A Digital Model-Based Method and System for Evaluating the Insulation Status of Motor Stator Windings

Abstract

This invention relates to the field of motor health management and predictive maintenance, specifically to a method and system for assessing the insulation status of motor stator windings based on a digital model. The method includes: S1, acquiring high-frequency data from the motor stator windings to generate real-time data of multiple physical quantities; S2, calculating a dynamic capacitance reference value from the real-time data of multiple physical quantities; S3, extracting an insulation degradation residual signal by subtracting the measured high-frequency equivalent capacitance value from the dynamic capacitance reference value; S4, constructing an adaptive dynamic detection threshold from the real-time data of multiple physical quantities; S5, determining whether the absolute value of the insulation degradation residual signal is greater than the adaptive dynamic detection threshold: if yes, an insulation degradation event is determined to have occurred; if no, it is determined to be in normal operating condition; S6, updating the insulation degradation index in response to the insulation degradation event, and generating an insulation status assessment based on the updated insulation degradation index. This invention avoids false alarms under drastic load fluctuations, significantly improving the accuracy and reliability of the assessment.

Description

Technical Field

[0001] This invention relates to the field of motor health management and predictive maintenance, specifically to a method and system for assessing the insulation status of motor stator windings based on digital models. Background Technology

[0002] In modern industrial production, electric motors, as core power equipment, are crucial for reliable operation. The insulation system of the motor stator windings is a key component ensuring long-term stable operation, and accurate assessment of its health status is of great significance for preventing sudden failures and enabling predictive maintenance. Using high-frequency equivalent capacitance as a sensitive parameter characterizing insulation status has become an effective technical approach.

[0003] Traditional insulation condition detection methods often rely on static models or fixed alarm thresholds. However, during actual motor operation, the equivalent capacitance of the windings not only undergoes irreversible changes due to the aging of the insulation material itself, but also experiences reversible and healthy dynamic changes due to fluctuations in real-time operating conditions such as winding temperature, load current, and rotor speed. Traditional methods struggle to accurately distinguish between these two completely different sources of parameter change, resulting in insufficient detection sensitivity when the motor is operating stably, failing to effectively identify early, weak degradation signals; while during periods of drastic fluctuation in operating conditions, normal parameter fluctuations are easily misinterpreted as insulation abnormalities, leading to numerous false alarms.

[0004] While some online monitoring solutions exist in existing technologies, they generally fail to establish a dynamic benchmark model that accurately describes the coupling relationship between multiple physical quantities and insulation capacitance. These methods neglect the combined effects of thermal expansion due to temperature, electromagnetic pressure effect of load current, and mechanical stress effect of rotational speed on the capacitance benchmark value. Furthermore, existing judgment logic mostly uses static thresholds, lacking dynamic judgment thresholds that adaptively adjust with the severity of operating conditions, which greatly limits the robustness and reliability of detection in complex and ever-changing industrial applications.

[0005] Therefore, how to provide a method that can couple the real-time operating conditions of the motor, effectively filter out the interference of fluctuations in normal operating conditions, and thus realize online, high-precision detection and evaluation of early and weak insulation degradation of the motor stator winding is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention discloses a simulation method and system for evaluating the insulation state of motor stator windings based on digital models. Specifically, the technical solution includes:

[0007] The method for evaluating the insulation status of motor stator windings based on digital models includes the following steps:

[0008] S1. Collect the high-frequency equivalent capacitance measurement value of the motor stator winding, the real-time temperature of the winding, the motor load current and the motor rotor speed, and generate real-time data of multiple physical quantities.

[0009] S2. Based on the real-time data of the multiple physical quantities, calculate the dynamic capacitance reference value through the working condition coupled dynamic reference model;

[0010] S3. Perform a subtraction operation between the measured value of the high-frequency equivalent capacitance and the reference value of the dynamic capacitance to extract the insulation degradation residual signal;

[0011] S4. Based on the real-time data of the multiple physical quantities, construct an adaptive dynamic detection threshold;

[0012] S5. Determine whether the absolute value of the insulation degradation residual signal is greater than the adaptive dynamic detection threshold: if yes, determine that an insulation degradation event has occurred; if no, determine that it is in normal operation.

[0013] S6. In response to the insulation degradation event, update the insulation degradation index and generate an insulation status assessment based on the updated insulation degradation index.

[0014] Preferably, S2 specifically includes:

[0015] The real-time temperature of the winding, the motor load current, and the motor rotor speed are extracted from the real-time data of the multiple physical quantities.

[0016] The root mean square (RMS) value of the motor load current is calculated to obtain the effective value of the current.

[0017] The real-time temperature of the winding, the effective value of the current, and the rotor speed of the motor are input into the operating condition coupled dynamic reference model to calculate the dynamic capacitance reference value.

[0018] Preferably, the operating condition coupled dynamic reference model decomposes the change in the equivalent capacitance of the winding into a linear superposition of the thermal expansion effect caused by temperature, the electromagnetic pressure effect caused by the load current, and the mechanical stress effect caused by the rotor speed.

[0019] Preferably, S4 specifically includes:

[0020] Extract the motor load current and the real-time winding temperature from the real-time data of the multiple physical quantities;

[0021] The load current of the motor and the real-time temperature of the winding are smoothed and numerically differentiated to obtain a smooth load change rate and temperature change rate.

[0022] The adaptive dynamic detection threshold is constructed by combining the preset basic detection threshold, the smooth load change rate, the temperature change rate, and the preset robustness adjustment coefficient.

[0023] Preferably, the determination step S5 further includes:

[0024] Before determining whether an insulation degradation event has occurred, it is also necessary to determine whether the duration of the event has reached the preset anti-jitter time window;

[0025] The insulation degradation event is finally confirmed and step S6 is triggered only when the duration reaches the anti-jitter time window.

[0026] Preferably, S6 specifically includes:

[0027] The insulation degradation index is set to an initial value of zero;

[0028] Whenever an insulation degradation event is received, the insulation degradation index is accumulated according to the severity of the event;

[0029] The current value of the insulation degradation index is mapped to the health status level, and slope analysis is performed on its historical data to calculate the degradation rate. Finally, the remaining service life when the preset failure threshold is reached is extrapolated.

[0030] Preferably, the insulation degradation event is correlated with the corresponding operating condition data at the time of occurrence to identify specific operating modes that accelerate insulation degradation, and operation optimization strategy suggestions are generated accordingly.

[0031] A digital model-based system for assessing the insulation condition of motor stator windings includes:

[0032] The multi-physical quantity real-time data acquisition module is used to acquire the high-frequency equivalent capacitance measurement value of the motor stator winding, the real-time temperature of the winding, the motor load current and the motor rotor speed, and generate multi-physical quantity real-time data.

[0033] The working condition coupled dynamic reference model module is used to calculate the dynamic capacitance reference value based on the real-time data of the multiple physical quantities.

[0034] An insulation degradation residual signal extraction module is used to perform a subtraction operation between the high-frequency equivalent capacitance measurement value and the dynamic capacitance reference value to extract the insulation degradation residual signal.

[0035] An adaptive threshold and state assessment module is used to construct an adaptive dynamic detection threshold based on the real-time data of the multiple physical quantities, and to determine the insulation state based on the comparison result between the insulation degradation residual signal and the adaptive dynamic detection threshold.

[0036] The degradation trend prediction and condition-based maintenance module is used to update the insulation degradation index and generate an insulation status assessment in response to the determination results of insulation degradation events.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] 1. By constructing a dynamic benchmark model of the real-time operating conditions of the coupled motor and an adaptive detection threshold, this invention can accurately distinguish between normal operating condition fluctuations and parameter changes caused by insulation degradation, thereby effectively avoiding false alarms under severe load fluctuations and significantly improving the accuracy and reliability of the assessment.

[0039] 2. The method of the present invention can effectively filter out normal fluctuations in capacitance caused by changes in temperature, load, and speed, thereby accurately capturing weak signals caused only by irreversible aging of insulation materials, and realizing high-precision online detection of early insulation degradation of motor stator windings.

[0040] 3. This invention introduces a quantified insulation degradation index, which accumulates discrete degradation events into a continuous health status indicator. Combined with historical data trend analysis, it realizes the calculation of degradation rate and prediction of remaining service life, providing a scientific basis for predictive maintenance.

[0041] 4. The system of the present invention can perform correlation analysis between insulation degradation events and the corresponding operating condition data at the time of occurrence, identify specific operating modes that accelerate insulation aging, and generate operation optimization strategy suggestions accordingly, helping users to shift from passive response to proactive prevention and extend the service life of motors. Attached Figure Description

[0042] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0043] Figure 1 This is a flowchart of the method of the present invention.

[0044] Figure 2 This is a flowchart of the system of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. Example 1

[0046] Please see Figure 1 A method for evaluating the insulation status of motor stator windings based on digital models includes the following steps:

[0047] S1. Collect the high-frequency equivalent capacitance measurement value of the motor stator winding, the real-time temperature of the winding, the motor load current and the motor rotor speed, and generate real-time data of multiple physical quantities.

[0048] S2. Based on real-time data of multiple physical quantities, the dynamic capacitance reference value is calculated through a dynamic reference model coupled with operating conditions.

[0049] S3. Perform a subtraction operation between the high-frequency equivalent capacitance measurement value and the dynamic capacitance reference value to extract the insulation degradation residual signal;

[0050] S4. Construct an adaptive dynamic detection threshold based on real-time data of multiple physical quantities;

[0051] S5. Determine whether the absolute value of the insulation degradation residual signal is greater than the adaptive dynamic detection threshold: if yes, determine that an insulation degradation event has occurred; if no, determine that it is in normal operation.

[0052] S6. In response to an insulation degradation event, update the insulation degradation index and generate an insulation status assessment based on the updated insulation degradation index;

[0053] This embodiment provides a method for evaluating the insulation status of motor stator windings based on a digital model. The purpose of this method is to construct a dynamic digital model that can couple the real-time operating conditions of the motor, accurately distinguish between parameter changes caused by fluctuations in normal operating conditions and parameter changes caused by insulation degradation itself. This ensures extremely high detection sensitivity while effectively avoiding false alarms under conditions of severe load fluctuations, thereby achieving accurate and online evaluation of the insulation status of motor stator windings.

[0054] The specific process of this method includes the following steps:

[0055] S1: The generation of real-time data on multiple physical quantities provides real-time and synchronous data input for subsequent dynamic benchmark calculations and threshold construction. In this embodiment, by deploying corresponding sensors at key parts of the motor, at least four physical quantities are collected in real time, which together constitute the multi-dimensional operating condition information required to assess the insulation status. This step involves collecting the high-frequency equivalent capacitance measurement value of the motor stator winding. Real-time winding temperature Motor load current and motor rotor speed The system performs validity checks on the acquired raw signals, removes outlier data that are outside the physical possible range (e.g., temperatures below -50°C or above 200°C), and performs interpolation processing on brief signal loss to generate stable and reliable real-time data of multiple physical quantities. These four physical quantities are recorded synchronously to form a time series dataset, namely real-time data of multiple physical quantities, which serves as a unified data source for all subsequent calculations.

[0056] S2: Calculation of dynamic capacitance reference value

[0057] Using real-time operating condition data and a pre-calibrated model, the capacitance value that a theoretically healthy motor should have under the current specific operating condition is calculated in real time. In this embodiment, based on the real-time data of multiple physical quantities generated by S1, a dynamic capacitance reference value is calculated through an operating condition coupled dynamic reference model. This model uses physical quantities such as temperature, current, and rotation speed as endogenous variables, and can accurately reproduce the capacitive response characteristics of healthy insulation under different working conditions.

[0058] S3: Extraction of Insulation Degradation Residual Signal

[0059] By removing the health fluctuations caused by changes in operating conditions from the actual measured values, a signal that purely reflects the irreversible degradation state of the insulating material is separated. In this embodiment, the high-frequency equivalent capacitance measurement value obtained in S1 is used as the basis for this separation. The dynamic capacitance reference value calculated by S2 Perform a subtraction operation, and the difference value is the insulation degradation residual signal. The residual signal has theoretically eliminated the influence of operating condition coupling, and its trend change can more directly characterize the early aging of insulation.

[0060] S4: Construction of Adaptive Dynamic Detection Threshold

[0061] A judgment threshold that can dynamically adjust with the degree of change in operating conditions is established to ensure high detection sensitivity when operating conditions are stable, while avoiding misjudgments due to transient errors in the model when operating conditions change drastically. In this embodiment, an adaptive dynamic detection threshold is constructed based on the operating condition change information in the real-time data of multiple physical quantities generated by S1. ;

[0062] S5: Determination of Insulation Status

[0063] The clean degradation signal is compared with a dynamic judgment threshold to make a final insulation state judgment. In this embodiment, the insulation degradation residual signal is continuously judged. Is the absolute value greater than the adaptive dynamic detection threshold constructed by S4? ,like If the abnormality is confirmed, it is preliminarily determined that an insulation degradation-related anomaly may have occurred; otherwise, it is determined to be a normal operating state. When the abnormal state continues to meet specific conditions, an insulation degradation event is finally confirmed.

[0064] S6: Generation of Insulation Condition Assessment

[0065] This approach transforms discrete degradation events into a quantitative, long-term assessment of insulation health status and provides predictive maintenance decision support. In this embodiment, the system updates a quantified insulation degradation index in response to the insulation degradation event determined in step S5. Based on the current value and historical trend of the index, a comprehensive insulation condition assessment report is generated, which includes the current health level, degradation rate and remaining service life prediction.

[0066] This embodiment fundamentally solves the contradiction between detection sensitivity and robustness in existing static models under dynamic conditions by introducing a dynamic benchmark model coupled with the real-time operating conditions of the motor and an adaptive detection threshold. It can effectively filter out normal capacitance fluctuations caused by drastic changes in temperature, load, and speed, thereby accurately capturing weak signals caused only by irreversible aging of the insulation material itself. This enables online, high-precision detection and evaluation of early insulation degradation of the motor stator winding, greatly improving the reliability and predictability of motor health management. Example 2

[0067] S2 specifically includes:

[0068] Extract real-time winding temperature, motor load current, and motor rotor speed from real-time data of multiple physical quantities;

[0069] The root mean square current of the motor load is calculated to obtain the effective value of the current.

[0070] The real-time winding temperature, effective current value, and motor rotor speed are input into the operating condition coupled dynamic reference model to calculate the dynamic capacitance reference value.

[0071] The operating condition coupled dynamic reference model decomposes the change in the equivalent capacitance of the winding into a linear superposition of the thermal expansion effect caused by temperature, the electromagnetic pressure effect caused by the load current, and the mechanical stress effect caused by the rotor speed.

[0072] Based on Example 1, this embodiment provides specific limitations on the working condition coupled dynamic benchmark model and its calculation process in S2;

[0073] S2 specifically includes extracting specific operating condition variables required for benchmark value calculation from real-time data of multiple physical quantities, namely, the real-time winding temperature. Motor load current and motor rotor speed ;

[0074] Due to the motor load current The signal is usually an alternating current (AC) signal, and the magnitude of the electromagnetic stress it generates is related to the effective value of the current. Therefore, it is necessary to perform root mean square (RMS) calculation on the motor load current to obtain the effective value of the current. ;

[0075] Real-time winding temperature RMS value of current and motor rotor speed These three processed operating condition variables are fed as inputs into the operating condition coupled dynamic reference model to calculate the dynamic capacitance reference value at the current moment. ;

[0076] In this embodiment, the construction of the operating condition coupled dynamic reference model follows a clear physical mechanism, and its core idea is to represent the change in the equivalent capacitance of the winding. The model decomposes into a linear superposition of three main physical effects: thermal expansion caused by temperature, electromagnetic pressure caused by load current, and mechanical stress caused by rotor speed. This model assumes no significant interaction between these physical effects and provides a sufficiently accurate approximation for most operating conditions. The model can be expressed by the following formula:

[0077]

[0078] in:

[0079] For a moment The dynamic capacitance reference value is the output of the model, and its unit is pF;

[0080] The reference capacitance value under reference operating conditions represents the basic capacitance of the motor under healthy, static, and standard temperature conditions. It is obtained through a one-time calibration experiment on a healthy motor and is measured in pF.

[0081] For reference temperature, it is usually taken as 25℃;

[0082] This is the root mean square value of the motor load current, reflecting the load magnitude, obtained from the data collected in previous steps. Calculated in real time, unit is A;

[0083] The temperature coupling coefficient characterizes the intensity of the thermal expansion effect on the capacitance. It is obtained through calibration experiments and its unit is pF / ℃.

[0084] is the load coupling coefficient, which characterizes the intensity of the electromagnetic pressure effect on the capacitor. It is obtained through calibration experiments and its unit is pF / A².

[0085] is the rotational speed coupling coefficient, which characterizes the intensity of the mechanical stress effect on the capacitor. It is obtained through calibration experiments and its unit is pF / RPM².

[0086] All key parameters in the model were obtained through a one-time full-condition characteristic calibration experiment on a confirmed healthy motor of the same model. In this experiment, the healthy motor was driven through various combinations of operating conditions within its design operating range, and complete records were simultaneously recorded. The dataset is used to identify parameters in the above formula using a multiple linear regression algorithm, solving for a set of optimal coefficients that minimize the sum of squared errors between the model's predicted values ​​and the actual measured values. These coefficients are then fixed as the health status baseline parameters for this model of motor, used for subsequent online evaluation of all motors of the same model. To address individual motor differences or long-term parameter drift, the system also supports online fine-tuning of the baseline parameters. For example, during the maintenance window confirming the motor's healthy condition, the system can automatically collect stable operating condition data over a period of time and use online parameter identification algorithms such as the least squares method to adjust the initial calibration parameters (…). Make minor adjustments to minimize the mean of the residual signal in the current state, thereby achieving adaptive model updates and ensuring the accuracy of long-term evaluations;

[0087] It should be noted that this model adopts the assumption of linear superposition, which is an engineering approximation based on the fact that there is no significant interactive coupling between the physical effects under most operating conditions. For special motors operating under extreme conditions (such as simultaneous operation of extremely high temperatures and extreme loads), in order to pursue higher accuracy, cross-coupling terms can be introduced into this model, for example:

[0088]

[0089] in: The temperature-force coupling coefficient is obtained through more complex calibration experiments. However, in general applications, the current model achieves a good balance between accuracy and complexity. Example 3

[0090] S4 specifically includes:

[0091] Extract motor load current and winding real-time temperature from real-time data of multiple physical quantities;

[0092] The motor load current and the real-time winding temperature are smoothed and numerically differentiated to obtain a smooth load change rate and temperature change rate.

[0093] An adaptive dynamic detection threshold is constructed by combining a preset basic detection threshold, a smooth load change rate, a temperature change rate, and a preset robustness adjustment coefficient.

[0094] This embodiment, based on embodiment 1, specifically defines the process of constructing the adaptive dynamic detection threshold in S4;

[0095] S4 specifically includes extracting the two variables most relevant to the degree of drastic change in operating conditions from real-time data of multiple physical quantities: motor load current. Real-time temperature of windings ;

[0096] To accurately quantify the rate of change in operating conditions and avoid interference from sensor noise in the derivative calculation, smoothing and numerical differentiation of the motor load current and real-time winding temperature are required. In this embodiment, a Savitzky-Golay filter can be used. This filter can fit the data using a polynomial within a sliding window and analytically calculate the derivative, thereby simultaneously achieving smoothing and differentiation to obtain a smooth load change rate. With temperature change rate ;

[0097] Set a preset base detection threshold The smoothed load change rate calculated above Temperature change rate and a set of preset robustness adjustment coefficients ( By combining these methods, an adaptive dynamic detection threshold can be constructed. The calculation formula is as follows:

[0098]

[0099] in:

[0100] For a moment The dynamic detection threshold is the decision threshold, and its unit is pF;

[0101] The basic detection threshold represents the detection sensitivity under absolutely stable operating conditions. In the calibration experiment, a data segment from a long period of stable motor operation is selected, and the residual signal is calculated. Standard deviation ,set up This is used to cover the vast majority of random noise, where k is the coverage coefficient, which is usually set based on experience or statistical principles. For example, when k=3, it can cover 99.7% of normal noise fluctuations, and the unit is pF.

[0102] The smoothed load current and temperature change rates are calculated from the previous steps, with units of A / s and ℃ / s, respectively.

[0103] κI and κT are the load robustness adjustment coefficient and temperature robustness adjustment coefficient, respectively, with units of s / A and s / ℃. They are used to adjust the sensitivity of the threshold to the rate of change of different operating conditions. In the calibration experiment, the data segment with the most drastic change in operating conditions is selected, and the transient peak value generated by the residual signal at this time is observed for calibration. The value of makes the calculated value at these worst transient points... It can just contain that transient peak.

[0104] The determination steps of S5 further include:

[0105] Before determining whether an insulation degradation event has occurred, it is also necessary to determine whether the duration of the event has reached the preset anti-jitter time window;

[0106] The insulation degradation event is finally confirmed and step S6 is triggered only when the duration reaches the anti-jitter time window;

[0107] This embodiment adds a time confirmation step to the determination logic in S5 of embodiment 1 to further improve the reliability of the determination result.

[0108] Specifically, in determining Therefore, it is preliminarily determined that after an insulation degradation event occurs, this embodiment will not immediately trigger the subsequent S6. Instead, it is necessary to determine whether the duration of the event reaches the preset anti-jitter time window. ;

[0109] It is a minimum duration set to filter out random pulse interference or sensor glitches. Its function is to ensure that the judged event has a certain time stability rather than instantaneous noise. It is set based on experience or statistical analysis of the system noise characteristics. For example, in this embodiment, it can be set to 5 seconds.

[0110] Only when the duration reaches the anti-jitter time window At that time, that is The state was maintained for at least 5 seconds, and the system finally confirmed that an insulation degradation event had occurred and triggered step S6.

[0111] S6 specifically includes:

[0112] Set the insulation degradation index to an initial value of zero;

[0113] Whenever an insulation degradation event is received, the insulation degradation index is accumulated according to the severity of the event;

[0114] The current value of the insulation degradation index is mapped to the health status level, and its historical data is used to perform slope analysis to calculate the degradation rate, and the remaining service life before reaching the preset failure threshold is extrapolated.

[0115] Correlation analysis is performed between insulation degradation events and the corresponding operating data at the time of their occurrence to identify specific operating modes that accelerate insulation degradation, and operational optimization strategy recommendations are generated accordingly.

[0116] This embodiment refines the insulation state assessment generation process in step S6 based on embodiment 1, and adds in-depth analysis and decision support functions based on historical data.

[0117] S6 specifically includes: maintaining a quantified insulation degradation index internally within the system. The purpose of this index is to accumulate discrete, instantaneous degradation events into a continuous indicator that can macroscopically characterize the overall health of the insulation. The insulation degradation index is set with an initial value of zero, meaning that for a newly commissioned or healthy motor... ,

[0118] During operation, whenever an insulation degradation event confirmed in step S5 is received, the system accumulates the insulation degradation index according to the severity of the event. The severity of the event is quantified by calculating the integral of the absolute value of the residual signal exceeding the dynamic threshold. The specific accumulation formula is as follows: ,in This represents the duration of the insulation degradation event. Insulation Degradation Index The unit is pF·s;

[0119] Current value of insulation degradation index The system will map these values ​​to a preset health status level, such as: Healthy [0-100], Attention [101-300], Warning [301-700], Danger [>700], providing maintenance personnel with an intuitive understanding of the status; the system will... Historical data were used to perform slope analysis to calculate the degradation rate, i.e. Based on the current index value and degradation rate, it is extrapolated that the index will reach a preset failure threshold. The remaining time, i.e., the remaining service life, is the failure threshold. It can be determined based on the destructive aging test data of the same model of motor, or set according to the definition of the hazard level of insulation condition in relevant industry standards;

[0120] To further uncover the underlying causes of degradation events, this embodiment further includes: performing correlation analysis between insulation degradation events and the corresponding operating condition data at the time of occurrence; specifically, the system records the timestamp, severity, and complete snapshots of operating condition data such as winding temperature, load current, and rotor speed at the time of each degradation event. Through statistical analysis of a large number of events and their accompanying operating condition data, the system can identify specific operating modes that accelerate insulation degradation; for example, it may be found that the combination of high ambient temperature and frequent heavy-load start-stop is strongly correlated with the frequency and severity of insulation degradation events.

[0121] It also generates operational optimization strategy suggestions. Based on the specific operating modes identified above, the system can automatically generate executable maintenance or operation suggestions. For example, it can prompt users that when the load exceeds 80% and rapid acceleration or deceleration is performed, the insulation degradation rate is significantly accelerated. It is recommended to optimize the process flow to reduce the occurrence of such operating conditions, thereby realizing intelligent maintenance from passive response to proactive prevention. Example 4

[0122] A digital model-based system for assessing the insulation condition of motor stator windings includes:

[0123] The multi-physical quantity real-time data acquisition module is used to acquire the high-frequency equivalent capacitance measurement value of the motor stator winding, the real-time temperature of the winding, the motor load current and the motor rotor speed, and generate multi-physical quantity real-time data.

[0124] The operating condition coupled dynamic reference model module is used to calculate the dynamic capacitance reference value based on real-time data of multiple physical quantities.

[0125] The insulation degradation residual signal extraction module is used to perform a difference operation between the high-frequency equivalent capacitance measurement value and the dynamic capacitance reference value to extract the insulation degradation residual signal.

[0126] The adaptive threshold and state assessment module is used to construct an adaptive dynamic detection threshold based on real-time data of multiple physical quantities, and to determine the insulation state based on the comparison results between the insulation degradation residual signal and the adaptive dynamic detection threshold.

[0127] The degradation trend prediction and condition-based maintenance module is used to update the insulation degradation index and generate an insulation status assessment in response to the determination result of insulation degradation events. This embodiment provides a digital model-based motor stator winding insulation status assessment system, which is configured to execute the assessment method in the aforementioned embodiment. The system can be integrated into the motor controller or used as a standalone edge computing device. Its internal logic function modules include:

[0128] Multi-physical quantity real-time data acquisition module: Provides basic data for the system. In this embodiment, this module is connected to the high-frequency capacitance sensor, temperature sensor, current transformer and speed encoder installed on the motor. It is used to acquire the high-frequency equivalent capacitance measurement value of the motor stator winding, the real-time temperature of the winding, the motor load current and the motor rotor speed, and generate multi-physical quantity real-time data. This module is responsible for the synchronous acquisition of data, timestamp marking and preliminary signal conditioning.

[0129] Operating Condition Coupled Dynamic Reference Model Module: This module calculates the theoretical capacitance reference of healthy insulation under the current operating conditions. In this embodiment, it is a module that solidifies the aforementioned formula. and calibration parameters ( The software or firmware unit is used to receive real-time data of multiple physical quantities from the data acquisition module and calculate the dynamic capacitance reference value based on the real-time data of multiple physical quantities. Its specific execution logic is completely consistent with the implementation of Embodiment 2.

[0130] Insulation degradation residual signal extraction module: Separates the pure degradation characterization signal. In this embodiment, the module is configured to receive the high-frequency equivalent capacitance measurement value and the dynamic capacitance reference value, and extracts the insulation degradation residual signal by performing a difference operation on the high-frequency equivalent capacitance measurement value and the dynamic capacitance reference value through a difference operation unit.

[0131] The adaptive threshold and state assessment module is designed to execute the core anomaly determination logic. In this embodiment, the module is configured to receive real-time data of multiple physical quantities and insulation degradation residual signals. It contains two sub-functions: first, to construct an adaptive dynamic detection threshold based on the real-time data of multiple physical quantities, the execution logic of which is consistent with that in the implementation of embodiment 4; second, to determine the insulation state based on the comparison result between the insulation degradation residual signal and the adaptive dynamic detection threshold, the determination logic of which includes the anti-jitter time window mechanism in the implementation of embodiment 5.

[0132] The degradation trend prediction and condition-based maintenance module aims to transform the underlying judgment results into high-level health assessment and decision support. In this embodiment, this module is configured to update the insulation degradation index and generate an insulation status assessment in response to the insulation degradation event judgment results output by the adaptive threshold and status assessment module. This module specifically implements all the functions in the implementation of embodiment 5, including the accumulation of degradation index, health status mapping, degradation rate and RUL calculation, as well as correlation analysis to generate operation optimization strategy suggestions.

[0133] The system in this embodiment, through the organic combination and collaborative work of the above modules, forms a complete technical closed loop from bottom-level data acquisition to top-level decision support. As an independent entity, it can automatically, online, and in real time execute all the aforementioned methods and steps, providing a powerful, reliable, and intelligent insulation health management solution for motors in industrial sites.

[0134] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention; any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0135] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for evaluating the insulation status of motor stator windings based on digital models, characterized in that, Includes the following steps: S1. Collect the high-frequency equivalent capacitance measurement value of the motor stator winding, the real-time temperature of the winding, the motor load current and the motor rotor speed, and generate real-time data of multiple physical quantities. S2. Based on the real-time data of the multiple physical quantities, calculate the dynamic capacitance reference value through the working condition coupled dynamic reference model; S3. Perform a subtraction operation between the measured value of the high-frequency equivalent capacitance and the reference value of the dynamic capacitance to extract the insulation degradation residual signal; S4. Based on the real-time data of the multiple physical quantities, construct an adaptive dynamic detection threshold; S5. Determine whether the absolute value of the insulation degradation residual signal is greater than the adaptive dynamic detection threshold: if yes, determine that an insulation degradation event has occurred; if no, determine that it is in normal operation. S6. In response to the insulation degradation event, update the insulation degradation index and generate an insulation status assessment based on the updated insulation degradation index.

2. The method for evaluating the insulation status of motor stator windings based on a digital model according to claim 1, characterized in that, S2 specifically includes: The real-time temperature of the winding, the motor load current, and the motor rotor speed are extracted from the real-time data of the multiple physical quantities. The root mean square (RMS) value of the motor load current is calculated to obtain the effective value of the current. The real-time temperature of the winding, the effective value of the current, and the rotor speed of the motor are input into the operating condition coupled dynamic reference model to calculate the dynamic capacitance reference value.

3. The method for evaluating the insulation status of motor stator windings based on a digital model according to claim 2, characterized in that, The operating condition coupled dynamic reference model decomposes the change in the equivalent capacitance of the winding into a linear superposition of the thermal expansion effect caused by temperature, the electromagnetic pressure effect caused by the load current, and the mechanical stress effect caused by the rotor speed.

4. The method for evaluating the insulation status of motor stator windings based on a digital model according to claim 1, characterized in that, S4 specifically includes: Extract the motor load current and the real-time winding temperature from the real-time data of the multiple physical quantities; The load current of the motor and the real-time temperature of the winding are smoothed and numerically differentiated to obtain a smooth load change rate and temperature change rate. The adaptive dynamic detection threshold is constructed by combining the preset basic detection threshold, the smooth load change rate, the temperature change rate, and the preset robustness adjustment coefficient.

5. The method for evaluating the insulation status of motor stator windings based on a digital model according to claim 1, characterized in that, The determination step of S5 further includes: Before determining whether an insulation degradation event has occurred, it is also necessary to determine whether the duration of the event has reached the preset anti-jitter time window; The insulation degradation event is finally confirmed and step S6 is triggered only when the duration reaches the anti-jitter time window.

6. The method for evaluating the insulation status of motor stator windings based on a digital model according to claim 1, characterized in that, S6 specifically includes: The insulation degradation index is set to an initial value of zero; Whenever an insulation degradation event is received, the insulation degradation index is accumulated according to the severity of the event; The current value of the insulation degradation index is mapped to the health status level, and slope analysis is performed on its historical data to calculate the degradation rate. Finally, the remaining service life when the preset failure threshold is reached is extrapolated.

7. The method for evaluating the insulation status of motor stator windings based on a digital model according to claim 6, characterized in that, Further includes: The insulation degradation events are correlated with the corresponding operating data at the time of occurrence to identify specific operating modes that accelerate insulation degradation, and operational optimization strategy recommendations are generated accordingly.

8. A digital model-based motor stator winding insulation condition assessment system, applied to the digital model-based motor stator winding insulation condition assessment method according to any one of claims 1-7, characterized in that, include: The multi-physical quantity real-time data acquisition module is used to acquire the high-frequency equivalent capacitance measurement value of the motor stator winding, the real-time temperature of the winding, the motor load current and the motor rotor speed, and generate multi-physical quantity real-time data. The working condition coupled dynamic reference model module is used to calculate the dynamic capacitance reference value based on the real-time data of the multiple physical quantities. An insulation degradation residual signal extraction module is used to perform a subtraction operation between the high-frequency equivalent capacitance measurement value and the dynamic capacitance reference value to extract the insulation degradation residual signal. An adaptive threshold and state assessment module is used to construct an adaptive dynamic detection threshold based on the real-time data of the multiple physical quantities, and to determine the insulation state based on the comparison result between the insulation degradation residual signal and the adaptive dynamic detection threshold. The degradation trend prediction and condition-based maintenance module is used to update the insulation degradation index and generate an insulation status assessment in response to the determination results of insulation degradation events.

Images