Industrial model effective management method and system based on artificial intelligence

Abstract

The invention relates to the technical field of model supervision, in particular to an industry model effective management method and system based on artificial intelligence. The method comprises the steps of obtaining a historical industry model and historical industry data, determining an industry data standard corresponding to the historical industry data according to the historical industry model, judging whether the historical industry data meets the industry data standard or not, and if yes, performing data analysis on the historical industry data to obtain correction performance data, and updating the industry data standard based on the correction performance data to obtain a correction data standard, and correcting and updating the historical industry model according to the correction performance data and the correction data standard to obtain a corrected industry model. The supervision efficiency of the industry model is improved.

Description

Technical Field

[0001] The present application relates to the technical field of model supervision, and in particular to an effective management method and system for industry models based on artificial intelligence. Background Art

[0002] In today's era of rapid technological development, artificial intelligence (AI) has become an important force driving innovation and progress in all walks of life. The application of AI technology has not only improved the operational efficiency of various industries, but also profoundly changed the operating model of traditional industries.

[0003] However, as the application time of industry application equipment corresponding to different industry models increases, the performance data restrictions that it can withstand will also change accordingly. For example, when an industry device is used for the first time, the application data displayed by the industry device according to a certain application time is X, and the performance data restriction range that it can withstand is Y. After a period of application, when an industry device is operated according to the same application time, the application data displayed is X1, and the performance data restriction range that it can withstand is Y1. This means that there is a certain internal loss in the application process of industry equipment, and neither its own performance data nor its own tolerable data is static.

[0004] At present, the existing industry model is to supervise the application of industry equipment with a data standard. The supervision method is to supervise the application of data displayed by industry equipment with an unchanging data standard. However, since the performance data of industry equipment and the restricted range of performance data themselves have certain variability, simply supervising industry equipment with an unchanging data standard has certain accuracy issues. Summary of the invention

[0005] In order to solve at least one of the above technical problems, the present application provides an effective management method and system for industry models based on artificial intelligence.

[0006] In the first aspect, the present application provides an effective management method of industry models based on artificial intelligence, which adopts the following technical solutions: An effective management method for industry models based on artificial intelligence, including: Acquire a historical industry model and historical industry data, wherein the historical industry model is an industry model trained based on the historical industry data, and the historical industry data is data generated by different production monitoring equipment in the industry during a historical period; Determine, according to the historical industry model, an industry data standard corresponding to the historical industry data, wherein the industry data standard is used to judge whether there is abnormal performance data in the historical industry data that does not meet a preset data range; Determine whether the historical industry data meets the industry data standard, and if so, perform data analysis on the historical industry data to obtain corrected performance data; Updating the industry data standard based on the correction performance data to obtain the correction data standard; The historical industry model is corrected and updated according to the corrected performance data and the corrected data standard to obtain a corrected industry model.

[0007] By adopting the above technical solution, the historical industry model and its corresponding historical industry data are obtained. The historical industry model is trained based on rich historical industry data, which is generated by different production monitoring equipment in the industry, ensuring the comprehensiveness and accuracy of the model. This step provides a solid foundation for the subsequent determination of industry data standards and model correction. The industry data standard is further determined based on the historical industry model. This standard, as a judgment benchmark, can accurately identify whether there are abnormal performance data in the historical industry data, thereby ensuring the accuracy and reliability of data analysis. The timely elimination of abnormal data avoids its interference with the subsequent data analysis results. The historical industry data is strictly analyzed, and after confirming that it meets the industry data standard, the corrected performance data is obtained, thereby realizing the correction of the industry equipment performance data. The industry data standard is updated based on the corrected performance data, and a more accurate and adapted to the current industry situation The corrected data standard is obtained. This makes the data standard closer to the actual application of the equipment, providing a more accurate judgment benchmark for the subsequent industry model correction. According to the corrected performance data and the corrected data standard, the historical industry model is corrected and updated, and finally the corrected industry model is obtained, thereby improving the supervision efficiency of the industry model.

[0008] In a preferred example, the present application may be further configured as follows: the data analysis of the historical industry data to obtain the corrected performance data includes: Determine whether there is actual abnormal feedback in the historical industry data. If so, record the abnormal device node, abnormal device data, and standard performance data corresponding to the abnormal device node corresponding to the actual abnormal feedback in the historical industry data. The standard performance data is the operating data that the device corresponding to the abnormal device node should output under different operating parameters; Perform information planning on the abnormal device nodes, abnormal device data and the standard performance data according to a time series to obtain an abnormal feedback matrix; The standard performance data is corrected and analyzed based on the abnormal feedback matrix to obtain corrected performance data corresponding to the abnormal device node.

[0009] In a preferred example, the present application may be further configured as follows: the correction analysis of the standard performance data based on the abnormal feedback matrix to obtain the corrected performance data corresponding to the abnormal device node includes: Perform information planning on the abnormal device nodes, abnormal device data and the standard performance data according to a time series to obtain an abnormal feedback matrix; Determine the device association relationship between the abnormal device nodes according to the abnormal device data; Reorganize the data in the abnormal feedback matrix based on the device association relationship and the time series to obtain multiple groups of reorganized feedback matrices; Performing data anomaly root cause analysis on the abnormal device data in each group of reorganized feedback matrices, and determining active abnormal devices and passive abnormal devices in each group of reorganized feedback matrices, wherein the active abnormal device is the first device in the reorganized feedback matrix to have data anomalies according to the time sequence, and the passive abnormal device is the device in the reorganized feedback matrix that is not the first device to have data anomalies according to the time sequence; The standard performance data is updated based on the abnormal data corresponding to the active abnormal device to obtain first performance data corresponding to the active abnormal device; Performing independent abnormality detection on the passive abnormal device to determine whether the abnormality of the passive abnormal device is caused by the active abnormal device and / or the adjacent abnormal device; if not, updating the standard performance data with the abnormal data corresponding to the passive abnormal device to obtain second performance data corresponding to the passive abnormal device; The first performance data and the second performance data in each group of reorganized feedback matrices are aggregated to obtain corrected performance data corresponding to the abnormal device node.

[0010] In a preferred example, the present application may be further configured as follows: the updating of the industry data standard based on the correction performance data to obtain the correction data standard includes: Determining a data standard correspondence relationship between the correction performance data and the industry data standard based on the abnormal device node; Determine the abnormal performance data corresponding to each model device node based on the industry data standard; According to the data standard correspondence, data critical value analysis is performed on the correction performance data and the abnormal performance data to obtain the correction data standard.

[0011] In a preferred example, the present application may be further configured as follows: performing data critical value analysis on the correction performance data and the abnormal performance data according to the data standard correspondence to obtain the correction data standard includes: determining a correction data critical value based on the correction performance data; Determining an abnormal data critical value according to the abnormal performance data; Based on the data standard corresponding relationship, the abnormal data critical value is updated by narrowing the range according to the correction data critical value to obtain the correction data standard.

[0012] In a preferred example, the present application may be further configured as follows: reorganizing the data in the abnormal feedback matrix based on the device association relationship and the time series to obtain multiple groups of reorganized feedback matrices, and then further comprising: Inputting the plurality of groups of reorganized feedback matrices into the trained vector extraction model to extract vector features, respectively, to obtain the number of matrix dimensions; The number of matrix dimensions and the multiple groups of reorganized feedback matrices are combined in a one-to-one correspondence process to obtain multiple groups of combined feedback matrices; Processing the data contained in the plurality of groups of combined feedback matrices to obtain comprehensive data; The comprehensive data is input into a preset deduction model for data deduction to obtain multiple groups of future abnormal nodes of the reorganized feedback matrix in future cycle time periods.

[0013] In a preferred example, the present application may be further configured as follows: the processing of the data contained in the plurality of groups of combined feedback matrices to obtain comprehensive data includes: Calculating a normal distribution mean and a normal distribution variance of the data contained in the plurality of groups of combined feedback matrices, and determining a 3σ range of the plurality of groups of combined feedback matrices based on the normal distribution mean and the normal distribution variance; Determining whether the data included in the multiple groups of combined feedback matrices are outside the 3σ range, and if the data included in the multiple groups of combined feedback matrices are outside the 3σ range, determining a first matrix sequence of the multiple groups of combined feedback matrices where the data is located; Calculate a sequence average value according to the first matrix sequence, and replace the data with the sequence average value to obtain a replaced second matrix sequence; The second matrix sequence is processed for missing values ​​and normalization to obtain comprehensive data.

[0014] In the second aspect, the present application provides an effective management system for industry models based on artificial intelligence, which adopts the following technical solutions: An effective management system for industry models based on artificial intelligence, including: A data acquisition module, used to acquire a historical industry model and historical industry data, wherein the historical industry model is an industry model trained based on the historical industry data, and the historical industry data is data generated by different production monitoring equipment in the industry during a historical period; A standard determination module, used to determine the industry data standard corresponding to the historical industry data according to the historical industry model, wherein the industry data standard is used to judge whether there is abnormal performance data in the historical industry data that does not meet a preset data range; A data analysis module, used to determine whether the historical industry data meets the industry data standard, and if so, to perform data analysis on the historical industry data to obtain corrected performance data; A standard updating module, used for updating the industry data standard based on the correction performance data to obtain the correction data standard; The model management module is used to calibrate and update the historical industry model according to the calibrated performance data and the calibrated data standard to obtain a calibrated industry model.

[0015] In a possible implementation, when the data analysis module performs data analysis on the historical industry data to obtain the corrected performance data, it is specifically used to: Determine whether there is actual abnormal feedback in the historical industry data. If so, record the abnormal device node, abnormal device data, and standard performance data corresponding to the abnormal device node corresponding to the actual abnormal feedback in the historical industry data. The standard performance data is the operating data that the device corresponding to the abnormal device node should output under different operating parameters; Perform information planning on the abnormal device nodes, abnormal device data and the standard performance data according to a time series to obtain an abnormal feedback matrix; The standard performance data is corrected and analyzed based on the abnormal feedback matrix to obtain corrected performance data corresponding to the abnormal device node.

[0016] In another possible implementation, when the data analysis module corrects and analyzes the standard performance data based on the abnormal feedback matrix to obtain the corrected performance data corresponding to the abnormal device node, it is specifically used to: Perform information planning on the abnormal device nodes, abnormal device data and the standard performance data according to a time series to obtain an abnormal feedback matrix; Determine the device association relationship between the abnormal device nodes according to the abnormal device data; Reorganize the data in the abnormal feedback matrix based on the device association relationship and the time series to obtain multiple groups of reorganized feedback matrices; Performing data anomaly root cause analysis on the abnormal device data in each group of reorganized feedback matrices, and determining active abnormal devices and passive abnormal devices in each group of reorganized feedback matrices, wherein the active abnormal device is the first device in the reorganized feedback matrix to have data anomalies according to the time sequence, and the passive abnormal device is the device in the reorganized feedback matrix that is not the first device to have data anomalies according to the time sequence; The standard performance data is updated based on the abnormal data corresponding to the active abnormal device to obtain first performance data corresponding to the active abnormal device; Performing independent abnormality detection on the passive abnormal device to determine whether the abnormality of the passive abnormal device is caused by the active abnormal device and / or the adjacent abnormal device; if not, updating the standard performance data with the abnormal data corresponding to the passive abnormal device to obtain second performance data corresponding to the passive abnormal device; The first performance data and the second performance data in each group of reorganized feedback matrices are aggregated to obtain corrected performance data corresponding to the abnormal device node.

[0017] In another possible implementation, when the standard updating module updates the industry data standard based on the correction performance data to obtain the correction data standard, it is specifically used to: Determining a data standard correspondence relationship between the correction performance data and the industry data standard based on the abnormal device node; Determine the abnormal performance data corresponding to each model device node based on the industry data standard; According to the data standard correspondence, data critical value analysis is performed on the correction performance data and the abnormal performance data to obtain the correction data standard.

[0018] In another possible implementation, when the standard updating module performs data critical value analysis on the corrected performance data and the abnormal performance data according to the data standard correspondence to obtain the corrected data standard, it is specifically used to: determining a correction data critical value based on the correction performance data; Determining an abnormal data critical value according to the abnormal performance data; Based on the data standard corresponding relationship, the abnormal data critical value is updated by narrowing the range according to the correction data critical value to obtain the correction data standard.

[0019] In another possible implementation, the system further includes: a vector extraction module, a data combination module, a data processing module and a data deduction module, wherein: The vector extraction module is used to input the multiple groups of reorganized feedback matrices into the trained vector extraction model to extract vector features and obtain the number of matrix dimensions; The data combining module is used to combine the number of matrix dimensions with the multiple groups of reorganized feedback matrices in a one-to-one correspondence to obtain multiple groups of combined feedback matrices; The data processing module is used to process the data contained in the multiple groups of combined feedback matrices to obtain comprehensive data; The data deduction module is used to input the comprehensive data into a preset deduction model for data deduction to obtain multiple groups of future abnormal nodes of the reorganized feedback matrix in future periodic time periods.

[0020] In another possible implementation, when the data processing module processes the data included in the multiple groups of combined feedback matrices to obtain the integrated data, it is specifically used to: Calculating a normal distribution mean and a normal distribution variance of the data contained in the plurality of groups of combined feedback matrices, and determining a 3σ range of the plurality of groups of combined feedback matrices based on the normal distribution mean and the normal distribution variance; Determining whether the data included in the multiple groups of combined feedback matrices are outside the 3σ range, and if the data included in the multiple groups of combined feedback matrices are outside the 3σ range, determining a first matrix sequence of the multiple groups of combined feedback matrices where the data is located; Calculate a sequence average value according to the first matrix sequence, and replace the data with the sequence average value to obtain a replaced second matrix sequence; The second matrix sequence is processed for missing values ​​and normalization to obtain comprehensive data.

[0021] In a third aspect, the present application provides an electronic device, which adopts the following technical solution: at least one processor; Memory; At least one application, wherein at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the above-mentioned artificial intelligence-based industry model effective management method.

[0022] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium stores a computer program, which, when executed in a computer, enables the computer to execute an effective management method for industry models based on artificial intelligence.

[0023] In summary, the present application includes at least one of the following beneficial technical effects: The historical industry model and its corresponding historical industry data were obtained. The historical industry model was trained based on rich historical industry data, which was generated by different production monitoring equipment in the industry, ensuring the comprehensiveness and accuracy of the model. This step provides a solid foundation for the subsequent determination of industry data standards and model correction. The industry data standard was further determined based on the historical industry model. This standard, as a judgment benchmark, can accurately identify whether there are abnormal performance data in the historical industry data, thereby ensuring the accuracy and reliability of data analysis. The timely elimination of abnormal data avoids its interference with the subsequent data analysis results. The historical industry data was strictly analyzed, and after confirming that it met the industry data standard, the corrected performance data was obtained, thereby realizing the correction of the industry equipment performance data. The industry data standard was updated based on the corrected performance data, and a more accurate and adapted corrected data standard was obtained. This makes the data standard closer to the actual application of the equipment, providing a more accurate judgment benchmark for the subsequent industry model correction. According to the corrected performance data and the corrected data standard, the historical industry model was corrected and updated, and finally the corrected industry model was obtained, thereby improving the supervision efficiency of the industry model. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A flowchart of an effective management method for industry models based on artificial intelligence provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of an effective management system for industry models based on artificial intelligence provided in an embodiment of the present application; Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The following is combined with Figure 1 To Attachment Figure 3 This application is described in further detail.

[0026] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make modifications to the present embodiment without any creative contribution as needed, but such modifications are protected by the patent law as long as they are within the scope of the claims of the present application.

[0027] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0028] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0029] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0030] The embodiment of the present application provides an effective management method for industry models based on artificial intelligence, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, and the embodiment of the present application does not limit this. Figure 1 As shown, the method includes: Step S10: Acquire historical industry models and historical industry data.

[0031] Among them, the historical industry model is an industry model trained based on historical industry data, and the historical industry data is the data generated by different production monitoring equipment in the industry during the historical period.

[0032] Specifically, the historical industry model refers to a model that is trained based on industry data in a specific historical period and can reflect the specific laws or characteristics of the industry. This model is usually used for tasks related to equipment monitoring and analysis. Historical industry data refers to the data generated by different production monitoring equipment in the industry during the historical period. These data cover various aspects such as the operating status, production efficiency, and fault records of the equipment, and are the basis for training historical industry models. The specific forms used to represent these data can be numbers, charts, logs, etc.

[0033] For the embodiment of the present application, historical data is obtained through an industry database or data warehouse. These databases or warehouses usually contain historical data of multiple companies in the industry, with higher data richness and diversity. In this case, data screening and integration are required to ensure the accuracy and consistency of the data. Then, the machine learning algorithm is also used to train the model to obtain a historical industry model.

[0034] Step S11: Determine the industry data standard corresponding to the historical industry data according to the historical industry model. The industry data standard is used to judge whether there is abnormal performance data in the historical industry data that does not meet the preset data range.

[0035] Specifically, the step of determining the correspondence with industry data standards usually occurs after the historical industry model is established as part of data quality management and analysis. By comparing historical industry data with industry data standards, abnormal performance in the data can be identified, and corresponding measures can be taken to correct or optimize. This step is crucial to ensure the accuracy and reliability of subsequent analysis.

[0036] Specifically, it is necessary to first formulate the normal range of industry equipment application data and the definition standards of abnormal values ​​based on the characteristics and rules of historical industry models. Then, the historical industry data is compared with these standards, and abnormal performance data is identified through algorithms or manual inspection.

[0037] Step S12: Determine whether the historical industry data meets the industry data standards. If so, perform data analysis on the historical industry data to obtain corrected performance data.

[0038] Specifically, determine whether there is actual abnormal feedback in the historical industry data. If so, record the abnormal device nodes, abnormal device data and standard performance data corresponding to the abnormal device nodes that correspond to the actual abnormal feedback in the historical industry data. The standard performance data is the operating data that the device corresponding to the abnormal device node should output under different operating parameters. The abnormal device nodes, abnormal device data and standard performance data are planned according to the time series to obtain the abnormal feedback matrix. The standard performance data is corrected and analyzed based on the abnormal feedback matrix to obtain the corrected performance data corresponding to the abnormal device nodes.

[0039] Specifically, the abnormal device nodes, abnormal device data and standard performance data are information planned according to the time series to obtain an abnormal feedback matrix, the device association relationship between the abnormal device nodes is determined according to the abnormal device data, the data in the abnormal feedback matrix is ​​reorganized based on the device association relationship and the time series to obtain multiple groups of reorganized feedback matrices, the abnormal device data in each group of reorganized feedback matrices are analyzed for data anomalies, and the active abnormal device and the passive abnormal device in each group of reorganized feedback matrices are determined. The active abnormal device is the device that firstly has data anomalies in the reorganized feedback matrix according to the time series, and the passive abnormal device is the device that is not the first to have data anomalies in the reorganized feedback matrix according to the time series. The standard performance data is updated based on the abnormal data corresponding to the active abnormal device to obtain first performance data corresponding to the active abnormal device, and the passive abnormal device is independently detected to determine whether the abnormal cause of the passive abnormal device is caused by the active abnormal device and / or the adjacent abnormal device. If not, the standard performance data is updated based on the abnormal data corresponding to the passive abnormal device to obtain second performance data corresponding to the passive abnormal device, and the first performance data and the second performance data in each group of reorganized feedback matrices are summarized to obtain the corrected performance data corresponding to the abnormal device node.

[0040] Specifically, the data standard correspondence between the correction performance data and the industry data standard is determined based on the abnormal device nodes, the abnormal performance data corresponding to each model device node is determined based on the industry data standard, and data critical value analysis is performed on the correction performance data and the abnormal performance data according to the data standard correspondence to obtain the correction data standard.

[0041] Specifically, data critical value analysis is performed on the corrected performance data and the abnormal performance data according to the data standard correspondence to obtain the corrected data standard, including: determining the corrected data critical value according to the corrected performance data, determining the abnormal data critical value according to the abnormal performance data, and based on the data standard correspondence, updating the abnormal data critical value by narrowing the range according to the corrected data critical value to obtain the corrected data standard.

[0042] Step S13: updating the industry data standard based on the correction performance data to obtain the correction data standard.

[0043] Step S14: Correct and update the historical industry model according to the corrected performance data and the corrected data standard to obtain a corrected industry model.

[0044] In the embodiment of the present application, a historical industry model and its corresponding historical industry data are obtained. The historical industry model is obtained by training based on rich historical industry data, which are derived from the generation of different production monitoring equipment in the industry, ensuring the comprehensiveness and accuracy of the model. This step provides a solid foundation for the subsequent determination of industry data standards and model correction. The industry data standard is further determined according to the historical industry model. This standard, as a judgment benchmark, can accurately identify whether there are abnormal performance data in the historical industry data, thereby ensuring the accuracy and reliability of data analysis. The timely removal of abnormal data avoids its interference with the subsequent data analysis results. The historical industry data is strictly analyzed, and after confirming that it meets the industry data standard, the corrected performance data is obtained, thereby realizing the correction of the industry equipment performance data. The industry data standard is updated based on the corrected performance data, and a more accurate and adapted to the current industry status of the corrected data standard is obtained. Thereby making the data standard closer to the actual application of the equipment, providing a more accurate judgment benchmark for the subsequent industry model correction. According to the corrected performance data and the corrected data standard, the historical industry model is corrected and updated, and finally the corrected industry model is obtained, thereby improving the regulatory efficiency of the industry model.

[0045] Furthermore, the data in the abnormal feedback matrix is ​​reorganized based on the device association relationship and the time series to obtain multiple groups of reorganized feedback matrices, and then it also includes: inputting the multiple groups of reorganized feedback matrices into the trained vector extraction model respectively to extract vector features to obtain the number of matrix dimensions, combining the number of matrix dimensions with the multiple groups of reorganized feedback matrices in a one-to-one correspondence to obtain multiple groups of combined feedback matrices, processing the data contained in the multiple groups of combined feedback matrices to obtain comprehensive data, and inputting the comprehensive data into a preset deduction model for data deduction to obtain future abnormal nodes of the multiple groups of reorganized feedback matrices in the future cycle time period.

[0046] In an embodiment of the present application, the preset inference model is a bidirectional LSTM prediction model, including but not limited to this prediction model.

[0047] Specifically, the normal distribution mean and normal distribution variance of the data contained in the multiple groups of combined feedback matrices are calculated, and the 3σ range of the multiple groups of combined feedback matrices is determined based on the normal distribution mean and normal distribution variance. It is judged whether the data contained in the multiple groups of combined feedback matrices are outside the 3σ range. If the data contained in the multiple groups of combined feedback matrices are outside the 3σ range, the first matrix sequence of the multiple groups of combined feedback matrices where the data is located is determined, and the sequence average is calculated according to the first matrix sequence. The data is replaced with the sequence average to obtain the replaced second matrix sequence. The second matrix sequence is processed for missing values ​​and normalization to obtain comprehensive data.

[0048] The above-mentioned embodiment introduces an effective management method of industry models based on artificial intelligence from the perspective of method flow. The following embodiment introduces an effective management system of industry models based on artificial intelligence from the perspective of virtual modules or virtual units. Please see the following embodiments for details.

[0049] The embodiment of the present application provides an effective management system 20 of industry models based on artificial intelligence, such as Figure 2 As shown, Figure 2 A schematic diagram of the structure of an effective management system of industry models based on artificial intelligence provided in an embodiment of the present application. The system 20 may specifically include: The data acquisition module 21 is used to acquire historical industry models and historical industry data. The historical industry model is an industry model trained based on historical industry data. The historical industry data is data generated by different production monitoring equipment in the industry during a historical period. A standard determination module 22 is used to determine the industry data standard corresponding to the historical industry data according to the historical industry model, and the industry data standard is used to judge whether there is abnormal performance data that does not meet the preset data range in the historical industry data; The data analysis module 23 is used to determine whether the historical industry data meets the industry data standards. If so, the historical industry data is analyzed to obtain corrected performance data. A standard updating module 24, used to update the industry data standard based on the correction performance data to obtain the correction data standard; The model management module 25 is used to calibrate and update the historical industry model according to the calibrated performance data and the calibrated data standard to obtain a calibrated industry model.

[0050] In a possible implementation of the embodiment of the present application, when the data analysis module 23 performs data analysis on historical industry data to obtain corrected performance data, it is specifically used to: Determine whether there is actual abnormal feedback in the historical industry data. If so, record the abnormal device node, abnormal device data, and standard performance data corresponding to the abnormal device node corresponding to the actual abnormal feedback in the historical industry data. The standard performance data is the operating data that the device corresponding to the abnormal device node should output under different operating parameters; The abnormal device nodes, abnormal device data and standard performance data are planned according to the time series to obtain the abnormal feedback matrix; Based on the abnormal feedback matrix, the standard performance data is corrected and analyzed to obtain the corrected performance data corresponding to the abnormal device node.

[0051] In another possible implementation of the embodiment of the present application, when the data analysis module 22 corrects and analyzes the standard performance data based on the abnormal feedback matrix to obtain the corrected performance data corresponding to the abnormal device node, it is specifically used to: The abnormal device nodes, abnormal device data and standard performance data are planned according to the time series to obtain the abnormal feedback matrix; Determine the device association relationship between abnormal device nodes according to abnormal device data; Reorganize the data in the abnormal feedback matrix based on the equipment association relationship and time series to obtain multiple groups of reorganized feedback matrices; Perform data anomaly root cause analysis on the abnormal device data in each group of reorganized feedback matrices to determine the active abnormal devices and passive abnormal devices in each group of reorganized feedback matrices. The active abnormal device is the first device in the reorganized feedback matrix to have data anomalies according to the time sequence, and the passive abnormal device is the device that is not the first device in the reorganized feedback matrix to have data anomalies according to the time sequence. The standard performance data is updated based on the abnormal data corresponding to the active abnormal device to obtain the first performance data corresponding to the active abnormal device; Perform independent abnormality detection on the passive abnormal device to determine whether the abnormality of the passive abnormal device is caused by the active abnormal device and / or the adjacent abnormal device. If not, update the standard performance data with the abnormal data corresponding to the passive abnormal device to obtain second performance data corresponding to the passive abnormal device; The first performance data and the second performance data in each group of reorganized feedback matrices are aggregated to obtain corrected performance data corresponding to the abnormal device node.

[0052] In another possible implementation of the embodiment of the present application, when the standard updating module 24 updates the industry data standard based on the correction performance data to obtain the correction data standard, it is specifically used to: Determine the data standard correspondence between the correction performance data and the industry data standards based on the abnormal device node; Determine the abnormal performance data corresponding to each model device node based on industry data standards; According to the data standard correspondence, data critical value analysis is performed on the corrected performance data and abnormal performance data to obtain the corrected data standard.

[0053] In another possible implementation of the embodiment of the present application, when the standard updating module 24 performs data critical value analysis on the correction performance data and the abnormal performance data according to the data standard correspondence relationship to obtain the correction data standard, it is specifically used to: determining a calibration data cutoff value based on the calibration performance data; Determine the abnormal data critical value based on the abnormal performance data; Based on the data standard correspondence, the abnormal data critical value is updated by narrowing the range according to the correction data critical value to obtain the correction data standard.

[0054] In another possible implementation of the embodiment of the present application, the system 20 further includes: a vector extraction module, a data combination module, a data processing module and a data deduction module, wherein: A vector extraction module is used to input multiple groups of reorganized feedback matrices into the trained vector extraction model to extract vector features and obtain the number of matrix dimensions; A data combining module is used to combine the number of matrix dimensions with multiple groups of reorganized feedback matrices in a one-to-one correspondence to obtain multiple groups of combined feedback matrices; A data processing module, used for processing the data contained in the multiple groups of combined feedback matrices to obtain comprehensive data; The data deduction module is used to input the comprehensive data into the preset deduction model for data deduction, and obtain multiple groups of future abnormal nodes of the reorganized feedback matrix in the future cycle time period.

[0055] In another possible implementation of the embodiment of the present application, when the data processing module processes the data included in the multiple groups of combined feedback matrices to obtain the comprehensive data, it is specifically used to: Calculating the normal distribution mean and the normal distribution variance of the data contained in the multiple sets of combined feedback matrices, and determining the 3σ range of the multiple sets of combined feedback matrices based on the normal distribution mean and the normal distribution variance; Determine whether the data included in the multiple groups of combined feedback matrices are outside the 3σ range, and if the data included in the multiple groups of combined feedback matrices are outside the 3σ range, determine the first matrix sequence of the multiple groups of combined feedback matrices where the data is located; Calculate the sequence average value according to the first matrix sequence, replace the data with the sequence average value, and obtain the replaced second matrix sequence; The second matrix sequence is processed for missing values ​​and normalization to obtain comprehensive data.

[0056] Technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific working process of the above-described artificial intelligence-based industry model effective management system 20 can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0057] An electronic device is provided in an embodiment of the present application, such as Figure 3 As shown, Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 3The electronic device 300 shown includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the electronic device 300 does not constitute a limitation on the embodiments of the present application.

[0058] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0059] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but it does not mean that there is only one bus or only one type of bus.

[0060] The memory 303 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0061] The memory 303 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the contents shown in the above method embodiment.

[0062] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 3 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0063] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0064] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0065] The above are only some implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An effective management method for industry models based on artificial intelligence, characterized in that: include: Acquire a historical industry model and historical industry data, wherein the historical industry model is an industry model trained based on the historical industry data, and the historical industry data is data generated by different production monitoring equipment in the industry during a historical period; Determine, according to the historical industry model, an industry data standard corresponding to the historical industry data, wherein the industry data standard is used to judge whether there is abnormal performance data in the historical industry data that does not meet a preset data range; Determine whether the historical industry data meets the industry data standard, and if so, perform data analysis on the historical industry data to obtain corrected performance data; Updating the industry data standard based on the correction performance data to obtain the correction data standard; The historical industry model is corrected and updated according to the corrected performance data and the corrected data standard to obtain a corrected industry model.

2. According to the effective management method of industry model based on artificial intelligence in claim 1, it is characterized in that: The data analysis of the historical industry data to obtain the corrected performance data includes: Determine whether there is actual abnormal feedback in the historical industry data. If so, record the abnormal device node, abnormal device data, and standard performance data corresponding to the abnormal device node corresponding to the actual abnormal feedback in the historical industry data. The standard performance data is the operating data that the device corresponding to the abnormal device node should output under different operating parameters; Perform information planning on the abnormal device nodes, abnormal device data and the standard performance data according to a time series to obtain an abnormal feedback matrix; The standard performance data is corrected and analyzed based on the abnormal feedback matrix to obtain corrected performance data corresponding to the abnormal device node.

3. The effective management method of industry models based on artificial intelligence according to claim 2 is characterized in that: The correcting and analyzing the standard performance data based on the abnormal feedback matrix to obtain the corrected performance data corresponding to the abnormal device node includes: Perform information planning on the abnormal device nodes, abnormal device data and the standard performance data according to a time series to obtain an abnormal feedback matrix; Determine the device association relationship between the abnormal device nodes according to the abnormal device data; Reorganize the data in the abnormal feedback matrix based on the device association relationship and the time series to obtain multiple groups of reorganized feedback matrices; Performing data anomaly root cause analysis on the abnormal device data in each group of reorganized feedback matrices, and determining active abnormal devices and passive abnormal devices in each group of reorganized feedback matrices, wherein the active abnormal device is the first device in the reorganized feedback matrix to have data anomalies according to the time sequence, and the passive abnormal device is the device in the reorganized feedback matrix that is not the first device to have data anomalies according to the time sequence; The standard performance data is updated based on the abnormal data corresponding to the active abnormal device to obtain first performance data corresponding to the active abnormal device; Performing independent abnormality detection on the passive abnormal device to determine whether the abnormality of the passive abnormal device is caused by the active abnormal device and / or the adjacent abnormal device; if not, updating the standard performance data with the abnormal data corresponding to the passive abnormal device to obtain second performance data corresponding to the passive abnormal device; The first performance data and the second performance data in each group of reorganized feedback matrices are aggregated to obtain corrected performance data corresponding to the abnormal device node.

4. The effective management method of industry models based on artificial intelligence according to claim 1 is characterized in that: The updating of the industry data standard based on the correction performance data to obtain the correction data standard includes: Determining a data standard correspondence relationship between the correction performance data and the industry data standard based on the abnormal device node; Determine the abnormal performance data corresponding to each model device node based on the industry data standard; According to the data standard correspondence, data critical value analysis is performed on the correction performance data and the abnormal performance data to obtain the correction data standard.

5. The effective management method of industry models based on artificial intelligence according to claim 4 is characterized in that: The step of performing data critical value analysis on the corrected performance data and the abnormal performance data according to the data standard correspondence to obtain the corrected data standard includes: determining a correction data critical value based on the correction performance data; Determining an abnormal data critical value according to the abnormal performance data; Based on the data standard corresponding relationship, the abnormal data critical value is updated by narrowing the range according to the correction data critical value to obtain the correction data standard.

6. The effective management method of industry models based on artificial intelligence according to claim 5 is characterized in that: The data in the abnormal feedback matrix is ​​reorganized based on the device association relationship and the time series to obtain multiple groups of reorganized feedback matrices, and then the following further includes: Inputting the plurality of groups of reorganized feedback matrices into the trained vector extraction model to extract vector features, respectively, to obtain the number of matrix dimensions; The number of matrix dimensions and the multiple groups of reorganized feedback matrices are combined in a one-to-one correspondence process to obtain multiple groups of combined feedback matrices; Processing the data contained in the plurality of groups of combined feedback matrices to obtain comprehensive data; The comprehensive data is input into a preset deduction model for data deduction to obtain multiple groups of future abnormal nodes of the reorganized feedback matrix in future cycle time periods.

7. The effective management method of industry models based on artificial intelligence according to claim 6 is characterized in that: The processing of the data contained in the plurality of groups of combined feedback matrices to obtain comprehensive data includes: Calculating a normal distribution mean and a normal distribution variance of the data contained in the plurality of groups of combined feedback matrices, and determining a 3σ range of the plurality of groups of combined feedback matrices based on the normal distribution mean and the normal distribution variance; Determining whether the data included in the multiple groups of combined feedback matrices are outside the 3σ range, and if the data included in the multiple groups of combined feedback matrices are outside the 3σ range, determining a first matrix sequence of the multiple groups of combined feedback matrices where the data is located; Calculate a sequence average value according to the first matrix sequence, and replace the data with the sequence average value to obtain a replaced second matrix sequence; The second matrix sequence is processed for missing values ​​and normalization to obtain comprehensive data.

8. An effective management system for industry models based on artificial intelligence, characterized in that: include: A data acquisition module, used to acquire a historical industry model and historical industry data, wherein the historical industry model is an industry model trained based on the historical industry data, and the historical industry data is data generated by different production monitoring equipment in the industry during a historical period; A standard determination module, used to determine the industry data standard corresponding to the historical industry data according to the historical industry model, wherein the industry data standard is used to judge whether there is abnormal performance data in the historical industry data that does not meet a preset data range; A data analysis module, used to determine whether the historical industry data meets the industry data standard, and if so, to perform data analysis on the historical industry data to obtain corrected performance data; A standard updating module, used for updating the industry data standard based on the correction performance data to obtain the correction data standard; The model management module is used to calibrate and update the historical industry model according to the calibrated performance data and the calibrated data standard to obtain a calibrated industry model.

9. An electronic device, characterized in that: It includes: One or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to: execute an artificial intelligence-based industry model effective management method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, it implements an effective management method of industry models based on artificial intelligence as described in any one of claims 1 to 7.

Images