A risk early warning method for a motor production process and a related device

Abstract

This invention discloses a risk early warning method and related apparatus for the electric motor production process, relating to the field of data analysis technology. The method includes: collecting real-time operating parameters of production equipment and real-time environmental parameters during the electric motor production process to analyze the impact on production efficiency; analyzing the impact on production accuracy using a behavioral response boundary database based on real-time operating parameters and real-time microstructure data of the electric motor's insulation layer; determining abnormal information using microstructure feature reference intervals based on the real-time microstructure data of the insulation layer; determining production quality risk information based on the abnormal information, production accuracy impact data, and production efficiency impact data; generating early warning information and target intervention measures based on the production quality risk information, and monitoring the execution effect of the target intervention measures to update the intervention strategy library. This invention effectively solves the problems of delayed risk identification and inaccurate early warning in existing technologies, achieving accurate risk early warning and proactive intervention in the production process.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a risk warning method and related device for the electric motor production process. Background Technology

[0002] In the modern production of electric motors, the stable operation of production equipment and the state of its environment are the core foundation for ensuring the final product quality and production efficiency. With the development of intelligent manufacturing, the demand for refined and intelligent control of the production process is becoming increasingly urgent. Real-time monitoring of the operating parameters of production equipment and environmental parameters, and assessment of the comprehensive impact of these parameters, has become a key link in achieving high-quality and high-efficiency production of electric motors.

[0003] However, existing methods for monitoring electric motor production often have shortcomings. Most current systems only monitor thresholds for single or a few parameters, lacking the ability to correlate and comprehensively analyze multi-source parameters. They cannot quantify and assess the specific impact of these parameters on production accuracy and efficiency. On the other hand, the failure to integrate and analyze data on the impact of accuracy and efficiency leads to a lag and inaccurate analysis of electric motor quality risks. Early warning information is often based on post-event alerts rather than pre-event predictions, making it difficult to achieve true early intervention and prevention of risks. This may result in significant quality deviations, efficiency losses, and increased costs. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a risk warning method and related device for the electric motor production process, which realizes accurate risk warning and proactive intervention in the production process, significantly improves the quality control level and production efficiency of the electric motor production process, and reduces production costs.

[0005] To address the aforementioned technical problems, this invention provides a risk warning method for the electric motor manufacturing process, the method comprising:

[0006] Real-time operating parameters and real-time environmental parameters of production equipment are collected during the electric motor production process, and the impact of these parameters on the production efficiency of the electric motor is analyzed to obtain production efficiency impact data.

[0007] Real-time microstructure data of the insulation layer of the electric motor are collected during the production process. Based on the real-time microstructure data of the insulation layer and the real-time operating parameters, the influence of the behavior response boundary database on the production accuracy of the electric motor is analyzed to obtain the production accuracy influence data.

[0008] Based on the real-time microstructure data of the insulation layer, anomaly information is determined using a reference interval of microstructure features;

[0009] Based on the aforementioned anomaly information, data affecting production accuracy, and data affecting production efficiency, production quality risk information is determined.

[0010] Based on the production quality risk information, early warning information and target intervention measures information are generated using the intervention strategy library, and the execution effect information of the target intervention measures information is monitored. The intervention strategy library is updated based on the data confidence level using the execution effect information.

[0011] Optionally, the step of using a behavioral response boundary database to analyze the impact of the insulation layer's real-time microstructure data and real-time operating parameters on the production accuracy of the motor, and obtaining production accuracy impact data, includes:

[0012] The microstructure data and operating parameters of the insulation layer in the production experiment of the electric motor are obtained, and the microstructure data of the insulation layer are extracted to obtain the microstructure feature data of the insulation layer.

[0013] Microscopic quality analysis of the insulation layer is performed based on the experimental microstructure characteristic data of the insulation layer, and the microscopic quality analysis results of the insulation layer are obtained.

[0014] A behavioral response boundary database is constructed based on the experimental microstructure feature data of the insulation layer, the microstructure quality analysis results of the insulation layer, and the experimental operation parameters.

[0015] Feature extraction is performed on the real-time microstructure data of the insulating layer to obtain real-time microstructure feature data of the insulating layer;

[0016] Based on the real-time microstructure feature data and real-time operating parameters of the insulation layer, a micro-behavioral deviation analysis is performed using a behavioral response boundary database to obtain micro-behavioral deviation information. Based on the micro-behavioral deviation information, an impact analysis on the production accuracy of the motor is performed to obtain production accuracy impact data.

[0017] Optionally, the step of determining anomaly information based on the real-time microstructure data of the insulating layer using a reference interval of microstructure features includes:

[0018] Based on the microstructural characteristics of the insulation layer in the production experiment of the electric motor, statistical features were determined, and a reference interval for microstructural features was constructed based on the statistical features.

[0019] Feature extraction is performed on the real-time microstructure data of the insulating layer to obtain real-time microstructure feature data of the insulating layer;

[0020] The real-time microstructure feature data of the insulating layer is compared with the reference range of the microstructure features to obtain the comparison results, and the abnormal information is determined based on the comparison results.

[0021] Optionally, the step of generating early warning information and target intervention measures information based on the production quality risk information using the intervention strategy library includes:

[0022] Based on the aforementioned production quality risk information, early warning information is determined using an intervention strategy library;

[0023] Based on the production quality risk information, several candidate intervention measures are identified using the intervention strategy library. The implementation effect of each candidate intervention measure is predicted by combining the real-time operating parameters of the production equipment and the real-time environmental parameters, thereby obtaining the implementation effect prediction information.

[0024] Based on the implementation effect prediction information of each candidate intervention, the target intervention information is selected from all candidate intervention information.

[0025] Optionally, the implementation effect information of the monitoring target intervention measures includes:

[0026] After obtaining the equipment operation information, environmental information, and insulation layer microstructure information after executing the target intervention measures, parameter change analysis is performed based on the equipment operation information, environmental information, and insulation layer microstructure information to obtain parameter change information;

[0027] Historical benchmark production data is obtained, and external disturbances are distinguished based on the parameter change information and historical benchmark production data to obtain the external disturbance distinction results;

[0028] The effectiveness of the target intervention measures is determined based on the parameter change information and the external disturbance differentiation results.

[0029] Optionally, updating the intervention strategy library based on the execution effect information using data confidence includes:

[0030] Acquire evaluation data on the performance information and determine the target confidence level information of the evaluation data;

[0031] The target confidence information is compared with a preset confidence threshold. If the target confidence information is less than the preset confidence threshold, supplementary data is obtained based on the data acquisition process. Based on the supplementary data, execution effect information, and evaluation data, fused data is generated. The intervention strategy library is updated based on the fused data.

[0032] If the target confidence information is greater than or equal to the preset confidence threshold, the intervention strategy library is updated based on the execution effect information and the evaluation data of the execution effect information.

[0033] Optionally, determining the target confidence information of the evaluation data includes:

[0034] Several data sources for the evaluation data are obtained, and data quality characteristic analysis is performed on each data source to obtain data quality characteristic information;

[0035] Based on the data quality characteristic information, the contribution degree of confidence is analyzed to obtain contribution degree information;

[0036] The first confidence level of the evaluation data is determined based on the data quality characteristics and contribution information of each data source.

[0037] The evaluation data is subjected to outlier identification to obtain outlier data; the evaluation data is subjected to missing value identification to obtain missing value data; and the evaluation data is subjected to noise data point identification to obtain target noise data points.

[0038] The second confidence level information of the evaluation data is determined based on the outlier data, missing value data, and target noise data points, and the target confidence level information of the evaluation data is determined based on the first confidence level information and the second confidence level information.

[0039] In addition, the present invention also provides a risk warning device for the electric motor production process, the device comprising:

[0040] Efficiency Impact Analysis Module: This module is used to collect real-time operating parameters and real-time environmental parameters of the production equipment during the electric motor production process, and to perform an impact analysis on the production efficiency of the electric motor based on these parameters to obtain production efficiency impact data.

[0041] Accuracy Impact Analysis Module: Used to collect real-time microstructure data of the insulation layer of the motor during the production process, and to perform production accuracy impact analysis of the motor based on the real-time microstructure data of the insulation layer and real-time operating parameters using a behavioral response boundary database to obtain production accuracy impact data;

[0042] Anomaly information determination module: used to determine anomaly information based on the real-time microstructure data of the insulating layer using a reference interval of microstructure features;

[0043] Quality risk determination module: used to determine production quality risk information based on the aforementioned anomaly information, production accuracy impact data, and production efficiency impact data;

[0044] Early warning and intervention module: This module is used to generate early warning information and target intervention measures information based on the production quality risk information using an intervention strategy library, monitor the execution effect information of the target intervention measures, and update the intervention strategy library based on the execution effect information using the data confidence level.

[0045] In addition, the present invention also provides an electronic device, which includes a processor and a memory. The memory is used to store instructions, and the processor is used to call the instructions in the memory to enable the electronic device to execute the risk warning method for the electric motor production process described above.

[0046] In addition, the present invention also provides a computer-readable storage medium that stores computer instructions that, when executed on an electronic device, cause the electronic device to perform the aforementioned risk warning method for the electric motor production process.

[0047] In this embodiment of the invention, the impact analysis on the production efficiency of the electric motor based on real-time operating parameters and real-time environmental parameters can accurately assess the impact on production efficiency. The impact analysis on the production precision of the electric motor based on real-time microstructure data of the insulation layer and real-time operating parameters using a behavioral response boundary database can more accurately identify the potential impact of microstructure changes on product precision, improving the reliability of precision impact data. Anomaly information is determined using microstructure feature reference intervals based on real-time microstructure data of the insulation layer, providing crucial data support for identifying production quality risks. The determination of production quality risk information based on anomaly information, production precision impact data, and production efficiency impact data effectively solves the problems of being unable to quantify the specific impact of assessment parameters on production precision and efficiency, and the lagging and inaccurate identification of quality risks. Based on production quality risk information, early warning information and target intervention measures are generated using an intervention strategy library. The execution effect information of the target intervention measures is monitored, and the intervention strategy library is updated based on data confidence using the execution effect information. This achieves accurate risk early warning and proactive intervention in the production process, significantly improving the quality control level and production efficiency of the electric motor production process, and reducing production costs. Attached Figure Description

[0048] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0049] Figure 1 This is a flowchart illustrating the risk warning method for the electric motor production process in an embodiment of the present invention;

[0050] Figure 2 This is a flowchart illustrating a risk warning method for the electric motor production process according to another embodiment of the present invention;

[0051] Figure 3This is a schematic diagram of the structural composition of the risk warning device for the electric motor production process in an embodiment of the present invention;

[0052] Figure 4 This is a schematic diagram of the structural composition of the electronic device in an embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Example 1

[0055] Please see Figure 1 , Figure 1 This is a flowchart illustrating a risk warning method for the electric motor production process according to an embodiment of the present invention. The method includes:

[0056] S11: Collect real-time operating parameters and real-time environmental parameters of the production equipment during the electric motor production process, and conduct an impact analysis on the production efficiency of the electric motor based on the real-time operating parameters and real-time environmental parameters to obtain production efficiency impact data;

[0057] In the specific implementation of this invention, real-time operating parameters and real-time environmental parameters of the production equipment are collected during the motor production process. Based on the preset process influence curve, the first efficiency influence information is determined using the real-time operating parameters. The deviation between the real-time environmental parameters and the preset environmental threshold is calculated. Based on the first efficiency influence information and the deviation, the production efficiency influence analysis of the motor is performed to more accurately assess the impact on production efficiency, thereby providing more accurate efficiency influence data for subsequent risk warning.

[0058] S12: Collect real-time microstructure data of the insulation layer of the electric motor during the production process, and use the behavior response boundary database to analyze the impact of the insulation layer on the production accuracy of the electric motor based on the real-time microstructure data and real-time operating parameters to obtain production accuracy impact data.

[0059] In the specific implementation of this invention, real-time microstructure data of the insulation layer of the electric motor are collected during the production process. Experimental microstructure data and operating parameters of the insulation layer in the production experiment are obtained, and feature extraction is performed on the experimental microstructure data to obtain characteristic data of the insulation layer. Microstructure quality analysis of the insulation layer is conducted based on this characteristic data. A behavioral response boundary database is constructed based on the experimental microstructure characteristic data, the microstructure quality analysis results, and the experimental operating parameters. Feature extraction is performed on the real-time microstructure data of the insulation layer to obtain real-time microstructure characteristic data. Microscopic behavioral deviation analysis is performed using the behavioral response boundary database based on the real-time microstructure characteristic data and real-time operating parameters to obtain microscopic behavioral deviation information. Based on this information, an analysis of the impact on the production accuracy of the electric motor is conducted, achieving a refined analysis of the production accuracy of the electric motor. This allows for more accurate identification of the potential impact of microstructure changes on product accuracy and improves the reliability of accuracy impact data.

[0060] S13: Based on the real-time microstructure data of the insulating layer, anomaly information is determined using the microstructure feature reference interval;

[0061] In the specific implementation of this invention, statistical features are determined based on the microstructural characteristics of the insulation layer in the production experiment of the electric motor, and a reference interval for microstructural features is constructed based on the statistical features. Features are extracted from the real-time microstructural data of the insulation layer to obtain real-time microstructural characteristic data of the insulation layer. The real-time microstructural characteristic data of the insulation layer is compared with the reference interval for microstructural features, and abnormal information is determined based on the comparison results. This allows for the timely detection of possible microstructural abnormalities in the production process, providing important abnormal information for the identification of production quality risks.

[0062] S14: Determine production quality risk information based on the aforementioned abnormal information, production accuracy impact data, and production efficiency impact data;

[0063] In the specific implementation of this invention, production quality risk information is determined based on the abnormal information, production accuracy impact data, and production efficiency impact data. This overcomes the limitations of single parameter monitoring and lack of correlation analysis of multi-source information in the prior art, and solves the problem of lagging and inaccurate identification of quality risks.

[0064] S15: Based on the production quality risk information, generate early warning information and target intervention measures information using the intervention strategy library, monitor the execution effect information of the target intervention measures information, and update the intervention strategy library based on the data confidence level using the execution effect information.

[0065] In the specific implementation of this invention, early warning information is determined using an intervention strategy library based on production quality risk information. Several candidate intervention measures are then identified using the same library, and the implementation effects of each candidate intervention measure are predicted to select the target intervention measure from all candidates, achieving more targeted and effective risk intervention. Parameter change analysis is performed based on equipment operation information, environmental information, and insulation layer microstructure information after the target intervention measure is implemented, obtaining parameter change information. Historical benchmark production data is acquired, and external disturbances are differentiated based on the parameter change information and the external disturbance differentiation results. The implementation effect information of the target intervention measure is determined based on the parameter change information and the external disturbance differentiation results, more accurately evaluating the actual effect of the intervention measures and providing reliable feedback data for updating the intervention strategy library. The system acquires evaluation data on execution effectiveness and determines the target confidence level for this data. It then compares the target confidence level with a pre-set confidence threshold. If the target confidence level is lower than the pre-set threshold, supplementary data is acquired. Based on this supplementary data, execution effectiveness information, and evaluation data, fused data is generated, and the intervention strategy library is updated using this fused data. If the target confidence level is greater than or equal to the pre-set confidence threshold, the intervention strategy library is updated based on the execution effectiveness information and its evaluation data. This ensures that the updates to the intervention strategy library are based on high-quality, high-confidence data, improving the reliability and effectiveness of the intervention strategies.

[0066] In this embodiment of the invention, the impact analysis on the production efficiency of the electric motor based on real-time operating parameters and real-time environmental parameters can accurately assess the impact on production efficiency. The impact analysis on the production precision of the electric motor based on real-time microstructure data of the insulation layer and real-time operating parameters using a behavioral response boundary database can more accurately identify the potential impact of microstructure changes on product precision, improving the reliability of precision impact data. Anomaly information is determined using microstructure feature reference intervals based on real-time microstructure data of the insulation layer, providing crucial data support for identifying production quality risks. The determination of production quality risk information based on anomaly information, production precision impact data, and production efficiency impact data effectively solves the problems of being unable to quantify the specific impact of assessment parameters on production precision and efficiency, and the lagging and inaccurate identification of quality risks. Based on production quality risk information, early warning information and target intervention measures are generated using an intervention strategy library. The execution effect information of the target intervention measures is monitored, and the intervention strategy library is updated based on data confidence using the execution effect information. This achieves accurate risk early warning and proactive intervention in the production process, significantly improving the quality control level and production efficiency of the electric motor production process, and reducing production costs.

[0067] Example 2

[0068] Please see Figure 2 , Figure 2 This is a flowchart illustrating a risk warning method for the electric motor production process according to another embodiment of the present invention, the method comprising:

[0069] S201: Collect real-time operating parameters and real-time environmental parameters of the production equipment during the electric motor production process, and conduct an impact analysis on the production efficiency of the electric motor based on the real-time operating parameters and real-time environmental parameters to obtain production efficiency impact data;

[0070] In a specific implementation of the present invention, the step of analyzing the impact of the real-time operating parameters and real-time environmental parameters on the production efficiency of the motor to obtain production efficiency impact data includes: determining first efficiency impact information based on the real-time operating parameters using a preset process impact curve; calculating the deviation between the real-time environmental parameters and a preset environmental threshold; and analyzing the impact of the first efficiency impact information and the deviation on the production efficiency of the motor to obtain production efficiency impact data.

[0071] Specifically, collecting real-time operating parameters and environmental parameters of the production equipment during motor production can be achieved by installing various sensors on the equipment. Real-time operating parameters refer to the operational status data of the production equipment at any given moment during motor production, such as vibration frequency, temperature, current, voltage, rotational speed, and processing speed. These parameters directly reflect the health status and workload of the equipment. Real-time environmental parameters refer to the environmental data of the environment in which the production equipment is located at any given moment during motor production, such as workshop temperature, humidity, cleanliness, dust concentration, and noise level. These parameters have a direct or indirect impact on the operational stability of the production equipment and product quality.

[0072] Based on a preset process influence curve, the first efficiency impact information is determined using the real-time operating parameters. The preset process influence curve can be understood as a mathematical model or lookup table pre-established for the process flow in motor production, reflecting the relationship between real-time operating parameters of production equipment and production efficiency. This curve is fitted and verified using a large amount of historical production data and experimental data, quantifying the potential impact of different combinations of operating parameters on production efficiency. Its purpose is to provide a benchmark for preliminary assessment of production efficiency performance under ideal environmental conditions. The first efficiency impact information refers to quantitative data reflecting the impact of the current operating state of production equipment on production efficiency, obtained based on the preliminary assessment of real-time operating parameters and the preset process influence curve. This information can be an efficiency percentage, an efficiency loss value, or an efficiency level, used to characterize the expected production efficiency performance under the current process operating conditions.

[0073] The deviation between real-time environmental parameters and preset environmental thresholds is calculated. This deviation refers to the degree of difference between the current production environment parameters and the environmental thresholds set to ensure optimal production efficiency. This deviation can be quantified by calculating the absolute difference, percentage difference, or specific environmental impact function between the real-time environmental parameters and the corresponding thresholds. The purpose is to identify the potential negative impact of environmental factors on production efficiency, because even if equipment operating parameters are normal, a harsh environment can still lead to a decrease in efficiency. Analyzing the impact of these efficiency impacts on motor production efficiency based on the first efficiency impact information and the deviation provides more accurate and comprehensive data on the impact of production efficiency, thus providing a more reliable basis for determining subsequent production quality risk information. This significantly improves the accuracy and timeliness of risk warnings, helping enterprises to more effectively identify and respond to the risk of declining production efficiency.

[0074] For example, suppose a motor production line operates at a real-time winding machine speed of 1500 RPM and a real-time temperature of 60°C. A preset process influence curve shows that at this speed and temperature, the baseline influence on production efficiency is 95% (i.e., the first efficiency influence information). Simultaneously, the real-time ambient temperature in the workshop is 30°C, while the preset environmental threshold is 25°C, and the humidity is 70%, with a preset environmental threshold of 60%. In this case, the ambient temperature deviation can be calculated as (30-25) / 25 = 20%, and the humidity deviation as (70-60) / 60 = 16.7%. Using a preset environmental influence model, for example, if every 10% temperature deviation leads to a 1% efficiency loss, and every 10% humidity deviation leads to a 0.5% efficiency loss, then the total efficiency loss due to environmental deviations is (20% / 10%)*1% + (16.7% / 10%)*0.5% ≈ 2% + 0.835% = 2.835%. Finally, the initial efficiency impact information (95%) is combined with the efficiency loss due to environmental deviation (2.835%), for example, by multiplying or subtracting, to obtain the final production efficiency impact data, such as 95% * (1 - 2.835%) ≈ 92.36%. Therefore, this method can comprehensively consider equipment operation and environmental factors, providing a more accurate assessment result of production efficiency impact.

[0075] S202: Collect real-time microstructure data of the insulation layer of the electric motor during the production process, and use the behavior response boundary database to analyze the impact of the insulation layer on the production accuracy of the electric motor based on the real-time microstructure data and real-time operating parameters to obtain production accuracy impact data.

[0076] In the specific implementation of this invention, the step of analyzing the impact of the real-time microstructure data and real-time operating parameters of the insulation layer on the production accuracy of the motor using a behavioral response boundary database to obtain production accuracy impact data includes: acquiring experimental microstructure data and experimental operating parameters of the insulation layer of the motor during production experiments, and extracting features from the experimental microstructure data of the insulation layer to obtain experimental microstructure feature data of the insulation layer; performing micro-quality analysis of the insulation layer based on the experimental microstructure feature data of the insulation layer to obtain micro-quality analysis results of the insulation layer; constructing a behavioral response boundary database based on the experimental microstructure feature data of the insulation layer, the micro-quality analysis results of the insulation layer, and the experimental operating parameters; extracting features from the real-time microstructure data of the insulation layer to obtain real-time microstructure feature data of the insulation layer; performing micro-behavioral deviation analysis using the behavioral response boundary database based on the real-time microstructure feature data of the insulation layer and the real-time operating parameters of the insulation layer to obtain micro-behavioral deviation information, and performing production accuracy impact analysis of the motor based on the micro-behavioral deviation information to obtain production accuracy impact data.

[0077] Specifically, real-time microstructure data of the motor insulation layer is collected during the production process. This real-time microstructure data refers to the data obtained by real-time detection of the microstructure of the motor insulation material during motor production. For example, information such as the material's crystal structure, porosity, defect distribution, and chemical composition are obtained through techniques like microscopy, spectral analysis, and X-ray diffraction. This data is a key indicator for evaluating the insulation layer's performance and lifespan.

[0078] Acquiring experimental microstructure data and operating parameters of the insulation layer of an electric motor during production experiments refers to manufacturing the motor in a controlled experimental environment and simultaneously collecting microstructure data of its insulation layer and operating parameters during the production process. This experimental data forms the basis for constructing a behavioral response boundary database, aiming to capture typical behavioral patterns of the insulation layer microstructure under different operating parameters. Feature extraction is performed on the experimental microstructure data of the insulation layer to obtain characteristic data of the insulation layer microstructure. This can be understood as identifying and quantifying key, representative features from the original microstructure data, such as the porosity, crack density, and grain size distribution of the insulation layer. These features effectively reflect the physical and chemical state of the insulation layer.

[0079] Microscopic quality analysis of the insulation layer based on the experimental microstructural feature data, and obtaining the microscopic quality analysis results, refers to evaluating the microscopic quality of the insulation layer based on the extracted microstructural feature data, combined with preset quality standards or expert experience, such as determining whether defects exist or the degree of aging. Its purpose is to provide a quality-level reference for the subsequent construction of a behavioral response boundary database.

[0080] Constructing a behavioral response boundary database based on the experimental microstructure feature data of the insulation layer, the microstructure quality analysis results of the insulation layer, and the experimental operating parameters refers to associating and storing the microstructure feature data of the insulation layer obtained in the experiment, the corresponding microstructure quality analysis results, and the experimental operating parameters to form a database containing multi-dimensional information. This database records the mapping relationship between the microstructure features of the insulation layer and its quality performance under different operating conditions, thereby defining the response boundary of the motor insulation layer under normal or abnormal behavior.

[0081] Feature extraction is performed on the real-time microstructure data of the insulation layer to obtain real-time microstructure feature data of the insulation layer. Similar to the feature extraction process of experimental data, the aim is to extract key features that can be used for comparative analysis from the real-time microstructure data of the insulation layer collected in the actual production process.

[0082] Based on the real-time microstructure feature data and real-time operating parameters of the insulation layer, micro-behavioral deviation analysis is performed using a behavioral response boundary database to obtain micro-behavioral deviation information. This involves using the real-time extracted insulation layer microstructure feature data and current real-time operating parameters as input to query or calculate in the behavioral response boundary database to assess the degree of deviation between the current insulation layer's micro-behavioral behavior and the normal behavioral boundaries defined in the database. The deviation information can quantify the degree of abnormality in the current insulation layer state. Based on this micro-behavioral deviation information, the production accuracy impact analysis of the motor is performed to obtain production accuracy impact data. This involves assessing the potential impact of this deviation on the motor's production accuracy based on the calculated micro-behavioral deviation information, combined with preset rules or models. For example, insulation layer defects may lead to uneven winding spacing, abnormal magnetic field distribution, etc., thereby obtaining specific production accuracy impact data. This allows for a more accurate assessment of the potential impact of the insulation layer's microstructure state on the motor's production accuracy, significantly improving the accuracy and reliability of the production accuracy impact data, and thus enhancing the effectiveness of the entire risk warning system. This helps to promptly identify and resolve potential production quality problems.

[0083] S203: Based on the real-time microstructure data of the insulating layer, anomaly information is determined using a reference interval of microstructure features;

[0084] In the specific implementation of this invention, the step of determining abnormal information based on the real-time microstructure data of the insulation layer using a microstructure feature reference interval includes: determining data statistical features based on the experimental microstructure feature data of the insulation layer of the motor in production experiments, and constructing a microstructure feature reference interval based on the data statistical features; extracting features from the real-time microstructure data of the insulation layer to obtain real-time microstructure feature data of the insulation layer; comparing the real-time microstructure feature data of the insulation layer with the microstructure feature reference interval to obtain a comparison result, and determining abnormal information based on the comparison result.

[0085] Specifically, determining statistical characteristics based on the microstructural features of the insulation layer in electric motor production experiments involves collecting and analyzing a large amount of microstructural data on the insulation layer of electric motors produced under controlled experimental conditions. This data extracts features that characterize the motor's normal state, including but not limited to insulation layer thickness, density, porosity, grain size distribution, and defect quantity. Statistical characteristics can be understood as parameters obtained after statistically processing these experimental microstructural feature data, such as mean, variance, standard deviation, median, and quartiles. Their purpose is to quantify the distribution pattern of the insulation layer's microstructure under normal production conditions. Constructing a microstructural feature reference interval based on these statistical characteristics means establishing one or more ranges for judging whether the insulation layer's microstructure is normal, using the determined statistical characteristics. For example, statistical methods (such as percentile method or confidence interval construction method) can be used to define the upper and lower limits of the microstructure characteristics of normal insulation layer. When the microstructure characteristic data collected in real time falls within the reference interval, it is considered to be in a normal state; otherwise, there may be an anomaly. The construction of this reference interval is intended to provide a reliable benchmark for subsequent real-time data comparison.

[0086] Feature extraction is performed on the real-time microstructure data of the insulating layer to obtain real-time microstructure feature data of the insulating layer. Using techniques such as image processing, signal analysis or pattern recognition, feature data of the same type as those determined in the experimental stage are extracted from the real-time microstructure data, such as real-time insulating layer thickness, density, porosity, etc., so as to be effectively compared with the pre-constructed microstructure feature reference range.

[0087] Comparing the real-time microstructure feature data of the insulation layer with the microstructure feature reference interval to obtain the comparison result means comparing the real-time extracted insulation layer microstructure feature data with the constructed microstructure feature reference interval one by one or comprehensively. For example, it can be determined whether the real-time feature value exceeds the upper or lower limit of the reference interval, or the deviation of the real-time feature value from the center value of the reference interval can be calculated. Determining abnormal information based on the comparison result means judging whether there is an anomaly in the current insulation layer microstructure. If the real-time feature data exceeds the reference interval, or the deviation reaches a preset threshold, it can be determined that there is abnormal information, indicating that the insulation layer may have potential quality problems or defects. Through the real-time comparison mechanism, subtle changes or potential defects in the insulation layer microstructure can be detected in a timely manner, thereby achieving early warning of potential quality problems in the motor production process, providing a solid data foundation for subsequent risk assessment and intervention measures, and significantly improving the quality control level and risk management capability of the motor production process.

[0088] S204: Determine production quality risk information based on the aforementioned anomaly information, production accuracy impact data, and production efficiency impact data;

[0089] In the specific implementation of this invention, production quality risk information is determined based on the abnormal information, production accuracy impact data, and production efficiency impact data. For example, a risk assessment model can be set up, which takes the abnormal information, production accuracy impact data, and production efficiency impact data as inputs, and comprehensively evaluates this information through weighted averaging, fuzzy logic reasoning, or machine learning algorithms, thereby quantifying the quality risk level of the current production process and generating specific production quality risk information.

[0090] S205: Based on the production quality risk information, generate early warning information and target intervention measures information using the intervention strategy library, monitor the execution effect information of the target intervention measures information, obtain evaluation data of the execution effect information, and determine the target confidence information of the evaluation data;

[0091] In a specific implementation of this invention, the step of generating early warning information and target intervention measures information based on the production quality risk information using an intervention strategy library includes: determining early warning information based on the production quality risk information using the intervention strategy library; determining several candidate intervention measures based on the production quality risk information using the intervention strategy library, and predicting the implementation effect of each candidate intervention measure based on the real-time operating parameters of the production equipment and real-time environmental parameters to obtain implementation effect prediction information; and selecting target intervention measures information from all candidate intervention measures based on the implementation effect prediction information of each candidate intervention measure.

[0092] Specifically, determining early warning information based on the aforementioned production quality risk information using the intervention strategy library means that the system retrieves and determines corresponding early warning information from a pre-established intervention strategy library based on the identified production quality risk information. This early warning information aims to promptly notify relevant personnel or the system that there are potential or actual production quality risks, requiring appropriate attention or action. The intervention strategy library stores early warning content and triggering conditions corresponding to different risk types and risk levels.

[0093] Based on the aforementioned production quality risk information, the system utilizes an intervention strategy library to identify several candidate intervention measures. This means that while determining early warning information, the system, based on the production quality risk information, filters all possible intervention measures applicable to the current risk situation from the intervention strategy library, forming a set of candidate intervention measures. These candidate measures may include various types such as adjusting equipment parameters, replacing components, optimizing environmental conditions, and suspending production. Combining the real-time operating parameters of the production equipment and real-time environmental parameters, the system predicts the implementation effect of each candidate intervention measure, obtaining implementation effect prediction information. This means that to ensure the effectiveness of the selected intervention measures, the system simulates or predicts for each candidate intervention measure. This prediction process fully considers the current real-time operating parameters of the production equipment and real-time environmental parameters, assessing the potential effects of each candidate measure under these specific conditions, such as its impact on production efficiency, product accuracy, and equipment lifespan. The implementation effect prediction information can be a quantitative indicator, such as the percentage reduction in risk, the increase in production efficiency, or the increase in cost.

[0094] The system selects the target intervention from among all candidate intervention measures based on the predicted implementation effects of each candidate intervention. This means that after predicting the implementation effects of all candidate interventions, the system comprehensively evaluates and compares these predictions according to preset optimization objectives (e.g., risk minimization, efficiency maximization, cost minimization, etc.), thereby selecting the optimal one or a group of interventions as the target intervention. This target intervention is the best response plan determined after effect prediction and optimization, tailored to the current specific production situation and risk context. This ensures that the generated early warnings and interventions are highly customized and optimized, significantly improving the accuracy and effectiveness of risk response.

[0095] For example, suppose that during the production of an electric motor, a production quality risk of localized overheating of the insulation layer is identified. Based on this risk information, the system will determine the corresponding early warning information from the intervention strategy library, such as "Insulation layer overheating risk, please pay immediate attention." Simultaneously, the system will filter several candidate intervention measures from the intervention strategy library, such as: 1. Reduce winding current by 10%; 2. Increase cooling fan speed by 20%; 3. Suspend the current production line for equipment inspection; 4. Adjust the ambient temperature to 22℃. Next, the system will combine the real-time operating parameters of the current production equipment and real-time environmental parameters to predict the implementation effect of each of the above candidate intervention measures. For example, it is predicted that "reducing winding current by 10%" may lead to a 5% decrease in production efficiency but an 80% reduction in overheating risk; "increasing cooling fan speed by 20%" may lead to a 2% increase in energy consumption but a 70% reduction in overheating risk with minimal impact on production efficiency; "suspending the production line" may completely eliminate the risk but cause significant production losses; and "adjusting the ambient temperature" may take a long time to take effect.

[0096] Finally, based on this predicted implementation effect information and combined with preset optimization objectives (e.g., minimizing the impact on production efficiency while ensuring effective risk control), the system selects the optimal solution from all candidate intervention measures. In this example, the system might choose "increasing the cooling fan speed by 20%" as the target intervention measure because it effectively reduces risk while minimizing the impact on production efficiency.

[0097] Furthermore, the execution effect information of the monitoring target intervention measures includes: acquiring equipment operation information, environmental information, and insulation layer microstructure information after the execution of the target intervention measures, and performing parameter change analysis based on the equipment operation information, environmental information, and insulation layer microstructure information to obtain parameter change information; acquiring historical benchmark production data, and distinguishing external disturbances based on the parameter change information and historical benchmark production data to obtain external disturbance distinction results; and determining the execution effect information of the target intervention measures based on the parameter change information and external disturbance distinction results.

[0098] Specifically, acquiring equipment operation information, environmental information, and insulation layer microstructure information after implementing the target intervention measures means that the system continuously collects various real-time data related to the motor production process after the intervention measures are implemented. Equipment operation information may include parameters such as temperature, pressure, speed, current, and voltage of the production equipment; environmental information may include temperature, humidity, and dust concentration in the production workshop; insulation layer microstructure information refers to the structural data of the motor insulation material at the microscopic level, such as the material crystal structure and defect distribution obtained through microscopy or spectral analysis. Parameter change analysis based on the aforementioned equipment operation information, environmental information, and insulation layer microstructure information aims to quantify the changes in these parameters before and after the implementation of the intervention measures. For example, statistical quantities such as the average value, standard deviation, and rate of change of each parameter can be calculated to identify the potential trend or magnitude of change caused by the intervention measures, thereby obtaining parameter change information.

[0099] Acquiring historical baseline production data refers to collecting data related to equipment operation, environment, and insulation layer microstructure under normal, stable production conditions, without being affected by specific intervention measures. Based on the parameter change information and historical baseline production data, external disturbance differentiation aims to identify and eliminate the impact of external factors caused by non-intervention measures on production process parameters. For example, seasonal changes in the production environment, batch differences in raw materials, and natural aging of equipment can all lead to parameter fluctuations. These factors need to be distinguished from the actual effects of intervention measures. By comparing parameter change information with historical baseline production data, statistical methods, machine learning models, or expert systems can be used to analyze the sources of parameter changes, thereby obtaining the results of external disturbance differentiation.

[0100] The effectiveness information of the target intervention measures is determined based on the parameter change information and the results of distinguishing external disturbances. This means that after excluding the influence of external disturbances, the portion of the parameter change information that is indeed caused by the target intervention measures is extracted and quantified into specific effectiveness. For example, if the intervention measures aim to reduce the operating temperature of a certain piece of equipment, then the actual decrease in the equipment's operating temperature after excluding external factors such as ambient temperature fluctuations is considered the effectiveness information of the intervention measures. This effectiveness information can be a quantitative indicator or a comprehensive evaluation result of the effectiveness, efficiency, or impact on production quality of the intervention measures. This provides a purer and more accurate assessment of the intervention effectiveness, avoiding misjudgments caused by confounding factors, thus providing a reliable basis for subsequent updates to the intervention strategy library.

[0101] Furthermore, determining the target confidence information of the evaluation data includes: acquiring several data sources of the evaluation data, and performing data quality characteristic analysis on each data source to obtain data quality characteristic information; performing confidence contribution analysis based on the data quality characteristic information to obtain contribution information; determining the first confidence information of the evaluation data based on the data quality characteristic information and contribution information of each data source; identifying outliers in the evaluation data to obtain outlier data, identifying missing values ​​in the evaluation data to obtain missing value data, identifying noisy data points in the evaluation data to obtain target noisy data points; determining the second confidence information of the evaluation data based on the outlier data, missing value data, and target noisy data points, and determining the target confidence information of the evaluation data based on the first confidence information and the second confidence information.

[0102] Specifically, obtaining evaluation data on the effectiveness of implementation refers to the quantitative or qualitative analysis of the effectiveness information after monitoring the implementation of target intervention measures, forming evaluation data. Several data sources can be used to obtain this evaluation data, including data from different sensors, manual records, historical databases, and other sources. These data sources may have different collection accuracy, reliability, and update frequency. Data quality characteristic analysis is performed on each data source to evaluate its accuracy, completeness, consistency, timeliness, and other characteristics, thereby obtaining data quality characteristic information. For example, it can analyze sensor calibration records, the standardization of manual record entry, and the degree of data cleansing in historical databases.

[0103] Based on the aforementioned data quality characteristics, a contribution analysis of the confidence level is performed to obtain contribution level information. This means that different data sources may contribute differently to the confidence level of the final evaluation data. For example, high-precision, real-time updated sensor data may have a higher contribution level, while data from manually entered sources with more errors will have a lower contribution level. The contribution level analysis can be conducted using methods such as expert scoring, historical data verification, or machine learning models.

[0104] The first confidence level of the evaluation data is determined based on the data quality characteristics and contribution information of each data source. The first confidence level mainly reflects the reliability of the data source and the quality characteristics of the data itself. For example, a weighted average method can be used to combine the quality characteristics and contribution information of each data source to calculate the first confidence level.

[0105] Outlier identification is performed on the evaluation data to obtain outlier data. Outliers are points in the dataset that significantly deviate from other observations. They may be caused by measurement errors, data entry errors, or real but rare events. Identification methods can include statistical methods, clustering methods, or distance-based methods. Missing value identification is performed on the evaluation data to obtain missing value data. Missing values ​​are instances where some observations in the dataset are missing or not recorded, which may lead to inaccurate analysis results. Identification methods typically include checking for null values, NaN values, or specific placeholders. Noise data point identification is performed on the evaluation data to obtain target noise data points. Noise data points are random or irrelevant errors in the data that may interfere with the identification of data patterns. Identification methods can include filtering techniques, wavelet transform, or density-based clustering methods.

[0106] Based on the outlier data, missing value data, and target noisy data points, a second confidence level is determined for the evaluation data. This second confidence level primarily reflects the completeness, accuracy, and purity of the data. For example, the number, severity, and impact on the overall data distribution of outliers, missing values, and noisy data points all reduce the second confidence level. Based on the first and second confidence levels, a target confidence level is determined for the evaluation data. By comprehensively considering both the first and second confidence levels, the target confidence level can be determined. This comprehensive approach can employ various fusion strategies such as weighted summation, fuzzy logic, or decision trees to obtain a final confidence value that comprehensively reflects the reliability of the data.

[0107] It should be noted that the determination of the second confidence information of the evaluation data based on the outlier data, missing value data, and target noise data points includes: performing an overall data distribution impact analysis based on the outlier data to obtain overall data distribution impact information; performing a statistical characteristic analysis of similar data based on the missing value data to obtain statistical characteristic information of similar data; performing a data pattern difference analysis based on the target noise data points to obtain data pattern difference information; and determining the second confidence information of the evaluation data based on the overall data distribution impact information, the statistical characteristic information of similar data, and the data pattern difference information.

[0108] Specifically, an overall data distribution impact analysis is performed based on the outlier data to obtain overall data distribution impact information. By analyzing the degree of influence of these outlier data on the overall data distribution, their potential negative impact on data reliability can be quantified. For example, the impact of outlier data on the mean, variance, skewness, or kurtosis can be calculated to obtain overall data distribution impact information. A statistical characteristic analysis of similar data is performed based on the missing value data to obtain similar data statistical characteristic information. By analyzing the statistical characteristics of missing value data in similar data, such as the missing rate, missing pattern (random missing, non-random missing, etc.), and the impact of missing values ​​on similar data statistics (such as mean, median), the impact of missing values ​​on data confidence can be assessed.

[0109] Data pattern difference analysis is performed on the target noise data points to obtain data pattern difference information. By analyzing the differences between noise data points and normal data patterns, such as through signal-to-noise ratio, data smoothness, or pattern recognition algorithms, this difference can be quantified, and the impact of noise on data quality can be assessed. Based on the overall data distribution impact information, the statistical characteristics of similar data, and the data pattern difference information, a second confidence level is determined for the evaluation data. These three types of information reflect data quality defects from different dimensions. By comprehensively considering them, the inherent reliability of the data can be assessed more comprehensively and accurately.

[0110] S206: Determine whether the target confidence information is greater than or equal to a preset confidence threshold;

[0111] In the specific implementation of this invention, the target confidence information is compared with a preset confidence threshold. The preset confidence threshold is a pre-defined standard used to determine whether the evaluation data is reliable enough to be directly used to update the intervention strategy library. This threshold can be flexibly configured according to the actual application scenario, data sensitivity, and requirements for the accuracy of strategy updates. If the target confidence information is less than the preset confidence threshold, the process proceeds to step S207 to obtain supplementary data. This supplementary data is then combined with the execution effect information and evaluation data to form fused data for updating the intervention strategy library. If the target confidence information is greater than or equal to the preset confidence threshold, the process proceeds to step S208 to update the intervention strategy library based on the execution effect information and the evaluation data of the execution effect information.

[0112] S207: Obtain supplementary data based on the data acquisition process, and generate fused data based on the supplementary data, execution effect information and evaluation data, and update the intervention strategy library based on the fused data;

[0113] In the specific implementation of this invention, if the target confidence level is less than a preset confidence threshold, it indicates that the existing evaluation data may be insufficient to support reliable strategy updates. In this case, the system will proactively trigger a data acquisition process to obtain additional supplementary data, such as more detailed sensor data, manual verification results, or historical similar case data. Subsequently, the supplementary data, execution effect information, and evaluation data are integrated to form fused data. Fusion data aims to provide a more comprehensive and reliable information foundation, thereby ensuring that updates to the intervention strategy library are based on more sufficient data, avoiding deviations in strategy updates due to insufficient or quality data.

[0114] S208: Update the intervention strategy library based on the execution effect information and the evaluation data of the execution effect information.

[0115] In the specific implementation of this invention, if the target confidence information is greater than or equal to a preset confidence threshold, it indicates that the existing evaluation data is sufficiently reliable. At this point, the intervention strategy library can be directly updated using the execution effect information and its evaluation data, without requiring additional data collection and fusion steps, thus improving update efficiency. This ensures that the intervention strategy library is always iteratively optimized based on high-quality data, thereby enhancing the robustness and adaptability of the entire risk warning method.

[0116] For example, if the target confidence level is 0.6, while the preset confidence threshold is 0.7, the system will determine that the current data reliability is insufficient because the target confidence level is lower than the preset confidence threshold. In this case, the system will initiate a data acquisition process, for example, by adding more detailed scanning electron microscopy analysis of the insulation layer of this batch of motors, or by cross-validating data from similar historical batches to obtain supplementary data. Subsequently, this supplementary data is fused with the original execution effect information and evaluation data to generate more comprehensive fused data. Based on this fused data, the strategies regarding the curing temperature of insulation materials in the intervention strategy library are updated to ensure that the updated strategies are based on more sufficient and reliable data support.

[0117] If, after analyzing the assessment data on the degree of noise reduction during motor operation, the target confidence level is determined to be 0.8, and the preset confidence threshold remains at 0.7, the system will determine that the current data has high reliability because the target confidence level is greater than or equal to the preset confidence threshold. In this case, the system will directly use the execution effect information and assessment data to update the noise control strategies in the intervention strategy library, without requiring additional data collection steps, thereby improving the efficiency of strategy updates.

[0118] In this embodiment of the invention, the impact analysis on the production efficiency of the electric motor based on real-time operating parameters and real-time environmental parameters can accurately assess the impact on production efficiency. The impact analysis on the production precision of the electric motor based on real-time microstructure data of the insulation layer and real-time operating parameters using a behavioral response boundary database can more accurately identify the potential impact of microstructure changes on product precision, improving the reliability of precision impact data. Anomaly information is determined using microstructure feature reference intervals based on real-time microstructure data of the insulation layer, providing crucial data support for identifying production quality risks. The determination of production quality risk information based on anomaly information, production precision impact data, and production efficiency impact data effectively solves the problems of being unable to quantify the specific impact of assessment parameters on production precision and efficiency, and the lagging and inaccurate identification of quality risks. Based on production quality risk information, early warning information and target intervention measures are generated using an intervention strategy library. The execution effect information of the target intervention measures is monitored, and the intervention strategy library is updated based on data confidence using the execution effect information. This achieves accurate risk early warning and proactive intervention in the production process, significantly improving the quality control level and production efficiency of the electric motor production process, and reducing production costs.

[0119] Example 3

[0120] Please see Figure 3 , Figure 3 This is a schematic diagram of the structural composition of a risk warning device for the electric motor production process in an embodiment of the present invention. The device includes:

[0121] Efficiency Impact Analysis Module 31: Used to collect real-time operating parameters and real-time environmental parameters of production equipment during the electric motor production process, and to perform an impact analysis on the production efficiency of the electric motor based on the real-time operating parameters and real-time environmental parameters, thereby obtaining production efficiency impact data;

[0122] Accuracy Impact Analysis Module 32: Used to collect real-time microstructure data of the insulation layer of the motor during the production process, and to perform production accuracy impact analysis of the motor based on the real-time microstructure data of the insulation layer and real-time operating parameters using a behavioral response boundary database to obtain production accuracy impact data;

[0123] Anomaly information determination module 33: used to determine anomaly information based on the real-time microstructure data of the insulating layer using a reference interval of microstructure features;

[0124] Quality risk determination module 34: used to determine production quality risk information based on the anomaly information, production accuracy impact data, and production efficiency impact data;

[0125] Early warning and intervention module 35: It is used to generate early warning information and target intervention measures information based on the production quality risk information using the intervention strategy library, monitor the execution effect information of the target intervention measures information, and update the intervention strategy library based on the data confidence level using the execution effect information.

[0126] In the specific implementation of this invention, the specific implementation of the device item can be referred to the implementation of the method item above, and will not be repeated here.

[0127] In this embodiment of the invention, the impact analysis on the production efficiency of the electric motor based on real-time operating parameters and real-time environmental parameters can accurately assess the impact on production efficiency. The impact analysis on the production precision of the electric motor based on real-time microstructure data of the insulation layer and real-time operating parameters using a behavioral response boundary database can more accurately identify the potential impact of microstructure changes on product precision, improving the reliability of precision impact data. Anomaly information is determined using microstructure feature reference intervals based on real-time microstructure data of the insulation layer, providing crucial data support for identifying production quality risks. The determination of production quality risk information based on anomaly information, production precision impact data, and production efficiency impact data effectively solves the problems of being unable to quantify the specific impact of assessment parameters on production precision and efficiency, and the lagging and inaccurate identification of quality risks. Based on production quality risk information, early warning information and target intervention measures are generated using an intervention strategy library. The execution effect information of the target intervention measures is monitored, and the intervention strategy library is updated based on data confidence using the execution effect information. This achieves accurate risk early warning and proactive intervention in the production process, significantly improving the quality control level and production efficiency of the electric motor production process, and reducing production costs.

[0128] This invention provides a computer-readable storage medium storing a computer program. When executed by a processor, this program implements the risk warning method for the electric motor production process described in any of the above embodiments. The computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic cards, or optical cards. In other words, the storage device includes any medium that stores or transmits information in a readable form by a device (e.g., a computer, a mobile phone), and can be a read-only memory, a disk, or an optical disk, etc.

[0129] Example 4

[0130] Please see Figure 4 , Figure 4 This is a schematic diagram of the structural composition of the electronic device in an embodiment of the present invention.

[0131] This invention also provides an electronic device, such as... Figure 4 As shown, the electronic device includes a memory 41, a processor 43, and a computer program 42 stored in the memory 41 and executable on the processor 43. Those skilled in the art will understand that... Figure 4The illustrated electronic device does not constitute a limitation on all devices and may include more or fewer components than illustrated, or combine certain components. Memory 41 can be used to store computer program 42 and various functional modules. Processor 43 runs the computer program 42 stored in memory 41, thereby performing various functional applications and data processing of the device. Memory can be internal memory or external memory, or both. Internal memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, or random access memory. External memory may include hard disks, floppy disks, ZIP disks, USB flash drives, magnetic tapes, etc. Processor 43 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, a single-chip microcomputer, or a processor 43, or any conventional processor, etc. The processors and memories disclosed in this invention include, but are not limited to, these types of processors and memories. The processors and memories disclosed in this invention are merely examples and not intended to be limiting.

[0132] As one embodiment, the electronic device includes: one or more processors 43, a memory 41, and one or more computer programs 42, wherein the one or more computer programs 42 are stored in the memory 41 and configured to be executed by the one or more processors 43, and the one or more computer programs 42 are configured to perform the risk warning method for the electric motor production process in any of the above embodiments. For specific implementation details, please refer to the above embodiments, which will not be repeated here.

[0133] In this embodiment of the invention, the impact analysis on the production efficiency of the electric motor based on real-time operating parameters and real-time environmental parameters can accurately assess the impact on production efficiency. The impact analysis on the production precision of the electric motor based on real-time microstructure data of the insulation layer and real-time operating parameters using a behavioral response boundary database can more accurately identify the potential impact of microstructure changes on product precision, improving the reliability of precision impact data. Anomaly information is determined using microstructure feature reference intervals based on real-time microstructure data of the insulation layer, providing crucial data support for identifying production quality risks. The determination of production quality risk information based on anomaly information, production precision impact data, and production efficiency impact data effectively solves the problems of being unable to quantify the specific impact of assessment parameters on production precision and efficiency, and the lagging and inaccurate identification of quality risks. Based on production quality risk information, early warning information and target intervention measures are generated using an intervention strategy library. The execution effect information of the target intervention measures is monitored, and the intervention strategy library is updated based on data confidence using the execution effect information. This achieves accurate risk early warning and proactive intervention in the production process, significantly improving the quality control level and production efficiency of the electric motor production process, and reducing production costs.

[0134] Furthermore, the above provides a detailed description of a risk warning method and related device for an electric motor production process provided by the embodiments of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A risk early warning method for electric motor production process, characterized in that, The method includes: Real-time operating parameters and real-time environmental parameters of production equipment are collected during the electric motor production process, and the impact of these parameters on the production efficiency of the electric motor is analyzed to obtain production efficiency impact data. Real-time microstructure data of the insulation layer of the electric motor are collected during the production process. Based on the real-time microstructure data of the insulation layer and the real-time operating parameters, the influence of the behavior response boundary database on the production accuracy of the electric motor is analyzed to obtain the production accuracy influence data. Based on the real-time microstructure data of the insulation layer, anomaly information is determined using a reference interval of microstructure features; Based on the aforementioned anomaly information, data affecting production accuracy, and data affecting production efficiency, production quality risk information is determined. Based on the production quality risk information, early warning information and target intervention measures information are generated using the intervention strategy library, and the execution effect information of the target intervention measures information is monitored. The intervention strategy library is updated based on the data confidence level using the execution effect information. The step of analyzing the impact of the insulation layer's real-time microstructure data and real-time operating parameters on the production accuracy of the electric motor using a behavioral response boundary database to obtain production accuracy impact data includes: acquiring experimental microstructure data and operating parameters of the insulation layer during production experiments, and extracting features from the experimental microstructure data to obtain microstructure feature data; performing microstructure quality analysis on the insulation layer based on the microstructure feature data to obtain microstructure quality analysis results; constructing a behavioral response boundary database based on the microstructure feature data, microstructure quality analysis results, and operating parameters; extracting features from the real-time microstructure data to obtain real-time microstructure feature data; and performing microstructure deviation analysis using the behavioral response boundary database based on the real-time microstructure feature data and operating parameters to obtain microstructure deviation information, and then performing production accuracy impact analysis on the electric motor based on the microstructure deviation information to obtain production accuracy impact data. The method of determining anomaly information based on the real-time microstructure data of the insulation layer using a microstructure feature reference interval includes: determining data statistical features based on experimental microstructure feature data of the insulation layer of the motor in production experiments, and constructing a microstructure feature reference interval based on the data statistical features; extracting features from the real-time microstructure data of the insulation layer to obtain real-time microstructure feature data of the insulation layer; comparing the real-time microstructure feature data of the insulation layer with the microstructure feature reference interval to obtain a comparison result, and determining anomaly information based on the comparison result.

2. The risk warning method for the electric motor production process according to claim 1, characterized in that, The process of generating early warning information and target intervention measures information based on the production quality risk information using the intervention strategy library includes: Based on the aforementioned production quality risk information, early warning information is determined using an intervention strategy library; Based on the production quality risk information, several candidate intervention measures are identified using the intervention strategy library. The implementation effect of each candidate intervention measure is predicted by combining the real-time operating parameters of the production equipment and the real-time environmental parameters, thereby obtaining the implementation effect prediction information. Based on the implementation effect prediction information of each candidate intervention, the target intervention information is selected from all candidate intervention information.

3. The risk warning method for the electric motor production process according to claim 1, characterized in that, The information on the effectiveness of the intervention measures for the monitored targets includes: After obtaining the equipment operation information, environmental information, and insulation layer microstructure information after executing the target intervention measures, parameter change analysis is performed based on the equipment operation information, environmental information, and insulation layer microstructure information to obtain parameter change information; Historical benchmark production data is obtained, and external disturbances are distinguished based on the parameter change information and historical benchmark production data to obtain the external disturbance distinction results; The effectiveness of the target intervention measures is determined based on the parameter change information and the external disturbance differentiation results.

4. The risk warning method for the electric motor production process according to claim 1, characterized in that, The step of updating the intervention strategy library based on data confidence using the execution effect information includes: Acquire evaluation data on the performance information and determine the target confidence level information of the evaluation data; The target confidence information is compared with a preset confidence threshold. If the target confidence information is less than the preset confidence threshold, supplementary data is obtained based on the data acquisition process. Based on the supplementary data, execution effect information, and evaluation data, fused data is generated. The intervention strategy library is updated based on the fused data. If the target confidence information is greater than or equal to the preset confidence threshold, the intervention strategy library is updated based on the execution effect information and the evaluation data of the execution effect information.

5. The risk warning method for the electric motor production process according to claim 4, characterized in that, The determination of the target confidence information for the evaluation data includes: Several data sources for the evaluation data are obtained, and data quality characteristic analysis is performed on each data source to obtain data quality characteristic information; Based on the data quality characteristic information, the contribution degree of confidence is analyzed to obtain contribution degree information; The first confidence level of the evaluation data is determined based on the data quality characteristics and contribution information of each data source. The evaluation data is subjected to outlier identification to obtain outlier data; the evaluation data is subjected to missing value identification to obtain missing value data; and the evaluation data is subjected to noise data point identification to obtain target noise data points. The second confidence level information of the evaluation data is determined based on the outlier data, missing value data, and target noise data points, and the target confidence level information of the evaluation data is determined based on the first confidence level information and the second confidence level information.

6. A risk warning device for the electric motor production process, characterized in that, The device includes: Efficiency Impact Analysis Module: This module is used to collect real-time operating parameters and real-time environmental parameters of the production equipment during the electric motor production process, and to perform an impact analysis on the production efficiency of the electric motor based on these parameters to obtain production efficiency impact data. Accuracy Impact Analysis Module: Used to collect real-time microstructure data of the insulation layer of the motor during the production process, and to perform production accuracy impact analysis of the motor based on the real-time microstructure data of the insulation layer and real-time operating parameters using a behavioral response boundary database to obtain production accuracy impact data; Anomaly information determination module: used to determine anomaly information based on the real-time microstructure data of the insulating layer using a reference interval of microstructure features; Quality risk determination module: used to determine production quality risk information based on the aforementioned anomaly information, production accuracy impact data, and production efficiency impact data; Early warning and intervention module: used to generate early warning information and target intervention measures information based on the production quality risk information using the intervention strategy library, monitor the execution effect information of the target intervention measures information, and update the intervention strategy library based on the data confidence level using the execution effect information; The step of analyzing the impact of the insulation layer's real-time microstructure data and real-time operating parameters on the production accuracy of the electric motor using a behavioral response boundary database to obtain production accuracy impact data includes: acquiring experimental microstructure data and operating parameters of the insulation layer during production experiments, and extracting features from the experimental microstructure data to obtain microstructure feature data; performing microstructure quality analysis on the insulation layer based on the microstructure feature data to obtain microstructure quality analysis results; constructing a behavioral response boundary database based on the microstructure feature data, microstructure quality analysis results, and operating parameters; extracting features from the real-time microstructure data to obtain real-time microstructure feature data; and performing microstructure deviation analysis using the behavioral response boundary database based on the real-time microstructure feature data and operating parameters to obtain microstructure deviation information, and then performing production accuracy impact analysis on the electric motor based on the microstructure deviation information to obtain production accuracy impact data. The method of determining anomaly information based on the real-time microstructure data of the insulation layer using a microstructure feature reference interval includes: determining data statistical features based on experimental microstructure feature data of the insulation layer of the motor in production experiments, and constructing a microstructure feature reference interval based on the data statistical features; extracting features from the real-time microstructure data of the insulation layer to obtain real-time microstructure feature data of the insulation layer; comparing the real-time microstructure feature data of the insulation layer with the microstructure feature reference interval to obtain a comparison result, and determining anomaly information based on the comparison result.

7. An electronic device, the electronic device comprising a processor and a memory, characterized in that, The memory is used to store instructions, and the processor is used to call the instructions in the memory to cause the electronic device to execute the risk warning method for the electric motor production process as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed on an electronic device, cause the electronic device to perform a risk warning method for the electric motor production process as described in any one of claims 1 to 5.

Images