An industrial control computer status testing system and method based on artificial intelligence

Abstract

The invention discloses an industrial control computer status test system and method based on artificial intelligence, belonging to the field of artificial intelligence technology. The invention comprises the following steps: matching historical test records with historical operation records to obtain associated records; dividing them into normal associated records and abnormal associated records; obtaining normal characteristic values ​​and abnormal characteristic values; obtaining the influence items of controlled equipment on target computers; obtaining the associated characteristics between the target computer and the controlled equipment; obtaining a comprehensive training set from the associated records; constructing a neural network model to carry out fault warning for the target computer; the invention can not only test the status of the industrial control computer through the test system itself, but also indirectly reflect the status of the industrial control computer through the operation data of the industrial equipment controlled by the industrial control computer, comprehensively understand the health status of the industrial control computer, improve the early warning of industrial control computer failure, reduce the risk of failure, and improve production efficiency.

Description

Technical Field

[0001] The present invention belongs to the technical field of artificial intelligence, and in particular relates to an industrial control computer status testing system and method based on artificial intelligence. Background Art

[0002] Industrial control computers, or industrial control computers, are computer systems specifically used in industrial environments. They are mainly used for industrial process control and data acquisition. They monitor, control and manage production equipment and processes by connecting to sensors, controllers and other equipment. Compared with ordinary commercial computers, industrial control computers have higher stability, durability and anti-interference capabilities.

[0003] The existing test system obtains the status of the industrial computer by monitoring the CPU usage, memory usage, temperature, power supply status and other test items of the industrial computer. This status test method focuses on checking the status of the industrial computer itself, ignoring the impact between the industrial computer and the industrial equipment it controls. It is impossible to fully understand the health status of the industrial computer, resulting in insufficient early warning of industrial computer failures by the test system, increasing the risk of failures and affecting production efficiency.

[0004] Therefore, people are in urgent need of an industrial control computer status testing system based on artificial intelligence to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to provide an industrial control computer status test system based on artificial intelligence to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] An industrial control computer status testing method based on artificial intelligence, the method comprising the following steps:

[0008] S1. Obtain historical test records of the target computer and historical operation records of the controlled device; match the historical test records with the historical operation records to obtain associated records;

[0009] S2. Mark the test records in the historical test records, and divide the associated records into normal associated records and abnormal associated records according to the test records included in the associated records;

[0010] S3, extracting the collection item data in the operation record, calculating the collection item data corresponding to the associated record, and obtaining normal characteristic values ​​and abnormal characteristic values; screening the collection items of the controlled device according to the normal characteristic values ​​and abnormal characteristic values, and obtaining the influence items of the controlled device on the target computer;

[0011] S4. According to the abnormal association record, the affected items are marked, and the items affecting the target computer are matched with the controlled device to obtain the association features between the target computer and the controlled device; according to the association features, a comprehensive training set is obtained from the association record;

[0012] S5. Construct a neural network model, train the neural network through a comprehensive training set, and issue fault warnings for the target computer.

[0013] According to the above technical solution, step S1 includes the following:

[0014] S1-1, each time a status test is performed on the target computer, a test record is generated, and all test records are extracted into the database to form historical test records;

[0015] The historical test records include test time, test items, test item values ​​and test results; the test items include CPU, memory, temperature, power supply and network; wherein, in each historical test record, the test items executed on the target computer are the same, each test item corresponds to a test result, and the test results include normal data and abnormal data;

[0016] S1-2. Each time a pair of controlled devices is subjected to data collection, an operation record is generated, and all operation records are extracted into the database to form a historical operation record;

[0017] The historical operation record includes the collection time, collection items and collection item values; the collection items include current, voltage, vibration frequency, operation frequency, rotation speed, load rate, temperature and humidity and data transmission rate; wherein, in each historical test record, the collection items executed for the same controlled device are the same;

[0018] S1-3, performing time matching according to the test time of the historical test record and the collection time of the historical operation record, taking the test time and the collection time being the same time point as a matching condition, matching the test record and the operation record, and marking the successfully matched test record and operation record as associated records;

[0019] The associated record includes a test record and several operation records, and each operation record corresponds to a controlled device;

[0020] By recording test data and operation data and matching these data into related records, an effective data basis is provided for obtaining the characteristic relationship between the target computer and the controlled device, which is helpful for subsequent characteristic analysis.

[0021] According to the above technical solution, step S2 includes the following:

[0022] S2-1. According to the historical test records, if in the test items of a certain test record, the test results corresponding to the test items are all normal data, then the test record is marked as a normal test record;

[0023] If, among the test items of a test record, any test result corresponding to any test item is abnormal data, the test record is marked as an abnormal test record;

[0024] S2-2, classifying the associated records according to the normal test records and the abnormal test records;

[0025] If the test record in a certain associated record is a normal test record, the associated record is classified as a normal associated record;

[0026] If a test record in a certain associated record is an abnormal test record, the associated record is classified as an abnormal associated record;

[0027] By classifying normal and abnormal test records, the two types of related records are effectively distinguished, laying the foundation for subsequent feature extraction and analysis.

[0028] According to the above technical solution, step S3 includes the following:

[0029] S3-1. Extract the collected item data in the operation record according to the historical operation record;

[0030] Match each operation record in all normal association records, and match all operation records of the same controlled device in all normal association records based on the operation record of the same controlled device;

[0031] According to all operation records of the same controlled device in all normal association records, the average value of the collection item data corresponding to the same collection item of the same controlled device in all normal association records is calculated as the normal characteristic value of the collection item of the controlled device;

[0032] Match each operation record in all abnormal association records, and match all operation records of the same controlled device in all abnormal association records based on the operation record of the same controlled device;

[0033] According to all operation records of the same controlled device in all abnormal association records, the average value of the collection item data corresponding to the same collection item of the same controlled device in all abnormal association records is calculated as the abnormal characteristic value of the collection item of the controlled device;

[0034] S3-2, calculating the ratio of the difference between the normal feature value and the abnormal feature value of the same collection item of the same controlled device to the normal feature value of the collection item of the controlled device; setting a difference ratio threshold, if the ratio of the difference between the normal feature value and the abnormal feature value of the same collection item of the same controlled device to the normal feature value of the collection item of the controlled device is greater than or equal to the difference ratio threshold, then the collection item of the controlled device is used as an influence item of the controlled device on the target computer;

[0035] For the collection items in each operation record included in all the abnormal association records, combined with the controlled devices corresponding to each operation record, the impact items in each operation record included in all the abnormal association records are marked;

[0036] Normal eigenvalues ​​and abnormal eigenvalues ​​are extracted by calculating the collection item data in the associated records, and the influencing items are determined by setting the difference ratio threshold. By extracting normal eigenvalues ​​and abnormal eigenvalues, the influence of noise can be eliminated, and the universality of the changes in the collection item data can be better seen through normal eigenvalues ​​and abnormal eigenvalues, so that the influencing items can be obtained more accurately.

[0037] According to the above technical solution, step S4 includes the following:

[0038] S4-1. Mark the test item whose test result in the test record included in a certain abnormality association record is a test item with data abnormality as an affected item;

[0039] Matching the affected item in the abnormal association record with the impact item corresponding to the running record contained in the abnormal association record, wherein the matching process is to use the impact item corresponding to the running record contained in the abnormal association record as the associated impact item of the affected item in the abnormal association record;

[0040] Count all associated impact items of different affected items in all abnormal associated records;

[0041] The same affected item and all the associated affected items corresponding thereto are used as a set of associated features between the target computer and the controlled device;

[0042] S4-2, according to a certain set of association features between the target computer and the controlled device, extract the collection item data corresponding to the set of association features from the normal association records to form a normal training set corresponding to the set of association features; extract the collection item data corresponding to the set of association features from the abnormal association records to form an abnormal training set corresponding to the set of association features;

[0043] After marking the data collected in the abnormal training set corresponding to the group of associated features, the data are combined with the normal training set corresponding to the group of associated features to form a comprehensive training set for the group of associated features;

[0044] By analyzing the associated features to construct a comprehensive training set, a reliable training data source is provided for the neural network model.

[0045] According to the above technical solution, step S5 includes the following:

[0046] Construct a neural network model and train the neural network model using a comprehensive training set of a set of related features;

[0047] The trained neural network model outputs the state information of the test items corresponding to the group of associated features according to the collection items corresponding to the group of associated features input; the state information includes normal state and abnormal state;

[0048] If the neural network model outputs that the state of the test item corresponding to the group of associated features is abnormal, fault warning information of the target computer is generated;

[0049] In the model training stage, a normal training set is formed by extracting the collection item data of the group of associated features from the normal associated records, and an abnormal training set is formed by extracting the collection item data of the group of associated features from the abnormal associated records. The data of the abnormal training set is marked and combined with the normal training set to form a comprehensive training set. The neural network model is trained using the comprehensive training set so that the model can identify the normal state and abnormal state of the corresponding test items.

[0050] Analyzing the impact between industrial computers and the industrial equipment they control can not only test the status of industrial computers through the test system itself, but also indirectly reflect the status of industrial computers through the operating data of industrial equipment controlled by industrial computers, so as to fully understand the health status of industrial computers, improve the early warning of industrial computer failures, reduce the risk of failures, and improve production efficiency;

[0051] Training the neural network based on the comprehensive training set and performing prediction and early warning on the target industrial control computer, combined with historical data and real-time monitoring, can effectively improve the accuracy and timeliness of the model.

[0052] An industrial control computer status test system based on artificial intelligence, the system includes an information acquisition module, an information processing module and an execution module;

[0053] The information acquisition module is used to acquire the information required by the system; the information processing module is used to store, analyze and transmit the information of each module; and the execution module is used to execute the information of the information processing module.

[0054] According to the above technical solution, the information acquisition module includes an industrial control computer unit and a controlled device unit;

[0055] The industrial control computer unit is used to collect test data information of the target computer; the controlled device unit is used to collect operation data information of the controlled device.

[0056] According to the above technical solution, the information processing module includes an information storage unit, an information analysis unit and an information transmission unit;

[0057] The information storage unit is used to store the information acquired by the information acquisition module; the information analysis unit is used to analyze the information acquired by the information acquisition module; and the information transmission unit is used for information transmission of each module in the system.

[0058] According to the above technical solution, the execution module includes a display unit and a control unit;

[0059] The display unit is used to form visual data from the state information of the target computer and the operating data of the controlled device; the control unit is used to control the sending of fault warning information.

[0060] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0061] The present invention monitors and analyzes the data between the industrial control computer and the industrial equipment it controls, obtains the impact items of the controlled equipment on the target computer, and the correlation characteristics between the target computer and the controlled equipment, and performs fault warning on the target computer through a neural network model, thereby realizing the influence judgment of the controlled equipment on the industrial control computer. Not only can the status of the industrial control computer be tested by the test system itself, but the status of the industrial control computer can also be indirectly reflected through the operating data of the industrial equipment controlled by the industrial control computer, so as to fully understand the health status of the industrial control computer, improve the early warning of industrial control computer failures, reduce the risk of failures, and improve production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0063] Figure 1 It is a flow chart of an industrial control computer status testing method based on artificial intelligence of the present invention;

[0064] Figure 2 The present invention is a structural schematic diagram of an industrial control computer status testing system based on artificial intelligence. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] See also Figure 1 , the present invention provides a technical solution:

[0067] An industrial control computer status testing method based on artificial intelligence, the method comprising the following steps:

[0068] S1. Obtain historical test records of the target computer and historical operation records of the controlled device; match the historical test records with the historical operation records to obtain associated records;

[0069] S2. Mark the test records in the historical test records, and divide the associated records into normal associated records and abnormal associated records according to the test records included in the associated records;

[0070] S3, extracting the collection item data in the operation record, calculating the collection item data corresponding to the associated record, and obtaining normal characteristic values ​​and abnormal characteristic values; screening the collection items of the controlled device according to the normal characteristic values ​​and abnormal characteristic values, and obtaining the influence items of the controlled device on the target computer;

[0071] S4. According to the abnormal association record, the affected items are marked, and the items affecting the target computer are matched with the controlled device to obtain the association features between the target computer and the controlled device; according to the association features, a comprehensive training set is obtained from the association record;

[0072] S5. Construct a neural network model, train the neural network through a comprehensive training set, and issue fault warnings for the target computer.

[0073] According to the above technical solution, step S1 includes the following:

[0074] S1-1, each time a status test is performed on the target computer, a test record is generated, and all test records are extracted into the database to form historical test records;

[0075] The historical test records include test time, test items, test item values ​​and test results; the test items include CPU, memory, temperature, power supply and network; wherein, in each historical test record, the test items executed on the target computer are the same, each test item corresponds to a test result, and the test results include normal data and abnormal data;

[0076] S1-2. Each time a pair of controlled devices is subjected to data collection, an operation record is generated, and all operation records are extracted into the database to form a historical operation record;

[0077] The historical operation record includes the collection time, collection items and collection item values; the collection items include current, voltage, vibration frequency, operation frequency, rotation speed, load rate, temperature and humidity and data transmission rate; wherein, in each historical test record, the collection items executed for the same controlled device are the same;

[0078] S1-3, performing time matching according to the test time of the historical test record and the collection time of the historical operation record, taking the test time and the collection time being the same time point as a matching condition, matching the test record and the operation record, and marking the successfully matched test record and operation record as associated records;

[0079] The associated record includes a test record and several operation records, and each operation record corresponds to a controlled device;

[0080] By recording test data and operation data and matching these data into related records, an effective data basis is provided for obtaining the characteristic relationship between the target computer and the controlled device, which is helpful for subsequent characteristic analysis.

[0081] According to the above technical solution, step S2 includes the following:

[0082] S2-1. According to the historical test records, if in the test items of a certain test record, the test results corresponding to the test items are all normal data, then the test record is marked as a normal test record;

[0083] If, among the test items of a test record, any test result corresponding to any test item is abnormal data, the test record is marked as an abnormal test record;

[0084] S2-2, classifying the associated records according to the normal test records and the abnormal test records;

[0085] If the test record in a certain associated record is a normal test record, the associated record is classified as a normal associated record;

[0086] If a test record in a certain associated record is an abnormal test record, the associated record is classified as an abnormal associated record;

[0087] By classifying normal and abnormal test records, the two types of related records are effectively distinguished, laying the foundation for subsequent feature extraction and analysis.

[0088] According to the above technical solution, step S3 includes the following:

[0089] S3-1. Extract the collected item data in the operation record according to the historical operation record;

[0090] Match each operation record in all normal association records, and match all operation records of the same controlled device in all normal association records based on the operation record of the same controlled device;

[0091] According to all operation records of the same controlled device in all normal association records, the average value of the collection item data corresponding to the same collection item of the same controlled device in all normal association records is calculated as the normal characteristic value of the collection item of the controlled device;

[0092] Match each operation record in all abnormal association records, and match all operation records of the same controlled device in all abnormal association records based on the operation record of the same controlled device;

[0093] According to all operation records of the same controlled device in all abnormal association records, the average value of the collection item data corresponding to the same collection item of the same controlled device in all abnormal association records is calculated as the abnormal characteristic value of the collection item of the controlled device;

[0094] For example:

[0095] There is an existing controlled device A, and the collected items of the controlled device A are current and voltage;

[0096] Match all the operation records of controlled device A in all normal association records. There are 2 operation records of controlled device A:

[0097] The first operation record is current value: 4.0, voltage value: 220;

[0098] The second operation record is current value: 5.0, voltage value: 218;

[0099] Calculate the average value of the collection item data corresponding to the same collection item of the same controlled device in all normal association records, that is, calculate the average value of the current value corresponding to the current of the controlled device A as (4.0+5.0) / 2=4.5;

[0100] The average value of the voltage corresponding to the voltage of the controlled device A is calculated to be (220+218) / 2=219;

[0101] 4.5 is taken as the normal characteristic value of the current of the controlled device A; 219 is taken as the normal characteristic value of the voltage of the controlled device A;

[0102] S3-2, calculating the ratio of the difference between the normal feature value and the abnormal feature value of the same collection item of the same controlled device to the normal feature value of the collection item of the controlled device; setting a difference ratio threshold, if the ratio of the difference between the normal feature value and the abnormal feature value of the same collection item of the same controlled device to the normal feature value of the collection item of the controlled device is greater than or equal to the difference ratio threshold, then the collection item of the controlled device is used as an influence item of the controlled device on the target computer;

[0103] For the collection items in each operation record included in all the abnormal association records, combined with the controlled devices corresponding to each operation record, the impact items in each operation record included in all the abnormal association records are marked;

[0104] Normal eigenvalues ​​and abnormal eigenvalues ​​are extracted by calculating the collection item data in the associated records, and the influencing items are determined by setting the difference ratio threshold. By extracting normal eigenvalues ​​and abnormal eigenvalues, the influence of noise can be eliminated, and the universality of the changes in the collection item data can be better seen through normal eigenvalues ​​and abnormal eigenvalues, so that the influencing items can be obtained more accurately.

[0105] According to the above technical solution, step S4 includes the following:

[0106] S4-1. Mark the test item whose test result in the test record included in a certain abnormality association record is a test item with data abnormality as an affected item;

[0107] Matching the affected item in the abnormal association record with the impact item corresponding to the running record contained in the abnormal association record, wherein the matching process is to use the impact item corresponding to the running record contained in the abnormal association record as the associated impact item of the affected item in the abnormal association record;

[0108] Count all associated impact items of different affected items in all abnormal associated records;

[0109] The same affected item and all the associated affected items corresponding thereto are used as a set of associated features between the target computer and the controlled device;

[0110] S4-2, according to a certain set of association features between the target computer and the controlled device, extract the collection item data corresponding to the set of association features from the normal association records to form a normal training set corresponding to the set of association features; extract the collection item data corresponding to the set of association features from the abnormal association records to form an abnormal training set corresponding to the set of association features;

[0111] After marking the data collected in the abnormal training set corresponding to the group of associated features, the data are combined with the normal training set corresponding to the group of associated features to form a comprehensive training set for the group of associated features;

[0112] By analyzing the associated features to construct a comprehensive training set, a reliable data source for training the neural network model is provided.

[0113] According to the above technical solution, step S5 includes the following:

[0114] Construct a neural network model and train the neural network model using a comprehensive training set of a set of related features;

[0115] The trained neural network model outputs the state information of the test items corresponding to the group of associated features according to the collection items corresponding to the group of associated features input; the state information includes normal state and abnormal state;

[0116] If the neural network model outputs that the state of the test item corresponding to the group of associated features is abnormal, fault warning information of the target computer is generated;

[0117] In the model training stage, a normal training set is formed by extracting the collection item data of the group of associated features from the normal associated records, and an abnormal training set is formed by extracting the collection item data of the group of associated features from the abnormal associated records. The data of the abnormal training set is marked and combined with the normal training set to form a comprehensive training set. The neural network model is trained using the comprehensive training set so that the model can identify the normal state and abnormal state of the corresponding test items.

[0118] Analyzing the impact between industrial computers and the industrial equipment they control can not only test the status of industrial computers through the test system itself, but also reflect the status of industrial computers through the operating data of industrial equipment controlled by industrial computers, so as to fully understand the health status of industrial computers, improve the early warning of industrial computer failures, reduce the risk of failures, and improve production efficiency;

[0119] Training the neural network based on the comprehensive training set and performing prediction and early warning on the target industrial control computer, combined with historical data and real-time monitoring, can effectively improve the accuracy and timeliness of the model.

[0120] See also Figure 2 , an industrial control computer status test system based on artificial intelligence, the system includes an information acquisition module, an information processing module and an execution module;

[0121] The information acquisition module is used to acquire the information required by the system; the information processing module is used to store, analyze and transmit the information of each module; and the execution module is used to execute the information of the information processing module.

[0122] According to the above technical solution, the information acquisition module includes an industrial control computer unit and a controlled device unit;

[0123] The industrial control computer unit is used to collect test data information of the target computer; the controlled device unit is used to collect operation data information of the controlled device.

[0124] According to the above technical solution, the information processing module includes an information storage unit, an information analysis unit and an information transmission unit;

[0125] The information storage unit is used to store the information acquired by the information acquisition module; the information analysis unit is used to analyze the information acquired by the information acquisition module; and the information transmission unit is used for information transmission of each module in the system.

[0126] According to the above technical solution, the execution module includes a display unit and a control unit;

[0127] The display unit is used to form visual data from the state information of the target computer and the operating data of the controlled device; the control unit is used to control the sending of fault warning information.

[0128] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0129] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An industrial computer status testing method based on artificial intelligence, characterized in that: The method comprises the following steps: S1. Obtain historical test records of the target computer and historical operation records of the controlled device; match the historical test records with the historical operation records to obtain associated records; S2. Mark the test records in the historical test records, and divide the associated records into normal associated records and abnormal associated records according to the test records included in the associated records; S3, extracting the collection item data in the operation record, calculating the collection item data corresponding to the associated record, and obtaining normal characteristic values ​​and abnormal characteristic values; screening the collection items of the controlled device according to the normal characteristic values ​​and abnormal characteristic values, and obtaining the influence items of the controlled device on the target computer; S4. According to the abnormal association record, the affected items are marked, and the items affecting the target computer are matched with the controlled device to obtain the association features between the target computer and the controlled device; according to the association features, a comprehensive training set is obtained from the association record; S5, construct a neural network model, train the neural network through a comprehensive training set, and provide fault warning for the target computer; The step S1 comprises: S1-1, each time a status test is performed on the target computer, a test record is generated, and all test records are extracted into the database to form historical test records; The historical test records include test time, test items, test item values ​​and test results; the test items include CPU, memory, temperature, power supply and network; wherein, in each historical test record, the test items executed on the target computer are the same, each test item corresponds to a test result, and the test results include normal data and abnormal data; S1-2. Each time a pair of controlled devices is subjected to data collection, an operation record is generated, and all operation records are extracted into the database to form a historical operation record; The historical operation record includes the collection time, collection items and collection item values; the collection items include current, voltage, vibration frequency, action frequency, rotation speed, load rate, temperature and humidity and data transmission rate; wherein, in each historical operation record, the collection items executed for the same controlled device are the same; S1-3, performing time matching according to the test time of the historical test record and the collection time of the historical operation record, taking the test time and the collection time being the same time point as a matching condition, matching the test record and the operation record, and marking the successfully matched test record and operation record as associated records; The association record includes a test record and several operation records, and each operation record corresponds to a controlled device.

2. The method for testing the status of an industrial computer based on artificial intelligence according to claim 1, characterized in that: The step S2 comprises: S2-1. According to the historical test records, if in the test items of a certain test record, the test results corresponding to the test items are all normal data, then the test record is marked as a normal test record; If, among the test items of a test record, any test result corresponding to any test item is abnormal data, the test record is marked as an abnormal test record; S2-2, classifying the associated records according to the normal test records and the abnormal test records; If the test record in a certain associated record is a normal test record, the associated record is classified as a normal associated record; If a test record in a certain associated record is an abnormal test record, the associated record is classified as an abnormal associated record.

3. The method for testing the status of an industrial control computer based on artificial intelligence according to claim 2, characterized in that: The step S3 comprises: S3-1. Extract the collected item data in the operation record according to the historical operation record; Match each operation record in all normal association records, and match all operation records of the same controlled device in all normal association records based on the operation record of the same controlled device; According to all operation records of the same controlled device in all normal association records, the average value of the collection item data corresponding to the same collection item of the same controlled device in all normal association records is calculated as the normal characteristic value of the collection item of the controlled device; Match each operation record in all abnormal association records, and match all operation records of the same controlled device in all abnormal association records based on the operation record of the same controlled device; According to all operation records of the same controlled device in all abnormal association records, the average value of the collection item data corresponding to the same collection item of the same controlled device in all abnormal association records is calculated as the abnormal characteristic value of the collection item of the controlled device; S3-2, calculating the ratio of the difference between the normal feature value and the abnormal feature value of the same collection item of the same controlled device to the normal feature value of the collection item of the controlled device; setting a difference ratio threshold, if the ratio of the difference between the normal feature value and the abnormal feature value of the same collection item of the same controlled device to the normal feature value of the collection item of the controlled device is greater than or equal to the difference ratio threshold, then the collection item of the controlled device is used as an influence item of the controlled device on the target computer; For the collection items in each operation record included in all the abnormal association records, combined with the controlled devices corresponding to each operation record, the impact items in each operation record included in all the abnormal association records are marked.

4. The method for testing the status of an industrial control computer based on artificial intelligence according to claim 3 is characterized in that: The step S4 comprises: S4-1. Mark the test item whose test result in the test record included in a certain abnormality association record is a test item with data abnormality as an affected item; Matching the affected item in the abnormal association record with the impact item corresponding to the running record contained in the abnormal association record, wherein the matching process is to use the impact item corresponding to the running record contained in the abnormal association record as the associated impact item of the affected item in the abnormal association record; Count all associated impact items of different affected items in all abnormal associated records; The same affected item and all the associated affected items corresponding thereto are used as a set of associated features between the target computer and the controlled device; S4-2, according to a certain set of association features between the target computer and the controlled device, extract the collection item data corresponding to the set of association features from the normal association records to form a normal training set corresponding to the set of association features; extract the collection item data corresponding to the set of association features from the abnormal association records to form an abnormal training set corresponding to the set of association features; After the collected item data in the abnormal training set corresponding to the group of associated features are marked, they are combined with the normal training set corresponding to the group of associated features to form a comprehensive training set of the group of associated features.

5. The method for testing the status of an industrial control computer based on artificial intelligence according to claim 4, characterized in that: The step S5 comprises: Construct a neural network model and train the neural network model using a comprehensive training set of a set of related features; The trained neural network model outputs the state information of the test items corresponding to the group of associated features according to the collection items corresponding to the group of associated features input; the state information includes normal state and abnormal state; If the neural network model outputs that the state of the test item corresponding to the group of associated features is abnormal, fault warning information of the target computer is generated.

6. An industrial control computer status test system based on artificial intelligence, using an industrial control computer status test method based on artificial intelligence as described in any one of claims 1 to 5, characterized in that: The system includes an information collection module, an information processing module and an execution module; The information acquisition module is used to acquire the information required by the system; the information processing module is used to store, analyze and transmit the information of each module; and the execution module is used to execute the information of the information processing module.

7. The artificial intelligence-based industrial computer status test system according to claim 6 is characterized in that: The information acquisition module includes an industrial control computer unit and a controlled equipment unit; The industrial control computer unit is used to collect test data information of the target computer; the controlled device unit is used to collect operation data information of the controlled device.

8. The artificial intelligence-based industrial computer status test system according to claim 6, characterized in that: The information processing module includes an information storage unit, an information analysis unit and an information transmission unit; The information storage unit is used to store the information acquired by the information acquisition module; the information analysis unit is used to analyze the information acquired by the information acquisition module; and the information transmission unit is used for information transmission of each module in the system.

9. The artificial intelligence-based industrial computer status test system according to claim 6, characterized in that: The execution module includes a display unit and a control unit; The display unit is used to form visual data from the state information of the target computer and the operating data of the controlled device; the control unit is used to control the sending of fault warning information.

Images