, descrambling and time synchronization processing on the discrete temperature sample sequence to obtain a preprocessed temperature sample sequence; The calculation module 303 is used to divide the preprocessed temperature sample sequence into the corresponding servo motor thermal balance window when the temperature change rate of the preprocessed temperature sample sequence is lower than the set threshold and continues for a preset time. The module uses the thermal balance window as input and the reverse cloud generator to calculate the expected value, entropy value and hyperentropy value respectively. The input module 304 is used to input the expected value, entropy value and hyperentropy value obtained for each servo motor into the forward cloud generator to generate a set of cloud droplets for the corresponding servo motor. The generation module 305 is used to calculate the cloud droplet cohesion based on the cloud droplet set and generate the temperature health assessment level corresponding to each servo motor according to the preset judgment rules. Output module 306 is used to output the temperature health assessment level to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor.

[0074] In some embodiments, Figure 3 The acquisition module 301 sets up a corresponding temperature sensing channel for each servo motor. The temperature acquisition device performs synchronous sampling on all channels at a preset sampling period. A timestamp is generated each time sampling, and the motor identifier of the corresponding servo motor is written into the sampling data frame. The temperature sampling data with motor identifier and timestamp is written into the circular buffer storage area, and an index is established according to the motor identifier. The data is then organized in chronological order to form a discrete temperature sample sequence that corresponds one-to-one with each servo motor.

[0075] In some embodiments, Figure 3 The calculation module 303 calculates the temperature change rate between adjacent sampling points in the preprocessed temperature sample sequence in chronological order; compares the temperature change rate with a set threshold, and determines the thermal equilibrium start time when the temperature change rate is continuously lower than the threshold and the duration is not less than the preset time; based on the thermal equilibrium start time, continuous temperature data is extracted according to the preset window length to generate the corresponding servo motor thermal equilibrium window; the thermal equilibrium window is associated with the motor identifier and stored, and used as the input of the reverse cloud generator.

[0076] In some embodiments, Figure 3 The calculation module 303 performs statistical processing on the temperature sampling data within the thermal balance window, calculates the sample mean and sample variance, and determines the sample mean as the expected value; calculates the entropy value based on the sample variance and the preset cloud model formula, and calculates the hyper-entropy value based on the entropy value and the preset cloud model formula; and stores the expected value, entropy value, hyper-entropy value and the corresponding servo motor motor identifier.

[0077] In some embodiments, Figure 3The input module 304 generates a first random number with an entropy value as the mean and a super-entropy value as the variance, and records the first random number as the entropy random number; it generates a second random number with an expected value as the mean and an entropy random number as the variance, and records the second random number as the temperature random number. It calculates the membership degree based on the expected value and the temperature random number according to the cloud model membership degree formula, and pairs the temperature random number with the corresponding membership degree to generate a single cloud droplet; the generation operation is repeated until the number of cloud droplets reaches a preset value, forming a set of cloud droplets corresponding to the servo motor, and storing it with the motor identifier.

[0078] In some embodiments, Figure 3 The generation module 305 counts the number of cloud droplets in the cloud droplet set whose membership degree is not lower than the preset membership threshold, and determines the cloud droplet cohesion degree by the ratio of the counted cloud droplet number to the total number of cloud droplets in the cloud droplet set; compares the cloud droplet cohesion degree with the evaluation threshold in the preset judgment rule to obtain the corresponding temperature health assessment level identifier, and stores the temperature health assessment level identifier after associating it with the motor identifier of the servo motor.

[0079] In some embodiments, Figure 3 The output module 306 converts the temperature health assessment level into a digital identification code corresponding to the servo motor identification and encapsulates it according to a predetermined data message format. The encapsulated data message is then sent to the operation and maintenance decision module via the industrial communication bus, allowing the module to parse the data message, obtain the servo motor identification and the corresponding temperature health assessment level, retrieve the preset operation strategy library based on the parsing results, select the operation management strategy matching the temperature health assessment level, and issue control commands to the servo drive unit to achieve the corresponding management. The temperature health assessment level and the selected operation management strategy are written into the operation and maintenance log database for subsequent querying and traceability.

[0080] It should be understood that the sequence number of each step in the above 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.

[0081] Figure 4 This is a schematic diagram of the electronic device 4 provided in an embodiment of this application. Figure 4 As shown, the electronic device 4 of this embodiment includes: a processor 401, a memory 402, and a computer program 403 stored in the memory 402 and executable on the processor 401. When the processor 401 executes the computer program 403, it implements the steps in the various method embodiments described above. Alternatively, when the processor 401 executes the computer program 403, it implements the functions of each module / unit in the various device embodiments described above.

[0082] Electronic device 4 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 4 may include, but is not limited to, processor 401 and memory 402. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or different components.

[0083] The processor 401 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0084] The memory 402 can be an internal storage unit of the electronic device 4, such as a hard disk or RAM of the electronic device 4. The memory 402 can also be an external storage device of the electronic device 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc., equipped on the electronic device 4. The memory 402 can also include both internal and external storage units of the electronic device 4. The memory 402 is used to store computer programs and other programs and data required by the electronic device.

[0085] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0086] If integrated modules / units are implemented as software functional units and sold or used as independent products, they can be stored in a readable storage medium (e.g., a computer-readable storage medium). Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which may be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable storage medium may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0087] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has 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 discrete temperature health assessment method for industrial multi-axis servo motors based on a cloud model, characterized in that, include: During the operation of the multi-axis servo control system, a temperature acquisition device is used to collect discrete temperature sample sequences of each servo motor, and these are archived according to the motor identification. The discrete temperature sample sequence is subjected to filtering, descrambling, and time synchronization processing to obtain a preprocessed temperature sample sequence. When the temperature change rate of the preprocessed temperature sample sequence is lower than the set threshold and continues for a preset duration, the preprocessed temperature sample sequence is divided into the corresponding servo motor thermal balance window, and the expected value, entropy value and hyperentropy value are calculated respectively using the thermal balance window as input and the reverse cloud generator. The expected value, entropy value, and hyperentropy value obtained for each servo motor are input into the forward cloud generator to generate a set of cloud droplets for the corresponding servo motor. The cloud droplet cohesion is calculated based on the cloud droplet set, and a temperature health assessment level corresponding to each servo motor is generated according to a preset judgment rule. The temperature health assessment level is output to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor.

2. The method according to claim 1, characterized in that, The process of collecting discrete temperature sample sequences from each servo motor using a temperature acquisition device and archiving them according to motor identification includes: Each servo motor is assigned a corresponding temperature sensing channel. The temperature acquisition device performs synchronous sampling on all channels at a preset sampling period. A timestamp is generated each time a sample is taken, and the motor identifier of the corresponding servo motor is written into the sampled data frame. Temperature sampling data with motor identification and timestamps are written into a circular buffer storage area, and an index is created according to the motor identification. The data is then organized in chronological order to form a discrete temperature sample sequence that corresponds one-to-one with each servo motor.

3. The method according to claim 1, characterized in that, The step of dividing the preprocessed temperature sample sequence into corresponding servo motor thermal balance windows includes: Calculate the rate of temperature change between adjacent sampling points in the preprocessed temperature sample sequence in chronological order; The temperature change rate is compared with a set threshold. When the temperature change rate is continuously lower than the threshold and the duration is not less than a preset time, the thermal equilibrium start time is determined. Based on the thermal equilibrium start time, continuous temperature data is extracted according to the preset window length to generate the corresponding servo motor thermal equilibrium window. The associated motor identifier of the thermal balance window is stored and used as input to the reverse cloud generator.

4. The method according to claim 1, characterized in that, The step of using the thermal equilibrium window as input and employing a reverse cloud generator to calculate the expected value, entropy value, and hyperentropy value includes: Statistical processing is performed on the temperature sampling data within the thermal equilibrium window to calculate the sample mean and sample variance, and the sample mean is determined as the expected value. The entropy value is calculated based on the sample variance and a preset cloud model formula, and the hyperentropy value is calculated based on the entropy value and the preset cloud model formula. The expected value, entropy value, and hyperentropy value are associated with the motor identifier of the corresponding servo motor and then stored.

5. The method according to claim 1, characterized in that, The process involves inputting the expected value, entropy value, and hyperentropy value obtained for each servo motor into the forward cloud generator to generate a cloud droplet set corresponding to the servo motor, including: A first random number is generated using the entropy value as the mean and the hyperentropy value as the variance, and the first random number is denoted as the entropy random number. A second random number is generated with the expected value as the mean and the entropy random number as the variance. The second random number is recorded as the temperature random number. The membership degree is calculated based on the expected value and the temperature random number according to the cloud model membership degree formula. The temperature random number is paired with the corresponding membership degree to generate a single cloud droplet. Repeat the generation operation until the number of cloud droplets reaches the preset value, forming a set of cloud droplets corresponding to the servo motor, and store it with the motor identifier.

6. The method according to claim 1, characterized in that, The calculation of cloud droplet cohesion based on the cloud droplet set, and the generation of temperature health assessment levels corresponding to each servo motor according to preset judgment rules, include: The number of cloud droplets with a membership degree not lower than a preset membership threshold in the cloud droplet set is counted, and the ratio of the counted number of cloud droplets to the total number of cloud droplets in the set is determined as the cloud droplet cohesion degree. The cloud droplet cohesion is compared with the evaluation threshold in the preset judgment rule to obtain the corresponding temperature health assessment level identifier, and the temperature health assessment level identifier is associated with the motor identifier of the servo motor and then stored.

7. The method according to claim 1, characterized in that, The step of outputting the temperature health assessment level to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor includes: The temperature health assessment level is converted into a digital identification code corresponding to the servo motor identification and encapsulated according to a predetermined data message format; The encapsulated data message is sent to the operation and maintenance decision module via the industrial communication bus, so that the operation and maintenance decision module can parse the data message and obtain the servo motor identifier and the corresponding temperature health assessment level. The operation and maintenance decision module retrieves the preset operation strategy library based on the analysis results, selects the operation management strategy that matches the temperature health assessment level, and sends control commands to the servo drive unit to achieve the corresponding management. The temperature health assessment level and the selected operation and management strategy are written into the operation and maintenance log database for subsequent querying and traceability.

8. A discrete temperature health assessment device for industrial multi-axis servo motors based on a cloud model, characterized in that, include: The data acquisition module is used to collect discrete temperature sample sequences from each servo motor using a temperature acquisition device during the operation of the multi-axis servo control system, and archive them according to the motor identification. The preprocessing module is used to perform filtering, descrambling, and time synchronization processing on the discrete temperature sample sequence to obtain a preprocessed temperature sample sequence. The calculation module is used to divide the preprocessed temperature sample sequence into the corresponding servo motor thermal balance window when the temperature change rate of the preprocessed temperature sample sequence is lower than the set threshold and continues for a preset time. The module uses the thermal balance window as input and a reverse cloud generator to calculate the expected value, entropy value and hyperentropy value respectively. The input module is used to input the expected value, entropy value and hyperentropy value obtained for each servo motor into the forward cloud generator to generate the cloud droplet set corresponding to the servo motor. The generation module is used to calculate the cloud droplet cohesion based on the cloud droplet set and generate the temperature health assessment level corresponding to each servo motor according to the preset judgment rules. The output module is used to output the temperature health assessment level to the operation and maintenance decision module of the industrial servo control system to manage the operating status of each servo motor.

9. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images