Industrial asset health monitoring management system and method based on AI

Abstract

The invention relates to the technical field of industrial asset monitoring, and particularly discloses an industrial asset health monitoring management system and method based on AI, and the system comprises a graph construction module which is used for obtaining historical industrial fault data, and constructing an industrial asset knowledge graph according to the historical industrial fault data; the health evaluation module is used for acquiring current industrial asset data and performing health degree evaluation on the current industrial asset data according to the industrial asset knowledge graph; and the health early warning module is used for monitoring the current industrial asset data in real time, and performing data health early warning according to the real-time monitoring value of the current industrial asset data in combination with the health degree evaluation result. The method can accurately evaluate the health state of the industrial asset data, carries out the real-time monitoring and early warning of the data in combination with the health state, and guarantees the safe and stable operation of the industrial assets.

Description

Technical Field

[0001] The present application relates to the field of industrial asset monitoring technology, and more specifically, to an AI-based industrial asset health monitoring management system and method. Background Art

[0002] Industrial assets are digital representations of industrial equipment, processes, and facilities. They can include PLCs, entire machine tools, production lines, rolling mills, vehicles, motors, and more. Each asset has associated data. For example, a piece of equipment might have a serial number, location, make and model, and installation date. It might also have time-series values for availability, performance, quality, temperature, pressure, and more. The management of industrial assets is crucial for improving production efficiency, reducing costs, and ensuring product quality. Effective asset management can help companies better understand the status of their assets.

[0003] Understanding the health of industrial assets is critical to achieving production and business goals. In existing technologies, the health monitoring of industrial assets relies heavily on manual measurements, which are not very frequent, perhaps once a month or quarter. The system only focuses on data collection, and data analysis must be manually performed by trained experts. Summary of the Invention

[0004] The present invention provides an AI-based industrial asset health monitoring management system and method to address the low efficiency of industrial asset health monitoring management in the prior art, including: A graph construction module is used to obtain historical industrial failure data and build an industrial asset knowledge graph based on the historical industrial failure data; The health evaluation module is used to obtain current industrial asset data and evaluate the health of the current industrial asset data based on the industrial asset knowledge graph; The health warning module is used to monitor the current industrial asset data in real time and issue data health warnings based on the real-time monitoring values of the current industrial asset data combined with the health evaluation results.

[0005] Furthermore, the graph construction module constructs an industrial asset knowledge graph based on historical industrial failure data, including: Discretize historical industrial failure data to obtain discretized industrial failure data; Based on the apriori algorithm, frequent item sets of discrete industrial failure data are mined, and strong association rules of historical industrial failure data are determined according to the frequent item sets of discrete industrial failure data. Construct an industrial asset knowledge graph based on the strong association rules of historical industrial failure data.

[0006] Furthermore, the method of mining frequent itemsets of the discretized industrial failure data based on the Apriori algorithm and determining strong association rules of the historical industrial failure data according to the frequent itemsets of the discretized industrial failure data includes: Construct the first candidate item set based on the discretized industrial fault data, set the minimum support threshold, and determine the item sets in the first candidate item set whose support is greater than the preset support threshold as frequent item sets; Construct a second candidate item set based on the frequent item set, and calculate the confidence and lift of the second candidate item set; A minimum confidence threshold and a minimum lift threshold are set, and association rules in the second candidate set whose confidence is greater than or equal to the minimum confidence threshold and whose lift is greater than the minimum lift threshold are determined as strong association rules of historical industrial failure data.

[0007] Furthermore, the construction of an industrial asset knowledge graph based on strong association rules of historical industrial failure data includes: The edge weights between data entities are calculated based on the strong association rules of historical industrial failure data. The data entities are linked according to the edge weights between them to obtain the industrial asset knowledge graph. The edge weight calculation formula is as follows:

[0008] in, is the edge weight between data entities, is the degree of improvement between data entities, is the number of occurrences of the data entity in all failure events, Tolerance for preset fault times, is the preset range coefficient.

[0009] Furthermore, the health evaluation module performs health evaluation on the current industrial asset data based on the industrial asset knowledge graph, including: Cluster the industrial asset knowledge graph based on the k-means clustering algorithm to obtain k industrial asset data categories. The weight value corresponding to each industrial asset data category is determined according to the severity of the fault event corresponding to the industrial asset data category. Obtain current industrial asset data, and calculate the ratio of the current industrial asset data to the corresponding industrial asset data in the industrial asset data category; The sum of the ratios of each industrial asset data in the industrial asset data category is calculated to obtain the matching degree of the industrial asset data category; According to the weight value corresponding to each industrial asset data category, the matching degrees of the industrial asset data categories corresponding to k industrial asset data categories are weighted and summed to obtain the comprehensive industrial asset data matching degree, and the health of the industrial asset data is evaluated based on the comprehensive industrial asset data matching degree.

[0010] Furthermore, clustering the industrial asset knowledge graph based on the K-means clustering algorithm includes: Randomly select k initial cluster centers of data entities in the industrial asset knowledge graph, calculate the Euclidean distance between the data entities in the industrial asset knowledge graph and the initial cluster centers, and divide the data entities into corresponding partitions based on the Euclidean distance between the data entities in the industrial asset knowledge graph and the initial cluster centers; Calculate the mean of the data entities in each partition, and recalculate the cluster center based on the mean of the data entities in each partition; Repeat the above steps until the cluster center no longer changes or the number of iterations reaches the maximum number of iterations, and obtain k industrial asset data categories.

[0011] Furthermore, the health evaluation of the industrial asset data according to the matching degree of the comprehensive industrial asset data includes: Obtaining the matching degree of the preset standard, and calculating the difference between the matching degree of the comprehensive industrial asset data and the matching degree of the preset standard; Determining whether a difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than a first preset threshold, and if the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than the first preset threshold, determining the health of the current industrial asset data to be a first health level; If the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is less than or equal to the first preset threshold, then determining whether the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than a second preset threshold; If the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than a second preset threshold, determining the health of the current industrial asset data to be a second health; If the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is less than or equal to the second preset threshold, the health level of the current industrial asset data is determined to be the third health level.

[0012] Furthermore, the health warning module performs data health warning based on the real-time monitoring value of the current industrial asset data in combination with the health evaluation result, including: Obtaining real-time monitoring values of industrial asset data, and drawing an industrial asset data change curve based on the real-time monitoring values of the industrial asset data; Perform curve fitting on the industrial asset data change curve to obtain the curve fitting equation of the industrial asset data; Determine the equation correction coefficient based on the health evaluation results of industrial asset data, and correct the curve fitting equation based on the equation correction coefficient; Predicting industrial asset data based on the curve fitting equation of the corrected industrial asset data, and determining the predicted value of the matching degree of the comprehensive industrial asset data based on the prediction results; Determine whether the difference between the predicted value of the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than the fourth preset threshold. If the difference between the predicted value of the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than the fourth preset threshold, issue a data health warning.

[0013] Furthermore, the correction of the curve fitting equation according to the equation correction coefficient includes: The equation correction coefficient is multiplied by the linear regression coefficient in the curve fitting equation to obtain the corrected linear regression coefficient, and the corrected curve fitting equation is determined based on the corrected linear regression coefficient.

[0014] To achieve the above objectives, the present invention also provides an AI-based industrial asset health monitoring and management method, comprising: Obtain historical industrial failure data and build an industrial asset knowledge graph based on the historical industrial failure data; Obtain current industrial asset data and evaluate its health based on the industrial asset knowledge graph; Conduct real-time monitoring of current industrial asset data, and issue data health warnings based on the real-time monitoring values of current industrial asset data combined with health evaluation results.

[0015] The beneficial effects of the present invention are: By applying the above technical solutions, the present invention constructs an industrial asset knowledge graph through historical industrial failure data, associates industrial asset data, solves the problems of scattered, diverse, complex and isolated industrial asset data, evaluates the health of industrial asset data through the industrial asset knowledge graph, and conducts real-time monitoring and early warning of industrial asset data in combination with the health evaluation, thereby improving the health monitoring and management efficiency of industrial assets and ensuring the safe and stable operation of industrial assets. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A schematic diagram of the structure of an AI-based industrial asset health monitoring and management system proposed in an embodiment of the present invention is shown; Figure 2The figure shows an overall flow chart of an AI-based industrial asset health monitoring and management method proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0019] The present application embodiment provides an AI-based industrial asset health monitoring and management system, such as Figure 1 As shown, it includes: a graph construction module, which is used to obtain historical industrial failure data and construct an industrial asset knowledge graph based on the historical industrial failure data; a health evaluation module, which is used to obtain current industrial asset data and evaluate the health of the current industrial asset data based on the industrial asset knowledge graph; a health warning module, which is used to monitor the current industrial asset data in real time and issue data health warnings based on the real-time monitoring values of the current industrial asset data combined with the health evaluation results.

[0020] In this embodiment, the graph construction module obtains historical industrial failure data by extracting the failure data of industrial assets in historical failure events and constructs an industrial asset knowledge graph, so that the health evaluation module can evaluate the health of the current industrial asset data according to the industrial asset knowledge graph, monitor the current industrial asset data in real time through the health warning module, and issue data health warnings for the industrial asset data through real-time monitoring values to ensure the stable operation of industrial assets.

[0021] In some embodiments of the present application, the graph construction module constructs an industrial asset knowledge graph based on historical industrial failure data, including: discretizing the historical industrial failure data to obtain discretized industrial failure data; mining frequent item sets of the discretized industrial failure data based on the apriori algorithm, and determining strong association rules of the historical industrial failure data based on the frequent item sets of the discretized industrial failure data; and constructing an industrial asset knowledge graph based on the strong association rules of the historical industrial failure data.

[0022] In some embodiments of the present application, the apriori algorithm is used to mine frequent item sets of discrete industrial failure data, and strong association rules of historical industrial failure data are determined based on the frequent item sets of the discrete industrial failure data, including: constructing a first candidate item set based on the discrete industrial failure data, setting a minimum support threshold, and determining the item sets in the first candidate item set whose support is greater than the preset support threshold as frequent item sets; constructing a second candidate item set based on the frequent item sets, and calculating the confidence and lift of the second candidate item set; setting a minimum confidence threshold and a minimum lift threshold, and determining the association rules in the second candidate item set whose confidence is greater than or equal to the minimum confidence threshold and whose lift is greater than the minimum lift threshold as strong association rules of historical industrial failure data.

[0023] In this embodiment, the data range of the industrial asset data when the historical fault event occurred is collected to obtain the historical industrial fault data. After the historical industrial fault data is discretized, the strong association rules therein are mined based on the apriori algorithm, thereby constructing an industrial asset knowledge graph and improving the closeness of the association between data entities.

[0024] In some embodiments of the present application, the method of constructing an industrial asset knowledge graph based on the strong association rules of historical industrial failure data includes: calculating the edge weights between each data entity based on the strong association rules of the historical industrial failure data, linking each data entity based on the edge weights between each data entity, and obtaining an industrial asset knowledge graph. The edge weight calculation formula is specifically:

[0025] in, is the edge weight between data entities, is the degree of improvement between data entities, is the number of occurrences of the data entity in all failure events, Tolerance for preset fault times, is the preset range coefficient.

[0026] In this embodiment, data entities with strong association relationships are linked, and the association strength between data entities is calculated based on the edge weight calculation formula. The edge weights between data entities are calculated by increasing the degree of association. The number of occurrences of the data entity in all failure events Derive, through Adjust the data range of the final edge weight to make the calculated edge weight more reasonable.

[0027] In some embodiments of the present application, the health evaluation module performs a health evaluation on the current industrial asset data based on the industrial asset knowledge graph, including: clustering the industrial asset knowledge graph based on the k-means clustering algorithm to obtain k industrial asset data categories, and determining the weight value corresponding to each industrial asset data category according to the severity of the fault event corresponding to the industrial asset data category; obtaining the current industrial asset data, and calculating the ratio of the current industrial asset data to the corresponding industrial asset data in the industrial asset data category; statistically calculating the sum of the ratios of each industrial asset data in the industrial asset data category to obtain the matching degree of the industrial asset data category; performing weighted summation of the matching degrees of the industrial asset data categories corresponding to the k industrial asset data categories according to the weight value corresponding to each industrial asset data category to obtain the comprehensive industrial asset data matching degree, and performing a health evaluation on the industrial asset data according to the comprehensive industrial asset data matching degree.

[0028] In this embodiment, the data entities in the industrial asset knowledge graph are clustered to obtain the industrial asset data category to which each data entity belongs. The weight value corresponding to each industrial asset data category is determined by the severity of the fault event corresponding to the industrial asset data category. The higher the severity, the larger the corresponding weight value. The comprehensive industrial asset data matching degree is obtained by the industrial asset data category matching degree of the current industrial asset data and the corresponding weight value. The current industrial asset data is matched with the industrial asset knowledge graph to facilitate the subsequent health evaluation of the industrial asset data.

[0029] In some embodiments of the present application, the clustering of the industrial asset knowledge graph based on the k-means clustering algorithm includes: randomly selecting k initial clustering centers of data entities in the industrial asset knowledge graph, calculating the Euclidean distance between the data entities in the industrial asset knowledge graph and the initial clustering centers, and dividing the data entities into corresponding partitions according to the Euclidean distance between the data entities in the industrial asset knowledge graph and the initial clustering centers; calculating the mean of the data entities in each partition, and recalculating the clustering centers according to the mean of the data entities in each partition; repeating the above steps until the clustering centers no longer change or the number of iterations reaches the maximum number of iterations, and obtaining k industrial asset data categories.

[0030] In this embodiment, clustering of the industrial asset knowledge graph is implemented based on the k-means clustering algorithm. The number of initial cluster centers k is set based on the number of data entities. The larger the number, the larger the corresponding k value.

[0031] In some embodiments of the present application, the health evaluation of industrial asset data based on the matching degree of comprehensive industrial asset data includes: obtaining the matching degree of preset standards, calculating the difference between the matching degree of comprehensive industrial asset data and the matching degree of preset standards; judging whether the difference between the matching degree of comprehensive industrial asset data and the matching degree of preset standards is greater than a first preset threshold value; if the difference between the matching degree of comprehensive industrial asset data and the matching degree of preset standards is greater than the first preset threshold value, determining that the health of the current industrial asset data is the first health degree; if the difference between the matching degree of comprehensive industrial asset data and the matching degree of preset standards is less than or equal to the first preset threshold value, judging whether the difference between the matching degree of comprehensive industrial asset data and the matching degree of preset standards is greater than a second preset threshold value; if the difference between the matching degree of comprehensive industrial asset data and the matching degree of preset standards is greater than the second preset threshold value, determining that the health of the current industrial asset data is the second health degree; if the difference between the matching degree of comprehensive industrial asset data and the matching degree of preset standards is less than or equal to the second preset threshold value, determining that the health of the current industrial asset data is the third health degree.

[0032] In this embodiment, the health of the current industrial asset data is obtained by comprehensively calculating the difference between the matching degree of the industrial asset data and the matching degree of the preset standard. The larger the difference, the lower the corresponding health.

[0033] In some embodiments of the present application, the health warning module performs data health warning based on the real-time monitoring value of the current industrial asset data in combination with the health evaluation result, including: obtaining the real-time monitoring value of the industrial asset data, and drawing the industrial asset data change curve based on the real-time monitoring value of the industrial asset data; performing curve fitting on the industrial asset data change curve to obtain the curve fitting equation of the industrial asset data; determining the equation correction coefficient based on the health evaluation result of the industrial asset data, and correcting the curve fitting equation based on the equation correction coefficient; predicting the industrial asset data based on the curve fitting equation of the corrected industrial asset data, and determining the comprehensive industrial asset data matching degree prediction value based on the prediction result; judging whether the difference between the comprehensive industrial asset data matching degree prediction value and the preset standard matching degree is greater than a fourth preset threshold value, and if the difference between the comprehensive industrial asset data matching degree prediction value and the preset standard matching degree is greater than the fourth preset threshold value, then performing a data health warning.

[0034] In some embodiments of the present application, the correction of the curve fitting equation according to the equation correction coefficient includes: multiplying the equation correction coefficient with the linear regression coefficient in the curve fitting equation to obtain a corrected linear regression coefficient, and determining the corrected curve fitting equation according to the corrected linear regression coefficient.

[0035] In this embodiment, an industrial asset data change curve is drawn through the real-time monitoring value of the industrial asset data, and the industrial asset data change curve is curve-fitted based on the least squares method. At the same time, the equation correction coefficient is determined according to the health evaluation result of the industrial asset data. The setting range of the equation correction coefficient is between 0 and 2. The lower the health, the larger the set equation correction coefficient. The industrial asset data is predicted by the curve fitting equation of the corrected industrial asset data, so as to obtain the comprehensive industrial asset data matching degree prediction value. The fourth preset threshold is set to be greater than the first preset threshold. When the difference between the comprehensive industrial asset data matching degree prediction value and the preset standard matching degree is greater than the fourth preset threshold, a data health warning is issued, thereby realizing real-time health monitoring of industrial asset data and ensuring the safe and stable operation of industrial assets.

[0036] Based on the same technical concept, such as Figure 2 As shown, the present invention also provides an AI-based industrial asset health monitoring and management method, including: S101, obtaining historical industrial failure data and constructing an industrial asset knowledge graph based on the historical industrial failure data; S102, obtaining current industrial asset data and performing a health evaluation on the current industrial asset data based on the industrial asset knowledge graph; S103, real-time monitoring of current industrial asset data, and data health warning based on the real-time monitoring value of the current industrial asset data combined with the health evaluation result.

[0037] By applying the above technical solutions, the present invention uses a graph construction module to obtain historical industrial failure data and construct an industrial asset knowledge graph based on this historical industrial failure data; a health assessment module to obtain current industrial asset data and perform a health assessment on the current industrial asset data based on the industrial asset knowledge graph; and a health warning module to monitor the current industrial asset data in real time and issue a data health warning based on the real-time monitoring values of the current industrial asset data combined with the health assessment results. This invention can accurately evaluate the health status of industrial asset data and conduct real-time monitoring and warning of the data based on the health status, thus ensuring the safe and stable operation of industrial assets.

[0038] Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented via hardware or via software combined with a necessary general-purpose hardware platform. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product. This software product can be stored on a non-volatile storage medium (such as a CD-ROM, USB flash drive, or external hard drive) and includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute the methods described in various implementation scenarios of the present invention.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An AI-based industrial asset health monitoring and management system, characterized by: include: A graph construction module is used to obtain historical industrial failure data and build an industrial asset knowledge graph based on the historical industrial failure data; The health evaluation module is used to obtain current industrial asset data and evaluate the health of the current industrial asset data based on the industrial asset knowledge graph; The health warning module is used to monitor the current industrial asset data in real time and issue data health warnings based on the real-time monitoring values of the current industrial asset data combined with the health evaluation results.

2. The AI-based industrial asset health monitoring and management system according to claim 1 is characterized in that: The graph construction module constructs an industrial asset knowledge graph based on historical industrial failure data, including: Discretize historical industrial failure data to obtain discretized industrial failure data; Based on the Apriori algorithm, frequent item sets of discrete industrial failure data are mined, and strong association rules of historical industrial failure data are determined according to the frequent item sets of discrete industrial failure data. Construct an industrial asset knowledge graph based on the strong association rules of historical industrial failure data.

3. The AI-based industrial asset health monitoring and management system according to claim 2 is characterized in that: The method of mining frequent itemsets of discretized industrial failure data based on the Apriori algorithm and determining strong association rules of historical industrial failure data based on the frequent itemsets of the discretized industrial failure data includes: Construct the first candidate item set based on the discretized industrial fault data, set the minimum support threshold, and determine the item sets in the first candidate item set whose support is greater than the preset support threshold as frequent item sets; Construct a second candidate item set based on the frequent item set, and calculate the confidence and lift of the second candidate item set; A minimum confidence threshold and a minimum lift threshold are set, and association rules in the second candidate set whose confidence is greater than or equal to the minimum confidence threshold and whose lift is greater than the minimum lift threshold are determined as strong association rules of historical industrial failure data.

4. The AI-based industrial asset health monitoring and management system according to claim 3 is characterized in that: The construction of the industrial asset knowledge graph based on the strong association rules of historical industrial failure data includes: The edge weights between data entities are calculated based on the strong association rules of historical industrial failure data. The data entities are linked according to the edge weights between them to obtain the industrial asset knowledge graph. The edge weight calculation formula is as follows: in, is the edge weight between data entities, is the degree of improvement between data entities, is the number of occurrences of the data entity in all failure events, Tolerance for preset fault times, is the preset range coefficient.

5. The AI-based industrial asset health monitoring and management system according to claim 1 is characterized in that: The health evaluation module evaluates the health of current industrial asset data based on the industrial asset knowledge graph, including: Cluster the industrial asset knowledge graph based on the k-means clustering algorithm to obtain k industrial asset data categories. The weight value corresponding to each industrial asset data category is determined according to the severity of the fault event corresponding to the industrial asset data category. Obtain current industrial asset data, and calculate the ratio of the current industrial asset data to the corresponding industrial asset data in the industrial asset data category; The sum of the ratios of each industrial asset data in the industrial asset data category is calculated to obtain the matching degree of the industrial asset data category; According to the weight value corresponding to each industrial asset data category, the matching degrees of the industrial asset data categories corresponding to k industrial asset data categories are weighted and summed to obtain the comprehensive industrial asset data matching degree, and the health of the industrial asset data is evaluated based on the comprehensive industrial asset data matching degree.

6. The AI-based industrial asset health monitoring and management system according to claim 5 is characterized in that: The clustering of the industrial asset knowledge graph based on the K-means clustering algorithm includes: Randomly select k initial cluster centers of data entities in the industrial asset knowledge graph, calculate the Euclidean distance between the data entities in the industrial asset knowledge graph and the initial cluster centers, and divide the data entities into corresponding partitions based on the Euclidean distance between the data entities in the industrial asset knowledge graph and the initial cluster centers; Calculate the mean of the data entities in each partition, and recalculate the cluster center based on the mean of the data entities in each partition; Repeat the above steps until the cluster center no longer changes or the number of iterations reaches the maximum number of iterations, and obtain k industrial asset data categories.

7. The AI-based industrial asset health monitoring and management system according to claim 5 is characterized in that: The health evaluation of industrial asset data based on the matching degree of comprehensive industrial asset data includes: Obtaining the matching degree of the preset standard, and calculating the difference between the matching degree of the comprehensive industrial asset data and the matching degree of the preset standard; Determining whether a difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than a first preset threshold, and if the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than the first preset threshold, determining the health of the current industrial asset data to be a first health level; If the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is less than or equal to the first preset threshold, then determining whether the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than a second preset threshold; If the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than a second preset threshold, determining the health of the current industrial asset data to be a second health; If the difference between the comprehensive industrial asset data matching degree and the preset standard matching degree is less than or equal to the second preset threshold, the health level of the current industrial asset data is determined to be the third health level.

8. The AI-based industrial asset health monitoring and management system according to claim 7 is characterized in that: The health warning module performs data health warning based on the real-time monitoring value of the current industrial asset data combined with the health evaluation results, including: Obtaining real-time monitoring values of industrial asset data, and drawing an industrial asset data change curve based on the real-time monitoring values of the industrial asset data; Perform curve fitting on the industrial asset data change curve to obtain the curve fitting equation of the industrial asset data; Determine the equation correction coefficient based on the health evaluation results of industrial asset data, and correct the curve fitting equation based on the equation correction coefficient; Predicting industrial asset data based on the curve fitting equation of the corrected industrial asset data, and determining the predicted value of the matching degree of the comprehensive industrial asset data based on the prediction results; Determine whether the difference between the predicted value of the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than the fourth preset threshold. If the difference between the predicted value of the comprehensive industrial asset data matching degree and the preset standard matching degree is greater than the fourth preset threshold, issue a data health warning.

9. The AI-based industrial asset health monitoring and management system according to claim 8, characterized in that: The step of correcting the curve fitting equation according to the equation correction coefficient includes: The equation correction coefficient is multiplied by the linear regression coefficient in the curve fitting equation to obtain the corrected linear regression coefficient, and the corrected curve fitting equation is determined based on the corrected linear regression coefficient.

10. An AI-based industrial asset health monitoring and management method, characterized in that: include: Obtain historical industrial failure data and build an industrial asset knowledge graph based on the historical industrial failure data; Obtain current industrial asset data and evaluate its health based on the industrial asset knowledge graph; Conduct real-time monitoring of current industrial asset data, and issue data health warnings based on the real-time monitoring values of current industrial asset data combined with health evaluation results.

Images