An industrial digital management method and system based on model construction

Abstract

The present invention relates to the technical field of industrial equipment management, and in particular to a model-based industrial digital management method and system. The method comprises the following steps: obtaining an industrial equipment operation log; performing a motor thermal overload analysis based on the industrial equipment operation log to obtain motor thermal overload data; performing a thermal isolation design based on the motor thermal overload data to obtain thermal isolation data; performing a ventilation path blockage analysis based on the motor thermal overload data to obtain ventilation path blockage data; performing a winding insulation aging test based on the ventilation path blockage data to obtain winding insulation aging data; performing an insulation paint film shedding test based on the winding insulation aging data to obtain insulation paint film shedding data; and performing a protective coating upgrade based on the insulation paint film shedding data to obtain protective coating upgrade data. The present invention improves the intelligent rate of industrial equipment protection status identification and operation and maintenance decision-making based on industrial equipment management technology.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial equipment management, and in particular to an industrial digital management method and system based on model building. Background Art

[0002] Traditional industrial digital management employs a unified, integrated processing mechanism for multi-source, heterogeneous data from industrial equipment, often relying on a single type of sensor or log data for state analysis. This results in low accuracy in identifying complex equipment operating states. It lacks causal reasoning-driven parameter correlation modeling, making it impossible to organically link multi-dimensional influencing factors such as thermal overload, ventilation blockage, and insulation aging across both temporal and spatial dimensions. This limits the comprehensive tracing and prediction of equipment aging paths and fault evolution chains. In thermal isolation design and protective coating upgrades, static models or expert rule-of-thumbnail models often are used, lacking the ability to adaptively adjust to the actual operating environment and material physical properties. This results in significant deviations between design results and actual operating conditions. In processes such as ventilation blockage detection, insulation paint crack identification, and adhesion degradation analysis, manual intervention or reliance on static image analysis are often employed, lacking high-frequency sampling and dynamic simulation mechanisms, resulting in delayed detection responses and limited identification accuracy. In the coating aging simulation and digital management of protective capabilities, a numerical simulation platform based on the linkage of coating material parameters, motor power characteristics, and environmental variables is lacking, making it impossible to dynamically reflect coating performance changes, making it difficult to implement graded early warnings for protective capabilities and precise operational and maintenance decisions. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide an industrial digital management method and system based on model building to solve at least one of the above technical problems.

[0004] To achieve the above objectives, a model-based industrial digital management method includes the following steps:

[0005] Step S1: Obtaining an industrial equipment operation log; performing a motor thermal overload analysis based on the industrial equipment operation log to obtain motor thermal overload data; performing a thermal isolation design based on the motor thermal overload data to obtain thermal isolation data;

[0006] Step S2: performing ventilation path blockage analysis based on the motor thermal overload data to obtain ventilation path blockage data; performing winding insulation aging detection based on the ventilation path blockage data to obtain winding insulation aging data; performing insulation paint film shedding detection based on the winding insulation aging data to obtain insulation paint film shedding data;

[0007] Step S3: upgrading the protective coating according to the insulating paint film shedding data to obtain protective coating upgrade data; constructing a motor coating thermal insulation model according to the protective coating upgrade data and the thermal insulation data; performing a coating aging simulation according to the motor coating thermal insulation model to obtain coating aging data;

[0008] Step S4: Perform coating hollowing detection based on the coating aging data to obtain coating hollowing data; perform coating adhesion degradation analysis based on the coating hollowing data to obtain coating adhesion degradation data; perform digital management of protective capability based on the coating adhesion degradation data to obtain digital management data of protective capability.

[0009] The present invention has achieved technological breakthroughs in multiple key links that are difficult to achieve with traditional technologies by constructing a full-link industrial digital processing flow covering "operation log acquisition - status diagnosis - material analysis - structural modeling - dynamic simulation - hierarchical management". By integrating data of different dimensions such as motor thermal overload, ventilation path blockage, insulation paint film peeling, coating hollowing, adhesion degradation, etc., the problem of low state recognition accuracy due to single data in the existing technology is solved; a causal relationship is established between the equipment operation status and the material degradation behavior, and a time-continuous aging evolution chain is constructed, which significantly improves the ability to track and predict potential failure paths of the equipment; a thermal isolation design method based on thermal load and material thermal conductivity parameters is introduced to replace the static rule base, thereby improving the adaptability of the thermal management module to actual operating conditions; dynamic simulation of ventilation resistance and high-frequency detection methods such as infrared thermal imaging are used to realize dynamic identification of winding insulation aging and paint film peeling, effectively improving the response speed and accuracy of detection; during the upgrade of the protective coating A dual-factor coupling method of thermal isolation data and coating material parameters is introduced in the process to significantly enhance the simulation accuracy of material behavior; coating aging behavior simulation is carried out based on a three-dimensional numerical thermal simulation platform, breaking the limitation of traditional reliance on manual experience to estimate aging life; a multimodal data fusion and discrimination strategy is introduced in the coating hollowing detection link to improve the hollowing recognition accuracy and eliminate interference responses; a regional distribution differentiation evaluation method is adopted in the adhesion degradation analysis to refine the local protection performance differences; finally, a hierarchical indicator system is constructed in the digital management of protection capabilities, and risk level division is realized based on core data such as adhesion, safety threshold and thermal field distribution parameters, providing reliable data support for intelligent decision-making, thereby realizing intelligent evaluation and closed-loop control of the protection status of industrial equipment.

[0010] Preferably, this specification also provides an industrial digital management system based on model building, which is used to execute the industrial digital management method based on model building as described above. The industrial digital management system based on model building includes:

[0011] Thermal isolation design module, used to obtain industrial equipment operation logs; perform motor thermal overload analysis based on the industrial equipment operation logs to obtain motor thermal overload data; perform thermal isolation design based on the motor thermal overload data to obtain thermal isolation data;

[0012] The insulation aging detection module is used to analyze ventilation path blockage based on motor thermal overload data to obtain ventilation path blockage data; perform winding insulation aging detection based on ventilation path blockage data to obtain winding insulation aging data; and perform insulation paint film shedding detection based on winding insulation aging data to obtain insulation paint film shedding data;

[0013] The motor coating thermal insulation model construction module is used to upgrade the protective coating based on the insulation paint film shedding data to obtain the protective coating upgrade data; construct the motor coating thermal insulation model based on the protective coating upgrade data and thermal insulation data; and perform coating aging simulation based on the motor coating thermal insulation model to obtain coating aging data;

[0014] The digital management module is used to perform coating hollowing detection based on coating aging data to obtain coating hollowing data; perform coating adhesion degradation analysis based on coating hollowing data to obtain coating adhesion degradation data; and perform digital management of protective capabilities based on coating adhesion degradation data to obtain digital management data of protective capabilities.

[0015] The industrial digital management system based on model construction of the present invention can implement any one of the industrial digital management methods based on model construction of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the industrial digital management method based on model construction. The internal modules of the system cooperate with each other to improve the intelligence rate of industrial equipment protection status identification and operation and maintenance decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0017] Figure 1 A schematic diagram of the steps of a model-based industrial digital management method of the present invention;

[0018] Figure 2 Detailed step flow diagram of step S1 in the present invention;

[0019] Figure 3 Detailed flowchart of step S16 in the present invention;

[0020] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0021] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative work are within the scope of protection of the present invention.

[0022] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0023] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0024] To achieve this, please refer to Figures 1 to 3 The present invention provides an industrial digital management method based on model building, the method comprising the following steps:

[0025] Step S1: Obtaining an industrial equipment operation log; performing a motor thermal overload analysis based on the industrial equipment operation log to obtain motor thermal overload data; performing a thermal isolation design based on the motor thermal overload data to obtain thermal isolation data;

[0026] In this embodiment, an on-site industrial control system is used to collect industrial equipment operation log data. The log content includes motor operating current (unit: A), voltage (unit: V), winding temperature (unit: ° C), operating time (unit: s), and load factor (percentage). The OPC UA protocol is used for real-time data collection, and the data sampling frequency is set to 1Hz. A thermal overload judgment standard is set. When the motor winding temperature is continuously greater than the set threshold of 135°C for more than 600 seconds, and the current exceeds 110% of the rated current (based on the motor nameplate rated current), a thermal overload event is recorded. All thermal overload events are numbered, categorized, and archived, and the frequency and duration of thermal overload events for each motor are counted weekly.

[0027] To design thermal isolation based on thermal overload data, we used ANSYS Icepak, a thermal field finite element analysis tool, to create a 3D thermal simulation model within the motor housing structure. We input the heat source location (set in the stator winding area), set the ambient temperature to 40°C, set the air-cooled convection heat transfer coefficient to 18 W / m²·K, and used a ceramic fiber layer with a thickness of 2 mm and a thermal conductivity of 0.04 W / (m·K) as the thermal isolation material. Thermal simulation compared the surface temperature rise with and without the thermal isolation layer, generating thermal isolation data including heat flux density maps, temperature gradient maps, and the locations of the highest temperature nodes.

[0028] Step S2: performing ventilation path blockage analysis based on the motor thermal overload data to obtain ventilation path blockage data; performing winding insulation aging detection based on the ventilation path blockage data to obtain winding insulation aging data; performing insulation paint film shedding detection based on the winding insulation aging data to obtain insulation paint film shedding data;

[0029] In this embodiment, based on the thermal overload data obtained in step S1, the air duct pressure difference value (unit: Pa) when the motor is running is extracted and collected by the pressure difference sensors installed at the air inlet and outlet, and the sampling frequency is set to 0.5Hz. The ventilation path blockage threshold is set to record a blockage event when the pressure difference is continuously less than 30Pa and the temperature rise is greater than 120°C. All blockage events are recorded synchronously with their corresponding timestamps. The blockage event data is input into the thermal aging analysis module of the stator winding. According to the IEC 60216 standard, the temperature data and operating time data are input to calculate the thermal aging index TI (Temperature Index). When the TI is lower than 105, it is considered that the insulation has begun to age. The insulation resistance of the motor winding insulation layer is tested. The phase-to-phase insulation resistance is measured using a megohmmeter. A DC 1000V voltage is applied and the resistance value is read. If the resistance value is lower than 2MΩ, it is identified as a serious insulation aging area. Continuing to scan the insulation surface based on areas of insulation aging, infrared thermal imaging equipment (320x240 resolution, temperature range: -20°C to 250°C) was used to identify areas with abnormal surface temperature differences. For areas with temperature differences greater than 8°C, a high-definition industrial camera (5 megapixel) with an image recognition algorithm was used to identify paint film detachment locations using Canny edge detection and grayscale thresholding. The detachment data was then recorded as annotated 2D images, with the image coordinate system corresponding to the motor winding locations.

[0030] Step S3: upgrading the protective coating according to the insulating paint film shedding data to obtain protective coating upgrade data; constructing a motor coating thermal insulation model according to the protective coating upgrade data and the thermal insulation data; performing a coating aging simulation according to the motor coating thermal insulation model to obtain coating aging data;

[0031] In this example, based on the paint film peeling data identified in step S2, a CAD 3D modeling tool was used to reconstruct the surface of the exposed parts of the motor. According to the protective coating thickness standards defined in the industrial standard JB / T 10918, the thickness of the first primer layer was set to 25 μm, the second insulating varnish layer to 40 μm, and the third anti-corrosion topcoat layer to 30 μm, for a total coating thickness of 95 μm. The protective coating upgrade solution used a polyesterimide insulating varnish with a thermal decomposition temperature set at 220°C. Based on the upgraded coating parameters and the thermal insulation data obtained in step S1, a thermal flow model of the combined coating and thermal insulation structure was constructed. This modeling was performed using the COMSOL Multiphysics Thermal Module, with inputs of the coating thermal conductivity (0.12 W / m·K), surface emissivity (0.92), and ambient wind speed (0.5 m / s). Natural convection was used as the boundary condition. The model solved the heat conduction equation and calculated the temperature field distribution at different locations under long-term operation. The simulation time period was set to 3600 seconds. Based on the aforementioned thermal model, a coating aging lifespan simulation module was constructed using the MATLAB programming language. Input variables included operating temperature, operating cycle, UV radiation intensity (set to a cumulative annual dose of 200 MJ / m²), and ambient humidity (80% RH). The aging process was simulated according to the Arrhenius aging acceleration equation, with the activation energy parameter set to 90 kJ / mol. The aging rate constant, k, was calculated under different environmental conditions, resulting in coating aging data, including the estimated aging lifespan, coating delamination probability curves, and distribution maps.

[0032] Step S4: Perform coating hollowing detection based on the coating aging data to obtain coating hollowing data; perform coating adhesion degradation analysis based on the coating hollowing data to obtain coating adhesion degradation data; perform digital management of protective capability based on the coating adhesion degradation data to obtain digital management data of protective capability.

[0033] In this embodiment, based on the coating aging data obtained in step S3, a handheld laser vibration detector (frequency range 20Hz to 10kHz, measurement accuracy ±0.5Hz) is used to perform a tapping response test on the surface of the area predicted to be severely aged. The response frequency and attenuation rate are recorded, and the actual response is compared with the ideal solid-state coating frequency response. If the peak frequency drops by more than 30%, it is marked as a coating hollowing area. The hollowing detection results are coordinate-marked, and based on the distribution of the hollowing area, an adhesion detection device (pull-off adhesion tester, standard ASTM D4541) is further used to perform a fixed-point adhesion test on the same area. During the test, a test block is glued to the surface to be tested, and a hydraulic system is used to apply force, and the maximum peeling force is recorded. If it is less than 1.5MPa, it is considered to be an adhesion deterioration area. The test results are summarized to form adhesion degradation data, including position, pull-off strength value and test time. Based on the adhesion degradation data, a coating protection capability dataset is constructed. A database management module written in Python standardized the aging age, hollowing frequency, and adhesion indicators for different regions, using the Z-score method for data normalization. The normalized results were entered into the digital protection capability management platform and stored in a MySQL database. Fields included coating location number, adhesion level (level 1 to level 5), estimated peeling time (months), hollowing frequency, and recommended maintenance intervals, forming a complete digital protection capability management data structure.

[0034] Preferably, step S1 is specifically as follows:

[0035] Step S11: Obtaining industrial equipment operation logs;

[0036] In this embodiment, the operation log of the industrial motor equipment is exported from the embedded industrial control unit. The log record sources include the industrial site SCADA system and the PLC controller, and the sampling frequency is set to 1Hz. The operation log contains a timestamp, equipment number, operation status identification, real-time current, voltage, speed, operation time, and fault identification code. Use a structured data format (such as CSV) to import it into the log processing platform, for example, use Apache Kafka for streaming data access, and Apache Flink for parsing and filtering, and select the operation logs corresponding to all motor numbers. The filtering rule is to eliminate data with an operation status identification of "standby" or "fault shutdown", and only retain the data corresponding to the "running" status identification. The output result is a pure motor operation log data set, which serves as the basis for subsequent data calculation and parameter extraction.

[0037] Step S12: Recording current operating parameters based on the industrial equipment operation log; recording voltage operating parameters based on the industrial equipment operation log; and calculating motor power based on the current operating parameters and the voltage operating parameters;

[0038] In this example, we use the operation log data and call a time series analysis tool (such as the Python pandas library) to extract the current and voltage operating parameters. The current fields are named "Ia", "Ib", and "Ic", and the voltage fields are named "Ua", "Ub", and "Uc". The root mean square (RMS) values ​​of the three-phase current and voltage are calculated using the following formula: , . Calculate the motor power P using the active power calculation formula: , where the power factor The default value is 0.85 based on the motor nameplate parameters. All calculations are processed line by line through a data processing script, and the real-time power value is output once every second in kilowatts (kW), forming power time series data.

[0039] Step S13: Estimating the winding temperature rise based on the motor power to obtain motor winding temperature data;

[0040] In this embodiment, the power time series data is used to estimate the winding temperature rise according to the energy conservation principle of the motor thermal model. The calculation formula is: , where P is the real-time power (kW), η is the motor thermal conversion coefficient, set to 0.92, t is the unit time period (seconds), m is the motor winding mass, set to 7.5 kg, and c is the specific heat capacity of copper, fixed at 0.385 kJ / kg·K. The calculated output is the motor winding temperature rise, ΔT, per second. ΔT is added to the initial winding temperature (set to an ambient temperature of 25°C) to form a motor winding temperature data series. To prevent error accumulation, temperature rise balancing is performed every 60 seconds, using an exponential sliding average to correct for abnormal peaks.

[0041] Step S14: performing a high temperature determination on the motor winding temperature data according to a preset motor winding temperature threshold to obtain motor winding high temperature data; calculating a high temperature duration based on the motor winding high temperature data to obtain high temperature duration data;

[0042] In this embodiment, the high temperature judgment threshold of the motor winding temperature is set to 125°C, which is based on the maximum operating temperature standard supported by the motor winding insulation grade F. The motor winding temperature data in S13 is judged second by second, and the time point when the temperature is greater than or equal to 125°C is marked as a high temperature state. The start and end time of the high temperature state are recorded, and the high temperature duration is calculated in seconds. For example, when the continuous high temperature state occurs from 10:01:00 to 10:06:20, the high temperature duration is 320 seconds. All high temperature states are continuously recorded to generate a high temperature duration data list in the format of: [start time, end time, duration]. The temperature data under all high temperature states are saved separately to form a motor winding high temperature data set.

[0043] Step S15: Counting the high temperature frequency based on the high temperature data of the motor winding to obtain high frequency and high temperature period data; performing thermal overload identification based on the high frequency and high temperature period data and the high temperature duration data to obtain motor thermal overload data;

[0044] In this embodiment, in the high-temperature data set of the motor winding output by S14, the frequency of high-temperature events in each 24-hour period is counted. The high-temperature frequency recognition threshold is set to 6 times / 24 hours, that is, if the high-temperature event occurs more than 6 times in any continuous 24 hours, it is marked as a high-frequency high-temperature period. Combined with the duration of each high temperature in S14, the total high-temperature duration in each high-frequency high-temperature period is calculated. The thermal overload recognition threshold is set to a single-segment continuous high-temperature time of more than 180 seconds, and the high-temperature segment is located in the high-frequency high-temperature period, which is determined to be a thermal overload event. Mark each time period that meets the conditions, and record the corresponding start and end time, high-temperature value, duration, etc., and output the motor thermal overload data table. The table structure includes: [start time, end time, maximum temperature, high-temperature duration, overload mark].

[0045] Step S16: Perform thermal isolation design based on the motor thermal overload data to obtain thermal isolation data.

[0046] In this embodiment, thermal isolation design is performed using the motor operating structure drawings based on thermal overload data. The heat transfer path and material thermal conductivity of the parts corresponding to the overload point are evaluated for the internal heat transfer path of the motor. The thermal isolation target temperature is set to be controlled within 95°C, and a ceramic insulation sheet with a thermal resistivity of not less than 0.4m²·K / W is selected as the main thermal isolation material. Based on the spatial structure of the overload point, the installation area is dimensioned using a three-dimensional modeling tool (such as SolidWorks), and a thermal isolation layer structure with a thickness of 10mm and an area of ​​150×100mm is designed to cover the heat transfer surface between the motor winding housing and the stator. The thermal isolation structure is assembled and simulated to verify the installation feasibility, and a CAD layout drawing and a material selection list are output to form thermal isolation data, which records the material number, size, thickness, coverage location, installation method, etc.

[0047] Preferably, step S16 is specifically as follows:

[0048] Step S161: identifying a heat radiation path based on the motor thermal overload data to obtain heat radiation path data;

[0049] In this embodiment, the time periods and heat accumulation areas corresponding to each overload event are extracted from the motor thermal overload data obtained in step S15. Heat radiation paths are then identified by combining the motor structural drawings with infrared thermal imaging images. Long-term thermal imaging of the motor is performed using an infrared thermal imager with a fixed wavelength band of 8μm to 14μm, with a capture frequency set to once every 60 seconds for at least 8 hours. After pixel grayscale calibration, the image temperature distribution histogram is used to extract the continuous diffusion direction of the highest temperature zone. The contour detection function findContours in OpenCV is used to identify the boundaries of the high-temperature zones. Heat radiation paths are constructed based on the shortest path between the boundary extension direction and the heat source center. Each path records the starting coordinates (heat source center), the ending coordinates (hot spot on the housing surface), the path length (in mm), and the temperature gradient information (in °C / mm). This ultimately creates a heat radiation path data table, including the following fields: [heat source number, path number, starting coordinates, ending coordinates, path length, temperature gradient, and direction angle].

[0050] Step S162: determining the heat source range according to the heat radiation path data to obtain heat source range data;

[0051] In this embodiment, based on the heat radiation path data, the area where all the path starting points are gathered is analyzed, and the average temperature within a radius of 20 mm around each starting point is calculated. If there are three or more overlapping starting points of heat radiation paths in a certain area, and their average temperature is higher than the set heat source judgment threshold of 150°C, then the area is calibrated as a heat source area. The judgment threshold is set based on the insulation level of the motor. If it is higher than this temperature, the insulation performance will be seriously affected. The heat density map of all starting point coordinate points is drawn through a two-dimensional plane thermal map, and the regional boundary is smoothed using Gaussian filtering. The closed hot zone boundary is extracted to form a heat source range. Each heat source range is represented in the form of a polygonal area, and the heat source number, circumscribed rectangle size, center coordinates, area size (unit: mm²), maximum temperature, and average temperature are recorded. Output the heat source range data table for subsequent demarcation of the insulation area.

[0052] Step S163: determining a thermal isolation boundary based on the heat source range data to obtain thermal isolation boundary data;

[0053] In this embodiment, the heat source range data is used to construct an outer thermal isolation boundary for each heat source area. The boundary construction method adopts the isothermal expansion principle: starting from the center of the heat source, the isolation target temperature is set to 95°C, and the heat conduction path is reversed through the thermal conductivity of copper or aluminum to infer the isolation radius. For example, under the conditions of the known copper thermal conductivity of 385W / (m·K), the initial heat source temperature of 155°C, and the ambient temperature of 35°C, the distance is inverted by Fourier's heat conduction law to make the end temperature of the heat transfer path ≤95°C, and the thermal isolation radius is approximately 38mm. On this basis, the heat source boundary is uniformly expanded outward by 38mm and polygon fitting is performed to construct a thermal isolation boundary. The thermal isolation boundary data structure is output, which contains the fields: [heat source number, boundary coordinate point sequence, maximum boundary length, minimum boundary width, closed area], and is exported to a CAD format file for structural design.

[0054] Step S164: Determine the structural dimensions of the heat shield based on the thermal isolation boundary data; select a material type based on the structural dimensions of the heat shield to obtain material type data; cut the plate according to the material type data to obtain heat shield plate data; perform air duct layout based on the heat shield plate data to obtain air duct structure data; reinforce the heat shield based on the air duct structure data to obtain heat shield data;

[0055] In this embodiment, based on the thermal isolation boundary data, the spatial structural dimensional information corresponding to the boundary is extracted from the three-dimensional mechanical structure diagram. Using a CAD drawing tool (such as AutoCAD or SolidWorks), the boundary outline is projected onto the surface of the actual structural model. The corresponding planar unfolded dimensions and 3D curved surface adaptation dimensions are measured to obtain the heat shield structural dimensional data. Parameters include unfolded length and width, curved surface radius, mounting screw hole spacing, and fixing edge thickness. Based on industrial thermal insulation specifications, a minimum board thickness of 5 mm and a target temperature control temperature of 95°C are set. Glass fiber composite boards with a thermal conductivity of no greater than 0.15 W / (m·K) and a density of no greater than 50 kg / m³ are selected from a standard industrial material database. These dimensional and material data are input into the control system of the sheet metal cutting equipment, such as a laser cutting machine (laser power 2.5 kW, positioning accuracy ±0.1 mm). The cutting trajectory and feed rate are set, and the cutting operation is completed to obtain the heat shield sheet material data. Based on the panel data and motor structure, the airflow path of the duct layout was optimized using CFD (computational fluid dynamics) simulation tools (such as ANSYS Fluent). The inlet diameter was set to 50mm and the outlet diameter to 80mm, ensuring a flow rate of at least 3.2m / s. The duct structure data sheet was then generated. Finally, the panels were assembled and 12 fixed reinforcement ribs were set along the main structural boundaries. These were secured with M6 stainless steel bolts, with a spacing of 60mm between the fixing points. After the reinforcement was completed, the complete heat shield data was generated.

[0056] Step S165: Evaluate the thermal isolation effect based on the thermal shield data to obtain thermal isolation data.

[0057] In this example, a thermal insulation effect evaluation was conducted on-site using an installed thermal insulation cover. Temperature sensors (PT100, ±0.3°C accuracy) were placed inside and outside the cover, with a sampling interval set to 30 seconds. The inner and outer surface temperatures of the cover, as well as the ambient temperature, were recorded over an 8-hour period while the motor ran at full load. The temperature difference between the inside and outside of the cover was statistically analyzed to determine whether the thermal insulation effect met the standard. Based on industrial thermal safety regulations, the thermal insulation effect standard is set as follows: the outer surface temperature of the cover must not exceed the ambient temperature +20°C. Data collected at 30-second intervals was analyzed for minimum, maximum, and mean values ​​on a timeline, marking time periods that did not meet the temperature difference standard. The evaluation ultimately outputs thermal insulation data, including the test time period, temperature difference distribution diagram, worst-case temperature zone number, location of insulation failure risk zone, and average temperature difference. All data is summarized and archived in a thermal insulation performance evaluation report.

[0058] Preferably, the ventilation path blockage analysis in step S2 is specifically as follows:

[0059] Count the motor heat according to the motor thermal overload data to obtain the motor heat data; identify the heat source concentration area according to the motor heat data; calculate the temperature difference between adjacent measuring points based on the heat source concentration area to obtain the measuring point temperature difference data;

[0060] In this embodiment, based on the motor thermal overload data generated in the previous step, the average current (unit: A), voltage (unit: V) and overload duration (unit: s) of each motor overload cycle are extracted. The heat input value corresponding to each overload stage is calculated using the formula Q=P×t=U×I×t (where Q is heat, unit: J). In a three-phase asynchronous motor, the motor power should be calculated as follows: Calculate the power factor The heat accumulation was set to 0.85. The total heat generated by each overload event was accumulated, and a motor heat data table was created using the event timestamp and heat value (J) as the record content. The fields included: [time, phase voltage U, phase current I, power P, overload time t, heat Q]. The data was exported in CSV format for the next step of heat source cluster analysis. An infrared thermal imaging camera (wavelength range 8-14 μm, thermal sensitivity ≤ 0.05°C) was placed on the motor housing surface for 24 hours of continuous monitoring. Based on the heat concentration periods recorded in the motor heat data, areas with temperatures above 140°C were extracted from the images. Using thermal image grayscale inversion technology, areas with grayscale values ​​above 220 were defined as high-temperature cores. Spatial cluster analysis was performed using the K-means clustering method, with a cluster number of 5, to identify high-frequency heat clusters. Each heat source cluster area must have an area ≥ 30 cm² and high temperature values ​​in the same coordinate area must appear at least 7 times out of 10 monitoring times. A temporal overlap of ≥ 70% was used as the heat source identification criterion. The resulting heat source concentration area data is formatted as [hot zone number, center coordinates (x, y, z), maximum temperature, average temperature, area] for subsequent temperature difference vector analysis. Multiple temperature sensors were symmetrically arranged around the heat source concentration area, using K-type thermocouples with an accuracy of ±0.3°C and a 10 cm distance between each measuring point, forming a dot matrix. Surface temperature data were collected at 16 measuring points around each heat source, sampling once per second for 300 seconds. The average value was taken as the steady-state temperature. The temperature difference ΔT (Tcenter - Tsurround) between the heat source center and each surrounding measuring point was calculated. The coordinates, temperatures, and corresponding temperature differences of each measuring point were then aggregated to form a measuring point temperature difference dataset. This dataset has the following structure: [heat source number, measuring point number, measuring point coordinates, measuring point temperature, ΔT]. All data were archived using a Matlab script, and a graphical heat distribution map was output.

[0061] Calculate the gradient vector direction based on the temperature difference data of the measuring points to obtain the temperature difference gradient direction data; obtain the motor heat dissipation path data; perform ventilation path mapping based on the motor heat dissipation path data and the temperature difference gradient direction data to obtain the ventilation path data;

[0062] In this embodiment, the gradient calculation method in a two-dimensional coordinate system is used to solve the vector direction of the temperature difference data of the measuring point. The temperature difference distribution matrix T(x,y) is used to calculate the horizontal and vertical temperature difference gradients respectively. , The central difference method is used to solve the numerical gradient, and the gradient direction Each measurement point is assigned a vector arrow pointing in the direction of decreasing temperature (i.e., heat flow). All data is uniformly generated into a vector field diagram, and the output data table contains: [measurement point number, coordinates (x, y), Gx, Gy, gradient direction θ (unit: degrees)] This directional information is used for subsequent matching of ventilation paths with heat flow vectors. A 3D structural scan of the motor's entire structure is performed, and a laser scanning device (accuracy ≤ 0.5 mm) is used to extract a 3D structural model of its housing and fan ventilation system. The air inlet, exhaust, internal duct, and cooling vents on the housing are identified, and the geometric parameters of each air flow path are recorded, including the path's starting and ending coordinates, cross-sectional diameter, and ventilation path length. A wind speed sensor (range 0-15 m / s, accuracy ±0.1 m / s) is set up to monitor air velocity within the duct and create a real-time ventilation velocity distribution map. All data is aggregated into a motor cooling path data table with the following structure: [path number, starting and ending coordinates, path length, cross-sectional area, airflow velocity]. This data output is used for subsequent duct mapping and blockage analysis. By superimposing the temperature gradient vector field obtained in step 4 with the three-dimensional structure of the heat dissipation path obtained in step 5, spatial alignment matching is performed. The matching angle threshold is set to within 15°. If the angle between the temperature gradient vector direction and the ventilation path direction is less than or equal to 15°, the path is considered to have effective ventilation. A Python script combined with a 3D modeling tool (such as ANSYS SpaceClaim) is used to compare each vector with the path direction, marking the effective ventilation path. The ventilation path data is output, including the following fields: [Path number, number of matching vectors, average angle, ventilation effectiveness determination], and a figure is attached to show the overlap between the heat flow direction and the air circulation direction.

[0063] Perform filter clogging detection based on ventilation path data to obtain filter clogging data;

[0064] In this embodiment, a particulate matter collector is installed at the air duct inlet. A PM filter structure and a differential pressure sensor (range 0-1000Pa, accuracy ±1Pa) are used to measure the inlet and outlet pressure differential in real time. The initial pressure differential is set at 80Pa. If the actual pressure differential exceeds 150% of the initial value (i.e., 120Pa) for more than five consecutive minutes, the filter is considered clogged. The detected clogged area number, pressure differential, sampling time, and inlet air velocity are recorded to form a filter clog data table. Fields include: [Filter number, Inlet air velocity, Pressure differential, Time of clog determination, and Severity of clog (classified into three levels: Mild (80-120Pa), Moderate (120-180Pa), and Severe (>180Pa)]. An alarm record is generated for each clogged event number.

[0065] According to the filter clogging data, a blockage sample is collected and a particle size analysis is performed on the blockage sample to obtain the blockage particle size data; based on the blockage particle size data, a surface roughness test is performed to obtain the blockage roughness data;

[0066] In this example, a high-efficiency dust classifier (such as a laser particle size analyzer with a measurement range of 0.1 μm to 500 μm) was used to measure the particle diameter of residual blockage on the filter. Sampling was performed at designated sampling points based on the filter area. 0.5 g of blockage dust was collected from each area. The average particle size was measured three times, and statistical parameters such as the peak size distribution, D10, D50, and D90 were recorded. A particle size data table was generated, including [sample number, D50 value, maximum particle size, minimum particle size, and distribution skewness coefficient]. This data was used to analyze whether the particles were source of blockage, such as fine suspended matter, grease adhesion, or metal oxide chips. The blockage sample was placed under a white light interferometer profilometer with a scanning area of ​​100 μm × 100 μm and a scanning resolution of 0.5 μm. A three-dimensional surface height map was obtained. Roughness parameters such as average roughness Ra, maximum height Rz, and peak-to-valley number Rp were extracted according to the ISO 4287 standard. Particles with an Ra value greater than 4 μm were considered to have high surface adhesion. The output roughness test data table structure is: [sample number, Ra, Rz, Rp, surface type (loose / smooth / sticky)], and the data is used for adhesion strength analysis.

[0067] Evaluate ventilation duct attachment strength based on blockage roughness data;

[0068] In this example, the minimum pull-off force required for particles to adhere to the metal surface of a ventilation duct is measured by combining roughness parameters with data from an adhesion tester (e.g., using the vertical pull-off method, measured in N / cm²). Sample particles are manually attached to a metal specimen (50 mm x 50 mm) of the ventilation duct. After drying, the specimen is vertically loaded until it falls off. Particles with an adhesion force ≥ 2 N / cm² are considered highly adherent. The average adhesion strength of samples with different roughness levels is calculated to create a relationship table: [Ra range, mean adhesion strength], which can be used to infer whether particles are likely to remain attached under current wind speeds and vibration intensities.

[0069] The ventilation path blockage analysis is performed based on the ventilation duct attachment strength to obtain the ventilation path blockage data.

[0070] In this example, all particles with high adhesion strength are marked as high clogging risk sources. The deposition behavior of particles in ventilation ducts is simulated by combining ventilation path wind speed data with particle migration velocity (calculated using the Stokes settling equation). A critical settling velocity is set at 0.8 m / s. If the actual wind speed is lower than this value and the adhesion strength is higher than 2.5 N / cm², the path segment is designated as a high clogging risk area. A ventilation path clogging data table is output, including the following fields: [Path ID, Clogging Risk Level, Adhesion Parameter, Deposition Simulation Results, Recommended Cleaning Cycle (in hours)].

[0071] Preferably, the winding insulation aging detection in step S2 is specifically as follows:

[0072] Calculate wind resistance based on ventilation path blockage data to obtain wind resistance data;

[0073] In this embodiment, the geometric parameter data of each ventilation duct section in the motor ventilation system is first obtained, including cross-sectional area, duct length, number of elbows, bend angle, branch location, etc. At the same time, combined with the collected blockage area location and blockage degree data, the effective ventilation area is calculated according to the ratio of the blockage volume to the duct cross-sectional area, and the residual ventilation channel width of the blocked section is accurately recorded. Then, combined with the real-time wind speed data (in m / s) and air density value (the air density is 1.225 kg / m³ under standard conditions) in the channel, the wind resistance value of each section is calculated based on the flow rate and equivalent cross-sectional area of ​​the ventilation section. The channel pressure loss calculation method in engineering thermodynamics is used to output the resistance value (in Pa) of each air duct section, and compared with the normal wind resistance baseline value to determine the wind resistance increment data, which is output as wind resistance data. The recording format includes the ventilation section number, blockage rate, wind speed, and pressure loss value.

[0074] Perform winding temperature rise analysis based on windage data to obtain winding temperature rise data;

[0075] In this embodiment, after obtaining wind resistance data, a temperature rise analysis is performed on the motor windings in the areas most affected by ventilation. The copper loss parameters of the windings are first obtained. This parameter can be obtained by multiplying the square of the current value by the winding resistance value, where the winding resistance is calculated based on the conductor length, cross-sectional area, and material resistivity. The heat generated per unit time and the heat dissipation capacity of the winding are then calculated, taking into account the change in the heat transfer coefficient in the ventilation path. The heat dissipation capacity is calculated using the convective heat transfer formula, where the heat transfer coefficient h is estimated by the wind speed reduction ratio (a 30% reduction in wind speed corresponds to an approximately 25% reduction in the heat transfer coefficient). The heat generation is compared with the heat transfer capacity to obtain the heat accumulation trend, and finally the winding temperature rise data (in °C) is calculated. This temperature rise data should be based on the standard operating winding temperature for temperature increment assessment, and the thermal stress exposure time should be calculated based on the operating time.

[0076] Conduct insulation material thermal stress testing based on winding temperature rise data to obtain insulation material thermal stress data;

[0077] In this embodiment, the material type of the insulating varnish or paper used in the winding is first determined. The elastic modulus (in GPa), thermal expansion coefficient (in 1 / K), glass transition temperature (Tg), and thermal decomposition onset temperature (Td) of the insulating material are then extracted from the material database. The actual winding temperature is compared with the material's Tg. If the temperature rise exceeds the material's Tg, the winding is considered to have entered the plastic deformation zone; if the temperature rise approaches Td, the winding is marked as entering the thermal decomposition risk zone. During this process, the maximum stress (in MPa) generated within the insulation layer due to thermal gradients and constrained deformation is calculated using the finite element thermal stress analysis method. This stress value is then proportionally compared to the material's tensile strength. If the thermal stress exceeds 50% of the yield strength, a high thermal stress state is recorded and output as the insulation material's thermal stress data.

[0078] Calculate material aging rate based on insulation material thermal stress data;

[0079] In this example, the Arrhenius aging characteristics of the insulation material, including the activation energy (in kJ / mol) and the frequency factor, are used. The temperature rise value is substituted into the aging rate formula to calculate the material's molecular degradation rate per unit time. The thermal stress duration and thermal cycle frequency are also considered. Accelerated aging time for each time period is accumulated according to the equivalent aging life calculation rules. Ultimately, the reduction in the material's actual service life under these thermal stress conditions is calculated, and the material aging rate data is recorded in hours.

[0080] The winding insulation aging is evaluated according to the material aging rate to obtain the winding insulation aging data.

[0081] In this embodiment, the actual operating time and accumulated aging time of the current winding are calculated, and the winding life assessment rules in the IEC standard are used for grading. For example, in IEC60034, the maximum allowable operating temperature for a Class F insulation system is 155°C, at which the insulation life is 20,000 hours. If the measured temperature rises to 170°C, the corresponding life is shortened to 10,000 hours, and so on. By comparing the actual aging time with the standard life limit, the life consumption ratio is calculated and the winding insulation aging level is assigned. For example, 0-30% is defined as Level I, 30%-60% as Level II, and above 60% as Level III. Ultimately, the winding insulation aging data is composed of four data items: winding number, accumulated aging time, remaining life percentage, and aging level, providing a basis for subsequent maintenance decisions.

[0082] Preferably, the insulating paint film shedding detection in step S2 is specifically as follows:

[0083] Extract insulation surface features based on winding insulation aging data to obtain insulation surface data;

[0084] In this embodiment, based on the winding numbers marked as Grade II or above in the winding insulation aging data, the corresponding equipment is located and high-precision images of the insulation layer are captured. The image capture device must utilize an industrial-grade imaging probe with a resolution of 600 dpi or higher, coupled with an adjustable-intensity ring-shaped LED light source to ensure clear, non-reflective surface textures. The image processing system uses preprocessing techniques such as grayscale normalization, mean filtering, and edge enhancement to remove noise and background interference. An image segmentation algorithm (such as the Otsu method based on maximum inter-class variance) is then used to extract the insulation surface region, obtaining image ROI coordinates and corresponding grayscale texture data. The extraction process quantifies surface feature parameters, including surface roughness factor (expressed as the grayscale change rate per unit area), local brightness difference coefficient, and edge density. All features are indexed by insulation number to form a structured insulation surface data table.

[0085] Performing paint film crack detection based on insulation surface data to obtain insulation paint film crack data;

[0086] In this embodiment, crack image recognition is performed within the extracted insulation surface image region. This process utilizes an image recognition method that combines multi-scale edge detection with a morphological algorithm. First, potential crack boundary features are extracted using Sobel gradient enhancement. Canny edge detection with an adaptive threshold is then used to precisely locate continuous crack contours. A morphological refinement algorithm is then used to restore the pixel skeleton structure of the crack path. After crack path extraction, metrics such as total crack length (in mm), maximum width (in μm), and crack density (in cracks / mm²) are calculated. Crack orientation analysis is also performed (to extract the main crack direction and the number of secondary crack branches). This completes the insulation paint film crack data, including fields such as crack number, starting point coordinates, direction angle, length, and width.

[0087] According to the crack propagation data of the insulating paint film, crack propagation data is obtained by simulating crack propagation;

[0088] In this embodiment, the mechanical parameters of the insulating material and the thermal stress history data are combined to calculate the crack stress field. The crack propagation analysis adopts the crack tip stress intensity factor (SIF) calculation method, and the fracture toughness K_IC (unit: ), the current crack length and thermal stress distribution are mapped to the crack tip, and the SIF value at the crack tip is estimated analytically. If the SIF exceeds the material's K_IC, the crack is considered to have a tendency to grow. The crack's extension path and growth rate are calculated based on the direction of thermal stress changes. The final output includes information such as the predicted crack growth length, predicted growth direction angle, and predicted growth rate (in mm / h), forming the crack growth data.

[0089] Evaluating the insulation paint film adhesion according to the crack growth data to obtain paint film adhesion data;

[0090] In this embodiment, the material of the adhesion surface (e.g., copper, aluminum, etc.) and the type of paint film material are identified to obtain the corresponding interface adhesion strength benchmark value. Adhesion assessment uses a method to infer the cause of crack propagation. The interface fracture pattern is analyzed by analyzing information such as the stress distribution and crack morphology changes (e.g., whether bifurcation occurs) during crack tip propagation. If separation of the adhesion interface is inferred, the energy required to release per unit area of ​​the interface is calculated. This is compared with the standard critical energy value (in J / m²) of this type of interface in existing literature to convert the actual adhesion strength estimate (in MPa) into a paint film adhesion data table. The content includes fields such as the measurement point number, the interface material combination, and the ratio of the calculated adhesion strength to the reference value.

[0091] The insulation paint film shedding test is performed based on the paint film adhesion data to obtain the insulation paint film shedding data.

[0092] In this embodiment, a lower adhesion threshold is set, determined by the insulation system design standard. For example, if the design adhesion strength of the insulation paint film is 8 MPa, the 70% critical adhesion strength is set at 5.6 MPa as the judgment basis. For all data points with adhesion below this value, the system further checks whether the crack length exceeds 5 mm, whether the crack density is greater than 2 cracks / mm², and whether the location is in a heat-concentrated area. If any two of these three criteria are met, the point is marked as a breakaway risk point, and the breakaway detection results are output, including the test number, crack characteristic indicators, adhesion value, and judgment result, forming a complete insulation paint film breakaway data set. This data can be used for subsequent structural protection level assessment and insulation repair decision-making.

[0093] Preferably, step S3 is specifically as follows:

[0094] Step S31: Calculating the shedding depth according to the insulation paint film shedding data to obtain shedding depth data; evaluating the degree of insulation layer damage according to the shedding depth data to obtain insulation layer damage data;

[0095] In the present embodiment, based on the image data of the falling-off area, a three-dimensional profile scanning device is used to perform micron-level high-precision topography measurement on the motor insulation surface. The profile scanning device adopts white light interferometry or laser confocal microscopy, and the scanning accuracy reaches within 0.1 μm. The scanning area needs to cover all identified falling-off numbered areas, and a relative height reference system is established with the non-falling-off peripheral surface as the reference plane. The three-dimensional height matrix data obtained by scanning is subjected to pixel-by-pixel subtraction operation, and the deepest point, average depth, and area-weighted depth of each falling-off area are counted to form the maximum falling-off depth data, average falling-off depth data, and comprehensive falling-off volume data respectively. All falling-off depth data are stored in a structured database with the falling-off number as the index. After completing the falling-off depth data statistics, the degree of damage to the insulation layer needs to be assessed according to the insulation system structural hierarchy. The motor winding insulation layer generally includes a three-layer structure of primer, intermediate layer, and topcoat, and the thickness of each layer is 10 μm, 30 μm, and 20 μm, respectively. If the maximum shedding depth exceeds 60μm, it is marked as full-thickness penetration damage; if the shedding depth is between 30-60μm, it is considered mid-layer and surface damage; if the shedding depth is less than 30μm and above the surface layer, it is classified as mild surface damage. Based on the damage area and depth, the comprehensive damage grade for each damage point is defined as Level I (mild), Level II (moderate), or Level III (severe). A data table for insulation layer damage is constructed, with fields including damage number, shedding depth, area, and grade.

[0096] Step S32: selecting a protective coating material based on the insulation layer damage data to obtain protective coating material data;

[0097] In this embodiment, data on industrial insulation repair materials registered in the enterprise technical standard library is used, including polyimide coating materials, modified silicone-acrylic coating materials, and two-component epoxy modified materials. Materials are directly matched according to the damage level. For example, for Level III damaged areas, a two-component epoxy coating with a high temperature resistance of ≥180°C and a volume resistivity of ≥10^13Ω·cm is selected; for Level II damaged areas, a polyimide material with a temperature resistance of ≥150°C and a flexibility of ≤1mm is selected; and for Level I damaged areas, a low-viscosity, fast-curing silicone-acrylic emulsion coating material is selected. The physical performance parameters of the selected materials, such as material number, thermal conductivity, electrical strength, dielectric constant, curing time, and adhesion strength, are imported into a protective coating material data table.

[0098] Step S33: determining the coating thickness according to the protective coating material data to obtain coating thickness data; determining the topcoat coating method according to the protective coating material data to obtain topcoat coating method data;

[0099] In this example, the standard single-layer thickness is calculated based on the solid content in the protective coating material data and the target coating thickness. For polyimide, for example, if the target dry film thickness is 30 μm and the material solids content is 60%, the wet film thickness is 30 μm ÷ 60% ≈ 50 μm. The coating thickness is precisely controlled by setting the movement speed based on spray equipment parameters (such as paint output and spray width). If brushing is used, the number of brush strokes and the interval between each coating stroke must be controlled. If the material is a two-component epoxy, it must be pre-mixed according to the specified ratio (for example, A:B = 4:1), allowed to stand for 3 minutes to degas, and then injected into the spray gun chamber. The spray pressure should be controlled between 0.3 and 0.5 MPa. The coating method data field records the coating process (e.g., air spray, high-pressure airless spray, brush coating) and, combined with the coating thickness value, forms the complete topcoat coating method data.

[0100] Step S34: Integrate the coating thickness data and the topcoat coating method data to obtain protective coating upgrade data;

[0101] In this example, coating thickness data and topcoat application method data are integrated into a unified data structure to create protective coating upgrade data. This data includes fields such as coating material number, substrate preparation requirements (e.g., polishing grit P800), single-layer thickness, total number of layers applied, coating method, curing conditions (e.g., temperature 80°C, time 2 hours), and final surface roughness Ra value. All parameters are used for coating upgrade process review and implementation records.

[0102] Step S35: constructing a motor coating thermal insulation model based on the protective coating upgrade data and the thermal insulation data;

[0103] In this embodiment, a motor coating thermal insulation model is constructed based on protective coating upgrade data and existing thermal isolation data (including heat shield structure data, thermal reflectivity, and thermal conductivity). In this model, the coating is defined as a multilayer structure, with the bottom layer representing the original winding surface, the middle layer being the insulation repair layer, and the top layer being the surface protective coating. The heat shield data is projected onto the winding region using heat flux density. The finite volume method is used to calculate the steady-state heat flux distribution between each layer, using thermal conductivity, thermal diffusivity, and layer thickness as input parameters. A steady-state thermal equilibrium solution is calculated, taking into account the duct velocity (2.5 m / s), ambient temperature (40°C), and the coating's outer surface temperature boundary (maximum 120°C). Thermal field data, such as the winding surface heat flux density (unit: W / m²) and interface temperature distribution, are output for subsequent aging calculations.

[0104] Step S36: performing coating aging simulation according to the motor coating thermal insulation model to obtain coating aging data.

[0105] In this example, an aging rate formula extracted from measured thermal aging data for insulating coatings was used. The aging test cycle was set to 10,000 hours, and the thermal field parameters were loaded into the simulation system based on the surface temperature gradient calculated from the steady-state solution in step S35. The aging simulation process was iterated sequentially, with each step consisting of 500 hours. The coating thickness change, adhesion loss percentage, and thermal conductivity increase were updated at each step. Ultimately, the aging rate (in μm / h), adhesion loss rate (in %), and thermal conductivity trend for each coating layer were output, forming a complete set of coating aging data. All results were stored in a time series and hierarchical structure, providing data support for coating maintenance and replacement cycles.

[0106] Preferably, step S36 is specifically as follows:

[0107] Step S361: uploading the motor coating thermal insulation model to the thermal aging simulation platform;

[0108] In this embodiment, uploading the motor coating insulation model to the thermal aging simulation platform requires the use of a dedicated thermal aging simulation platform with a 3D structural modeling input interface and a thermal conduction simulation solver, such as a customized module platform based on Ansys Workbench or COMSOL Multiphysics. The upload operation requires first exporting the motor coating insulation model to a supported standard geometry file format, such as STEP (.stp) or Parasolid (.x_t). The model should include clear thermal property distinctions between various structural layers, such as the inner winding area, insulation repair layer, protective surface layer, and external thermal shield structure. In the model upload module, the "Structure Definition Verification" function must be invoked to perform integrity checks on issues such as the closure of thermal boundaries between structures and the absence of missing material properties. The regional discretization operation is then completed using meshing parameters. The minimum side length of the mesh element must be controlled within 0.1 mm, and the maximum side length must not exceed 2 mm to ensure that the subsequent heat flux distribution accuracy meets the requirements of aging calculations.

[0109] Step S362: setting the ambient temperature range to 40°C-120°C and the humidity range to 30%-80%;

[0110] In this embodiment, the ambient temperature range and humidity range are set in the environmental parameter setting module of the thermal aging simulation platform. The ambient temperature range is set to 40°C to 120°C, and a step loading method with a step value of 10°C is adopted, with a total of 9 temperature levels. Each temperature level corresponds to an independent thermal aging simulation cycle to form a group simulation scheme. The humidity range is set to 30% to 80%, and a fixed relative humidity level combination (30%, 50%, 65%, 80%) is used for loading, and the temperature level is fully permuted and combined, and finally a total of 36 groups of working condition combination parameter tables are generated. Each group of working conditions is organized in a numbered manner, and the fields contain temperature value, humidity value, simulation number and working condition label. The system calls the table for simulation loading in sequence.

[0111] Step S363: Setting the motor operating power range to 50kW-150kW and the duty cycle to 8 hours-24 hours;

[0112] In this embodiment, the motor operating power range and duty cycle are set. The operating power range is 50kW to 150kW, divided into 5 gears with a step of 25kW: 50, 75, 100, 125, 150kW; the duty cycle is set to five gears of 8 hours, 12 hours, 16 hours, 20 hours and 24 hours. The power and duty cycle are arranged crosswise to construct 25 types of motor operating load condition parameters. Each load condition needs to correspond to the definition of the heat source term. For example, in COMSOL, the heat source term is injected through the "Heat Source" node. The unit is W / m³, and the calculation method is the total power value divided by the volume of the winding material. Assume a power of 100kW and a winding volume of 0.01m 3 , the injected heat source term is 1×10 7 W / m³. The simulation duration for each operation cycle is set in hours through the "Time Dependent Study" module, and the thermal field evolution simulation is performed according to the set cycle.

[0113] Step S364: Input the coating material thermal conductivity of 0.15-0.35 W / (m·K) and thickness of 0.5-2.0 mm;

[0114] In this embodiment, the thermal conductivity range is set to 0.15-0.35W / (m·K), with a step of 0.05, a total of 5 values: 0.15, 0.20, 0.25, 0.30, 0.35W / (m·K), which are entered into the "Thermal Conductivity" field of the material library, and the unit is maintained in W / (m·K). The coating thickness is used as a geometric parameter in the model structure. The thickness value range is set to 0.5mm to 2.0mm in the modeling module. Four structural versions of 0.5mm, 1.0mm, 1.5mm, and 2.0mm are generated with a step of 0.5mm. The thickness variable is defined in the "Geometry Parameters" mode in the parametric simulation platform to achieve automatic batch simulation. All combination results cross-arrange thermal conductivity and thickness to generate 20 thermal parameter combinations, each of which is executed as an independent simulation task.

[0115] Step S365: Run the thermal aging simulation program to obtain coating aging data.

[0116] In this example, the thermal aging simulation program is officially run. The platform calls upon all combined input parameters, sequentially loading operating condition data (temperature, humidity), operating load data (power, cycle), and thermal parameter data (thermal conductivity, thickness). The thermal field distribution is solved in the steady-state heat conduction module. The aging rate curve is then loaded into the "Material Degradation" module, using the Arrhenius equation to control the time-temperature dependence. The iterative simulation cycle is set to update every 100 hours, with a total cumulative duration of 10,000 hours. After each time step, target metrics are extracted, including material thickness change (μm), residual adhesion percentage (%), thermal conductivity increase (%), and surface crack density (units / mm²). Finally, all simulation results are numbered and mapped to the parameter combination table. These results are stored in the "Aging Results Data Table" in the database. Fields include simulation number, operating condition group number, time step number, and thermal aging indicator value. This completes the generation of coating aging data.

[0117] Preferably, step S4 is specifically as follows:

[0118] Step S41: performing heat conduction simulation according to the coating aging data to obtain heat conduction data;

[0119] In this embodiment, the execution of the heat conduction simulation requires input parameters based on the coating aging data, and a heat conduction simulation project is constructed using a platform with finite element thermal field solving function (such as COMSOL Multiphysics). Boundary conditions and material properties should be set during the simulation process, where the thermal conductivity of the coating after aging is input as the main variable. This parameter should be derived from the change in thermal conductivity after aging obtained in step S365, for example, from the original thermal conductivity of 0.25W / (m·K) to 0.18W / (m·K). The heat source position is set in the winding area, and the total thermal power value is injected through the "Heat Source" module, for example, it is set to 100kW, corresponding to a volume heat source intensity of 1×10 7 W / m³ (assuming a winding volume of 0.01 m³). Convection heat transfer conditions are applied to the outer boundary, with an ambient temperature of 80°C and a heat transfer coefficient of 25 W / (m²·K). The propagation path of the overall heat flux in the interlayer structure is simulated. The calculation period is a steady-state solution. The output is temperature distribution data along each cross-section of the structure, in °C. The spatial resolution must be less than 0.2 mm to meet the requirements of subsequent thermal gradient calculations.

[0120] Step S42: drawing a thermal gradient distribution map based on the heat conduction data; performing interface temperature difference offset identification based on the thermal gradient distribution map to obtain temperature difference offset data; and identifying an abnormal interface adhesion area based on the temperature difference offset;

[0121] In this embodiment, the thermal gradient distribution map is drawn based on the derived three-dimensional temperature field data. The temperature gradient between adjacent grids is calculated for each discrete grid node, and the gradient vector distribution is calculated using the central difference method. The calculation is performed using the SciPy library in MATLAB or Python. The temperature gradient magnitude (K / m) and the gradient direction (vector unit direction) are mapped to a three-dimensional graph, and pseudo-color strips are used to represent the area of ​​concentrated heat distribution. On this basis, the interface temperature difference offset phenomenon is identified by comparing the normal thermal gradient of each interface point by point. The offset is defined as the angle difference between the upper and lower temperature gradient directions of the interface exceeding 15°, or the absolute temperature difference between the center and edge of the interface exceeding 5°C. These areas are marked as offset anomaly areas. Finally, the coordinates of the offset anomaly area are exported to form "temperature difference offset data", the format of which includes fields such as three-dimensional spatial position, interface normal angle offset, and absolute temperature difference.

[0122] Step S43: performing coating hollowing detection according to the interface adhesion abnormality area to obtain coating hollowing data;

[0123] In this embodiment, abnormal adhesion interface areas are located based on temperature offset data, and then a thermoacoustic excitation detection device is used to detect coating hollowing. This device consists of a thermal excitation module and an infrared detection module. The thermal excitation module rapidly heats the identified area to 90°C for one second using a directed laser pulse. A high-resolution infrared thermal imager (such as the FLIR X8501sc) then captures a thermal response image sequence at 100 frames per second for a total of 10 seconds. Hollowing is determined by the time and amplitude differences between the surface temperature rise and recovery curves in the image sequence. If the recovery time of an area lags behind the normal area by more than 1.5 seconds, and the temperature rise exceeds 1.2 times the standard deviation, it is determined to be a hollowing area. The detection results are exported as a "hollowing distribution map," recording information such as the center coordinates of the hollowing area, area (mm²), and hollow thickness (mm) in JSON or CSV format. This organized information becomes "coating hollowing data."

[0124] Step S44: performing coating adhesion degradation analysis based on the coating hollowing data to obtain coating adhesion degradation data;

[0125] In this embodiment, the hollowing ratio per unit area is calculated by extracting the area, thickness and distribution density of the hollowing region. For example, if a hollowing region with a total area of ​​18 cm² is detected within a range of 100 cm², the hollowing ratio is 18%. The artificial indentation adhesion measurement method is used for verification. A pressure needle is used to apply a constant load of 1N to the hollowing area, and the ratio of the coating rupture area to the original indentation area is recorded to evaluate the adhesion residual rate. For example, if the indentation area is 1 mm² and the peeling area is 0.6 mm², the adhesion residual rate is 40%. On this basis, the adhesion degradation level is comprehensively calculated in combination with the hollowing distribution density, area ratio and residual rate. The adhesion drop of more than 50% and the hollowing density exceeding 2 cm² are defined as "severe degradation". Below this standard, they are divided into "mild degradation" and "moderate degradation". The final result forms "adhesion degradation data", and the fields include degradation level, hollowing rate, residual rate, etc.

[0126] It is particularly important that step S44 includes the following steps:

[0127] Step S441: Scan the surface morphology according to the coating hollowing data to obtain a coating microstructure image;

[0128] In this embodiment, a high-resolution scanning electron microscope (SEM) or a laser confocal microscope is used to scan the coating surface and the hollow area point by point to collect microstructure images. During the scanning process, the scanning range is set to at least 5 mm of the hollow area and its surrounding area, and the resolution is set to 0.5 microns to ensure that detailed microscopic morphological information is obtained. During the scanning process, the height information of each scanning point is recorded to form microscopic three-dimensional surface morphological image data of the coating. After the image acquisition is completed, the data is subjected to noise filtering to remove artifacts caused by instrument vibration and environmental influences to ensure the accuracy of subsequent analysis. Special image processing software is used to enhance and adjust the contrast of the collected image to make the tiny pores and crack features more obvious. The output of this step is a high-resolution image containing the hollow area of ​​the coating and its microscopic morphological features, which provides a data basis for subsequent pore detection.

[0129] Step S442: performing pore detection based on the coating microstructure image to obtain coating pore data;

[0130] In this embodiment, image segmentation techniques, such as threshold-based binarization, are used to distinguish the pore area from the solid coating area in the image. The threshold parameter is determined by pre-calibrated standard samples. The grayscale threshold is usually set between 80 and 120 to ensure that the pore boundaries are clearly discernible. An edge detection algorithm (such as Canny edge detection) is used to accurately locate the pore contour and eliminate noise and tiny impurities. Pore morphological parameters include pore diameter, shape factor, and pore area, which are calculated using image processing software. The pore detection process also needs to eliminate misjudged areas of non-porous structure, such as cracks or coating peeling edges, to ensure the accuracy of the pore data. The pore detection results are output as a list of digitized pore coordinates and dimensions, which serve as input data for subsequent pore distribution density calculations.

[0131] Step S443: Calculating the pore distribution density based on the coating pore data to obtain pore distribution density data;

[0132] In this embodiment, the ratio of the total number of detected pores to the scanned area is used as the basic parameter for pore distribution density. The scanned area is determined by the scan range in step S441 and is typically measured in square millimeters. The uniformity of pore distribution within the entire scanned area is statistically analyzed, and point pattern analysis methods in spatial statistics (such as Ripley's K function) are used to assess pore aggregation or dispersion. The pore volume fraction is calculated as a supplementary parameter and combined with the pore size distribution to obtain more detailed pore structure information. The pore distribution density data is output as the number of pores per square millimeter and the pore volume fraction, along with a distribution uniformity index, providing a quantitative basis for mechanical strength testing.

[0133] Step S444: performing a coating mechanical strength test based on the pore distribution density data to obtain coating mechanical strength data;

[0134] In this embodiment, a coating tensile tester or a micro-nano indenter is used to measure the mechanical properties of the coating sample. Sample preparation includes cutting a coating specimen of standard size (such as 10mm×10mm) from the test area, and the thickness is kept within the standard range (0.5mm to 2.0mm). The test point layout is adjusted according to the pore distribution density data to ensure coverage of pore-dense areas and pore-sparse areas. The test parameters set the tensile speed to 1mm / min and the indentation load to 100mN, and the mechanical properties of the coating, such as elastic modulus, fracture strength and hardness, are measured. The data records include stress-strain curves and indentation depths, and the influence of pores on mechanical strength is evaluated in combination with pore distribution density analysis. The mechanical strength data is output in numerical form as a basic parameter for evaluating the degradation of coating adhesion.

[0135] Step S445: performing adhesion degradation evaluation based on the coating mechanical strength data to obtain coating adhesion degradation data.

[0136] In this embodiment, a standard adhesion test method, such as the grid method or the pull-out method, is used, combined with a comprehensive analysis of the mechanical strength test results. The grid method is carried out in accordance with the GB / T9286 standard, the grid size is set to 1mm×1mm, the line spacing is 0.2mm, and a special tape is used to perform a pull-out test, and the percentage of the peeling area is recorded. The pull-out method uses a puller, the pulling speed is set to 0.5 mm / min, and the maximum pull-out force is measured. The mechanical strength data is combined with the adhesion test data, and a multi-factor weighted calculation model is used to calculate the adhesion degradation index. When the threshold is set to less than 50% of the maximum strength, it is judged to be significantly degraded. The coating adhesion degradation data is output, and the value range is 0 to 1, representing the degree from no degradation to complete degradation. This data provides a quantitative basis for subsequent coating maintenance and repair.

[0137] Step S45: performing risk level classification based on the coating adhesion degradation data, thereby obtaining protection capability classification data;

[0138] In this embodiment, the risk level classification is based on the degradation level, where mild degradation corresponds to protection level I, moderate degradation corresponds to protection level II, and severe degradation corresponds to protection level III. In addition, the thermal conductivity degradation coefficient is also introduced. As auxiliary reference parameters, It is defined as the ratio of thermal conductivity after aging to the original thermal conductivity. For example, if the original thermal conductivity is 0.25W / (m·K) and after aging it is 0.18W / (m·K), then =0.72. If the value falls below 0.7, the protection level is automatically increased by one level to compensate for the unstable adhesion boundary caused by the decline in thermal performance. By entering all assessment fields (hollowing density, residual adhesion rate, and thermal conductivity degradation coefficient) into the risk level assessment rule table, the "protection capability classification data" is automatically generated. Each result includes information such as the protection level number, comprehensive score, and the coordinates of the abnormal area.

[0139] It is particularly important that step S45 includes the following steps:

[0140] Step S451: calculating a protection capability index based on coating adhesion degradation data;

[0141] In this example, specific numerical parameters of coating adhesion degradation are obtained. These data are derived from coating adhesion strength curves previously obtained through nondestructive testing techniques (such as ultrasonic guided waves, thermal imaging, or mechanical peel testing). Based on the actual measured adhesion strength values, a weighted average method is used to calculate a comprehensive protective capability index, combining the coating area, thickness uniformity, and degradation distribution. The protective capability index is represented by a normalized value between 0 and 1, with 0 indicating complete failure and 1 indicating complete integrity. Weighting coefficients are pre-set based on the importance of different coating regions, such as assigning a weight of 0.6 to critical structural areas and 0.4 to non-critical areas. The overall protective capability index is calculated by multiplying the local adhesion degradation degree by the corresponding weight and summing the results. This step utilizes professional data processing software to process and statistically analyze the multi-point sampling data to ensure the accuracy and representativeness of the index.

[0142] Step S452: Comparing the preset adhesion safety threshold with the protection capability index to obtain a protection capability critical point;

[0143] In this embodiment, a preset adhesion safety threshold is clearly defined, which is generally set between 0.3 and 0.5 MPa. The protection capability index calculated in step S451 is compared point by point with this threshold, and the specific value at which the index first falls below the threshold is identified as the protection capability critical point. During data processing, statistical analysis tools are used to screen out intervals with a clear downward trend in protection capability. Combined with time series data analysis, the stability and effectiveness of the critical point are confirmed, eliminating interference from transient abnormal data. This step emphasizes the precise application of the threshold to ensure that the critical point reflects the actual critical failure state of the coating.

[0144] Step S453: Divide the protection risk level according to the protection capability critical point;

[0145] In this embodiment, the protection capability index range of 0-1 is divided into five risk level intervals: Level 1 (indicator 0.8-1.0), Level 2 (indicator 0.6-0.79), Level 3 (indicator 0.4-0.59), Level 4 (indicator 0.2-0.39), and Level 5 (indicator 0-0.19). Using the critical points determined in step S452, data samples are mapped to the interval standards one by one to clearly define the risk level assigned to each coating area. During the risk level classification process, a hierarchical clustering algorithm is used to statistically aggregate the risk intervals. This is combined with spatial distribution data to create a risk heat map to assist in identifying high-risk concentrated areas. This step also includes the removal of extreme values ​​and outliers to ensure that the risk level classification is scientific and objective.

[0146] Step S454: perform protection capability grading according to the protection risk level to obtain protection capability grading data.

[0147] In this embodiment, the protection risk level data is used as input, and the physical performance indicators and environmental exposure parameters of the coating material are combined to systematically grade the overall protection capability of the coating. The grading standard is set to five levels: Level A corresponds to level one risk, indicating that the coating is intact and has strong protection capabilities; Level B corresponds to level two risk, with a slight decrease in protection capabilities; Level C corresponds to level three risk, with obvious weakening of protection; Level D corresponds to level four risk, with serious damage to the protection function; Level E corresponds to level five risk, with the protection function basically lost. The grading operation is implemented through the database association function, combined with the coating life prediction model data for cross-validation, and the final protection capability grading data is output. The graded data is stored in a structured format to support subsequent automated management system calls and maintenance decision-making.

[0148] Step S46: Generate digital labels based on the protection capability classification data to obtain digital protection capability management data.

[0149] In this embodiment, the label generation module calls the label encoding function and defines the field sequence according to a unified structure: equipment number, motor number, area number, protection level, risk score, update time, validity period, and additional notes. The protection level field is represented by level numbers I to III, and the risk score field is generated on a percentage basis. The update cycle is set to once every three months, and the label validity period does not exceed six months. The label generation format is a two-dimensional data matrix code (such as DataMatrix ECC200). The ZXing open source library is used to generate the QR code image with an image resolution of 300×300 pixels. Before encoding, the field content is encrypted with SHA-256 to ensure uniqueness and security. The generated digital label is embedded in the equipment archive database and bound to the coating number and location coordinates, realizing the organized output of "digital management data of protection capabilities."

[0150] Preferably, this specification also provides an industrial digital management system based on model building, which is used to execute the industrial digital management method based on model building as described above. The industrial digital management system based on model building includes:

[0151] Thermal isolation design module, used to obtain industrial equipment operation logs; perform motor thermal overload analysis based on the industrial equipment operation logs to obtain motor thermal overload data; perform thermal isolation design based on the motor thermal overload data to obtain thermal isolation data;

[0152] The insulation aging detection module is used to analyze ventilation path blockage based on motor thermal overload data to obtain ventilation path blockage data; perform winding insulation aging detection based on ventilation path blockage data to obtain winding insulation aging data; and perform insulation paint film shedding detection based on winding insulation aging data to obtain insulation paint film shedding data;

[0153] The motor coating thermal insulation model construction module is used to upgrade the protective coating based on the insulation paint film shedding data to obtain the protective coating upgrade data; construct the motor coating thermal insulation model based on the protective coating upgrade data and thermal insulation data; and perform coating aging simulation based on the motor coating thermal insulation model to obtain coating aging data;

[0154] The digital management module is used to perform coating hollowing detection based on coating aging data to obtain coating hollowing data; perform coating adhesion degradation analysis based on coating hollowing data to obtain coating adhesion degradation data; and perform digital management of protective capabilities based on coating adhesion degradation data to obtain digital management data of protective capabilities.

[0155] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0156] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. An industrial digital management method based on model building, characterized in that: The following steps are involved: Step S1: Obtaining an industrial equipment operation log; performing a motor thermal overload analysis based on the industrial equipment operation log to obtain motor thermal overload data; performing a thermal isolation design based on the motor thermal overload data to obtain thermal isolation data; Step S2: performing ventilation path blockage analysis based on the motor thermal overload data to obtain ventilation path blockage data; performing winding insulation aging detection based on the ventilation path blockage data to obtain winding insulation aging data; performing insulation paint film shedding detection based on the winding insulation aging data to obtain insulation paint film shedding data; Step S3: upgrading the protective coating according to the insulation paint film shedding data to obtain protective coating upgrade data; Build a motor coating thermal insulation model based on protective coating upgrade data and thermal insulation data; perform coating aging simulation based on the motor coating thermal insulation model to obtain coating aging data; Step S4: performing coating hollowing detection according to the coating aging data to obtain coating hollowing data; The coating adhesion degradation analysis is performed based on the coating hollowing data to obtain the coating adhesion degradation data; the protection capability digital management is performed based on the coating adhesion degradation data to obtain the protection capability digital management data.

2. The industrial digital management method based on model building according to claim 1 is characterized in that: Step S1 is specifically as follows: Step S11: Obtaining industrial equipment operation logs; Step S12: Recording current operating parameters based on the industrial equipment operation log; recording voltage operating parameters based on the industrial equipment operation log; Calculate motor power based on current operating parameters and voltage operating parameters; Step S13: Estimating the winding temperature rise based on the motor power to obtain motor winding temperature data; Step S14: performing a high temperature determination on the motor winding temperature data according to a preset motor winding temperature threshold to obtain motor winding high temperature data; calculating a high temperature duration based on the motor winding high temperature data to obtain high temperature duration data; Step S15: Counting the high temperature frequency based on the high temperature data of the motor winding to obtain high frequency and high temperature period data; Thermal overload identification is performed based on high-frequency and high-temperature period data and high-temperature duration data to obtain motor thermal overload data; Step S16: Perform thermal isolation design based on the motor thermal overload data to obtain thermal isolation data.

3. The industrial digital management method based on model building according to claim 2 is characterized in that: Step S16 is specifically as follows: Step S161: identifying a heat radiation path based on the motor thermal overload data to obtain heat radiation path data; Step S162: determining the heat source range according to the heat radiation path data to obtain heat source range data; Step S163: determining a thermal isolation boundary based on the heat source range data to obtain thermal isolation boundary data; Step S164: Determine the structural dimensions of the heat shield based on the thermal isolation boundary data; select a material type based on the structural dimensions of the heat shield to obtain material type data; cut the plate according to the material type data to obtain heat shield plate data; perform air duct layout based on the heat shield plate data to obtain air duct structure data; reinforce the heat shield based on the air duct structure data to obtain heat shield data; Step S165: Evaluate the thermal isolation effect based on the thermal shield data to obtain thermal isolation data.

4. The industrial digital management method based on model building according to claim 1 is characterized in that: The ventilation path blockage analysis in step S2 is specifically as follows: Count the motor heat according to the motor thermal overload data to obtain the motor heat data; identify the heat source concentration area according to the motor heat data; calculate the temperature difference between adjacent measuring points based on the heat source concentration area to obtain the measuring point temperature difference data; Calculate the gradient vector direction based on the temperature difference data of the measuring points to obtain the temperature difference gradient direction data; obtain the motor heat dissipation path data; perform ventilation path mapping based on the motor heat dissipation path data and the temperature difference gradient direction data to obtain the ventilation path data; Perform filter clogging detection based on ventilation path data to obtain filter clogging data; According to the filter clogging data, a blockage sample is collected and a particle size analysis is performed on the blockage sample to obtain the blockage particle size data; based on the blockage particle size data, a surface roughness test is performed to obtain the blockage roughness data; Evaluate ventilation duct attachment strength based on blockage roughness data; The ventilation path blockage analysis is performed based on the ventilation duct attachment strength to obtain the ventilation path blockage data.

5. The industrial digital management method based on model building according to claim 1 is characterized in that: The winding insulation aging detection in step S2 is specifically as follows: Calculate wind resistance based on ventilation path blockage data to obtain wind resistance data; Perform winding temperature rise analysis based on windage data to obtain winding temperature rise data; Conduct insulation material thermal stress testing based on winding temperature rise data to obtain insulation material thermal stress data; Calculate material aging rate based on insulation material thermal stress data; The winding insulation aging is evaluated according to the material aging rate to obtain the winding insulation aging data.

6. The industrial digital management method based on model building according to claim 1 is characterized in that: The insulation paint film peeling detection in step S2 is specifically as follows: Extract insulation surface features based on winding insulation aging data to obtain insulation surface data; Performing paint film crack detection based on insulation surface data to obtain insulation paint film crack data; According to the crack propagation data of the insulating paint film, crack propagation data is obtained by simulating crack propagation; Evaluating the insulation paint film adhesion according to the crack growth data to obtain paint film adhesion data; The insulation paint film shedding test is performed based on the paint film adhesion data to obtain the insulation paint film shedding data.

7. The industrial digital management method based on model building according to claim 1 is characterized in that: Step S3 is specifically as follows: Step S31: Calculating the shedding depth according to the insulation paint film shedding data to obtain shedding depth data; evaluating the degree of insulation layer damage according to the shedding depth data to obtain insulation layer damage data; Step S32: selecting a protective coating material based on the insulation layer damage data to obtain protective coating material data; Step S33: determining the coating thickness according to the protective coating material data to obtain coating thickness data; Determine a topcoat coating method according to protective coating material data to obtain topcoat coating method data; Step S34: Integrate the coating thickness data and the topcoat coating method data to obtain protective coating upgrade data; Step S35: constructing a motor coating thermal insulation model based on the protective coating upgrade data and the thermal insulation data; Step S36: performing coating aging simulation according to the motor coating thermal insulation model to obtain coating aging data.

8. The industrial digital management method based on model building according to claim 1 is characterized in that: Step S36 is specifically as follows: Step S361: uploading the motor coating thermal insulation model to the thermal aging simulation platform; Step S362: setting the ambient temperature range to 40°C-120°C and the humidity range to 30%-80%; Step S363: Setting the motor operating power range to 50kW-150kW and the duty cycle to 8 hours-24 hours; Step S364: Input the coating material thermal conductivity of 0.15-0.35 W / (m·K) and thickness of 0.5-2.0 mm; Step S365: Run the thermal aging simulation program to obtain coating aging data.

9. The industrial digital management method based on model building according to claim 1 is characterized in that: Step S4 is specifically as follows: Step S41: performing heat conduction simulation according to the coating aging data to obtain heat conduction data; Step S42: drawing a thermal gradient distribution map based on the heat conduction data; performing interface temperature difference offset identification based on the thermal gradient distribution map to obtain temperature difference offset data; and identifying an abnormal interface adhesion area based on the temperature difference offset; Step S43: performing coating hollowing detection according to the interface adhesion abnormality area to obtain coating hollowing data; Step S44: performing coating adhesion degradation analysis based on the coating hollowing data to obtain coating adhesion degradation data; Step S45: performing risk level classification based on the coating adhesion degradation data, thereby obtaining protection capability classification data; Step S46: Generate digital labels based on the protection capability classification data to obtain digital protection capability management data.

10. An industrial digital management system based on model building, characterized in that: For executing the industrial digital management method based on model building as claimed in claim 1, the industrial digital management system based on model building comprises: Thermal isolation design module, used to obtain industrial equipment operation logs; perform motor thermal overload analysis based on the industrial equipment operation logs to obtain motor thermal overload data; perform thermal isolation design based on the motor thermal overload data to obtain thermal isolation data; The insulation aging detection module is used to analyze ventilation path blockage based on motor thermal overload data to obtain ventilation path blockage data; perform winding insulation aging detection based on ventilation path blockage data to obtain winding insulation aging data; and perform insulation paint film shedding detection based on winding insulation aging data to obtain insulation paint film shedding data; The motor coating thermal insulation model construction module is used to upgrade the protective coating based on the insulation paint film shedding data to obtain the protective coating upgrade data; construct the motor coating thermal insulation model based on the protective coating upgrade data and thermal insulation data; and perform coating aging simulation based on the motor coating thermal insulation model to obtain coating aging data; The digital management module is used to perform coating hollowing detection based on coating aging data to obtain coating hollowing data; perform coating adhesion degradation analysis based on coating hollowing data to obtain coating adhesion degradation data; and perform digital management of protective capabilities based on coating adhesion degradation data to obtain digital management data of protective capabilities.

Images