Fault diagnosis system based on AI data driving

Abstract

The embodiment of the invention discloses a fault diagnosis system based on AI data driving, and belongs to the technical field of equipment fault diagnosis, and the system comprises a sensing module which is installed on target equipment, collects the operation signal data of the target equipment, transmits the signal data to a monitoring station through a signal cable, and completes the local data processing through the monitoring station. The data is transmitted to a server through a network cable and an optical fiber; the analysis module displays, stores, distributes and models signal data according to the obtained equipment signal data, and field equipment maintenance and management personnel can remotely access the intelligent monitoring and diagnosis system to obtain the real-time operation state of the equipment; the application module carries out remote data analysis and fault diagnosis on all monitoring equipment and remotely monitors equipment operation signal data, the data can be connected with a large screen, and the current state and statistical information of the equipment are displayed in real time. According to the embodiment of the invention, the consumption of spare parts and the maintenance workload can be reduced, the maintenance cost is minimized, and the production efficiency is improved.

Description

Technical Field

[0001] The present application belongs to the technical field of equipment fault diagnosis, and specifically, relates to a fault diagnosis system driven by AI data. Background Art

[0002] Predictive monitoring of equipment status is a technology based on rotating machinery status assessment, which can accurately determine the operating status and remaining life of the equipment. The new predictive monitoring system for auxiliary equipment status, based on the establishment of a fault model, determines the vibration alarm value, writes the fault model and judgment rules into the monitoring system, corresponds various faults to characteristic frequencies and alarm values ​​one by one, establishes a fault diagnosis library, and automatically outputs the diagnosis results, achieving the purpose of reducing maintenance costs and improving production efficiency.

[0003] At present, the vibration protection system in the industry mainly provides passive shutdown protection for the unit, and can only provide numerical parameters. It is impossible to perform spectrum analysis in this way, which makes most of the monitoring functions of the monitoring point abandoned. At the same time, in the existing online protection system applications of key units, most of them only play the role of monitoring alarm values, and the alarm value setting is a mean alarm. In the face of complex working conditions, there are often no alarms or false alarms, making the system "useless"; in fault diagnosis, since most equipment suppliers do not have professional vibration theory knowledge and cannot establish a fault diagnosis model, they do not have an expert self-diagnosis function that can be used, and only rely on professionals for analysis and diagnosis. When professionals analyze, due to the complex design and operation of the software interface and incomplete analysis functions, they are often unable to diagnose the exact cause of the fault. Summary of the invention

[0004] In order to solve the above problems and technical defects, the embodiment of the present application adopts the following technical solution, a fault diagnosis system based on AI data drive, including:

[0005] The sensing module is installed on the target device to collect the operating signal data of the target device and transmit the signal data to the monitoring station through the signal cable. The monitoring station completes local data processing and then transmits the data to the server through the network cable and optical fiber.

[0006] The analysis module is used to display, store, distribute and model the acquired equipment signal data. On-site equipment maintenance and management personnel can remotely access the intelligent monitoring and diagnosis system to obtain the real-time operating status of the equipment.

[0007] The application module is used for remote data analysis and fault diagnosis of all monitoring equipment, remote monitoring equipment operation signal data, and the data can be connected to the large screen to display the current status and statistical information of the equipment in real time.

[0008] Preferably, the perception module is provided with an online monitoring module, the analysis module is provided with a fault diagnosis module, and the application module is provided with a login registration module, a system setting module and a data query module;

[0009] The online monitoring module is used to collect monitoring point signals, and to perform operation analysis based on the time domain, frequency domain, trend, phase, time 3D, speed 3D, axis trajectory, Bode diagram and Nyquist diagram of the monitoring point signals to obtain the operating status of the target equipment. At the same time, corresponding alarm trigger thresholds are set according to different speeds and powers, and different alarm status color labels are set according to different alarm trigger thresholds;

[0010] Fault diagnosis module, used to establish fault diagnosis rule base, symptom fact base and equipment database, and perform fault reasoning diagnosis on the monitored and collected equipment signals;

[0011] The login and registration module is used to register and log in the user's account, allowing the user to establish a connection with the server through the client;

[0012] The system settings module is used to complete user management, organization, department, role and responsibility definition, and define and manage user viewing permissions, data permissions, and operation permissions;

[0013] The data query module is used to display data, query equipment status data and abnormal data, provide data overview, and support data export.

[0014] Furthermore, the online monitoring module is to install a monitoring sensor on the target device, and use the monitoring sensor as a monitoring point. The monitoring point signal is uploaded to the main board through a signal line. After the main board compresses and filters the collected signal data, the signal data is uploaded to the server through optical fiber. The host computer system establishes a three-dimensional model of the device and the monitoring point, binds the uploaded signal data one by one, sets the alarm to observe the operating data of the equipment, and realizes real-time monitoring of the setting operating status.

[0015] Furthermore, the fault diagnosis module collects device signal data, and the back end compares it according to preset rules, performs core algorithm calculations, obtains diagnostic results and possible values, and then summarizes large amounts of data and uses machine learning to make the diagnostic results more accurate.

[0016] Furthermore, the fault diagnosis rule base first obtains characteristic parameters of the target device type and then is established based on the characteristic parameters;

[0017] The symptom fact base is established based on historical failure data;

[0018] The equipment database is established based on the equipment structure, bearing type, model and speed, and the diagnostic logic mode is preset in the equipment database;

[0019] Fault reasoning diagnosis of equipment signals collected by monitoring is to first determine the diagnostic logic mode according to the hardware information of the target equipment, and then combine the fault diagnosis rule base and symptom fact base and perform fault reasoning on the fault signals obtained by monitoring based on the fuzzy fault diagnosis principle.

[0020] Furthermore, the specific process of the fault reasoning diagnosis is as follows:

[0021] Input feature parameters, match cases according to the feature parameters, and output the fault cause when the case matches completely within the range;

[0022] If there is no matching fault case in the database, rule matching will be started until the best result is obtained;

[0023] If there is no matching case or rule, the fault event will be memorized and stored in the database, and a prompt dialog box will be displayed to select manual diagnosis or AI intelligent diagnosis.

[0024] Furthermore, the process of the AI ​​intelligent diagnosis is as follows:

[0025] Edit the basic component information of the target device;

[0026] Select the shaft system component in the basic components and set the basic parameters of the shaft system;

[0027] Edit the corresponding shaft system diagnostic model according to the set shaft system basic parameters;

[0028] Select the shaft system diagnosis model corresponding to the monitoring point that needs to be diagnosed;

[0029] The shaft system diagnosis model performs AI intelligent diagnosis on the corresponding monitoring signals to obtain the corresponding fault reasoning conclusions.

[0030] Furthermore, the login registration module performs user login by calling the interface for obtaining the token, transmitting the summary of the user name and password to the server, and the server compares the user information in the database. If a match is found, the token bound to the user is returned. When the server-side interface is called, the IP address and user-agent information of the initiating request are compared to realize single sign-on in multiple services, and one token can access each service.

[0031] Furthermore, the system setting module adopts a role-based permission access control strategy, creates various roles according to work requirements, associates permissions with roles, and users are assigned to corresponding roles based on responsibilities and qualifications, so that users become members of the target role and obtain corresponding role permissions. Roles can be granted new permissions based on new requirements and system mergers. The role-based permission access control strategy supports the principle of least privilege, the principle of separation of duties, and the principle of data abstraction.

[0032] Furthermore, the data query module transmits query conditions by sending requests, queries the qualified data from the database, and then assembles the data through association and responds to the front end to complete the display, search and viewing of the data. It also supports a variety of fuzzy search matching, time end matching, keyword matching search, and optimizes the database storage structure by adding database query indexes.

[0033] Compared with the prior art, the beneficial effects of the embodiments of the present application are:

[0034] This application accumulates industrial data on equipment operation, uses big data to analyze equipment degradation, establishes a fault model, determines the vibration alarm value, writes the fault model and judgment rules into the monitoring system, and writes diagnostic information such as the vibration frequency and alarm value of each component into the online monitoring and diagnosis system, so that the system can realize self-diagnosis and have an early warning function, thereby establishing reliable predictive maintenance, maximizing the service life of the equipment, reducing the number of overhauls, extending the overhaul cycle, and controlling excessive maintenance, thereby reducing spare parts consumption and maintenance workload, and ultimately achieving the goal of minimizing maintenance costs and improving production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In the attached picture:

[0036] Figure 1 Schematic diagram of the system structure of an embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions 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. Generally, the components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0038] Example

[0039] like Figure 1 As shown, a fault diagnosis system based on AI data drive includes:

[0040] The sensing module is installed on the target device to collect the operating signal data of the target device and transmit the signal data to the monitoring station through the signal cable. The monitoring station completes local data processing and then transmits the data to the server through the network cable and optical fiber.

[0041] The analysis module is used to display, store, distribute and model the acquired equipment signal data. On-site equipment maintenance and management personnel can remotely access the intelligent monitoring and diagnosis system to obtain the real-time operating status of the equipment.

[0042] The application module is used for remote data analysis and fault diagnosis of all monitoring equipment, remote monitoring equipment operation signal data, and the data can be connected to the large screen to display the current status and statistical information of the equipment in real time.

[0043] The perception module is equipped with an online monitoring module, the analysis module is equipped with a fault diagnosis module, and the application module is equipped with a login and registration module, a system setting module, and a data query module;

[0044] The login and registration module is used to register and log in the user's account, allowing the user to establish a connection with the server through the client;

[0045] The login registration module performs user login by calling the interface for obtaining tokens, and transmits the summary of the username and password to the server. The server compares the user information in the database, and returns the token bound to the user if a match is found. When the server-side interface is called, the IP address and user-agent information of the initiating request are compared to implement single sign-on in multiple services, and one token can be used to access each service.

[0046] Even if the token is intercepted, it will not cause too much security risk because there is no username and password. Then, if a hacker uses this token to forge a user request, when the server-side interface is called, the IP address, user-agent and other information that initiated the request can be compared to prevent forgery. If the validity period of the token is set to a short time, it will expire after a while, which solves the risk of the user's login username and password being hijacked. In addition, it can simply implement single sign-on in multiple services, and one token can access each service.

[0047] The system settings module is used to complete user management, organization, department, role and responsibility definition, and define and manage user viewing permissions, data permissions, and operation permissions;

[0048] The system setting module adopts a role-based permission access control strategy. Various roles are created according to work requirements, and permissions are associated with roles. Users are assigned to corresponding roles based on responsibilities and qualifications, making users members of the target roles and obtaining corresponding role permissions. Roles can be granted new permissions based on new requirements and system mergers. The role-based permission access control strategy supports the principle of least privilege, the principle of separation of duties, and the principle of data abstraction.

[0049] The online monitoring module is used to collect monitoring point signals, and to perform operation analysis based on the time domain, frequency domain, trend, phase, time 3D, speed 3D, axis trajectory, Bode diagram and Nyquist diagram of the monitoring point signals to obtain the operating status of the target equipment. At the same time, corresponding alarm trigger thresholds are set according to different speeds and powers, and different alarm status color labels are set according to different alarm trigger thresholds;

[0050] The online monitoring module installs monitoring sensors on the target device and uses the monitoring sensors as monitoring points. The monitoring point signals are uploaded to the mainboard through signal lines. After the mainboard compresses and filters the collected signal data, the signal data is uploaded to the server through optical fiber. The host computer system establishes a three-dimensional model of the device and monitoring points, binds the uploaded signal data one by one, sets alarms to observe the operating data of the equipment, and realizes real-time monitoring of the operating status.

[0051] Data query module, used for data display, querying equipment status data and abnormal data, providing data overview, and supporting data export;

[0052] The data query module passes the query conditions by sending requests, queries the database for data that meets the conditions, then assembles the data through associations and responds to the front end to complete the display, search and viewing of the data. It also supports a variety of fuzzy search matching, time end matching, keyword matching search, and optimizes the database storage structure by adding database query indexes.

[0053] Fault diagnosis module, used to establish fault diagnosis rule base, symptom fact base and equipment database, and perform fault reasoning diagnosis on the monitored and collected equipment signals;

[0054] The fault diagnosis module collects device signal data, and the back-end compares it according to preset rules and performs core algorithm calculations to obtain diagnostic results and possible values. It then aggregates large amounts of data and uses machine learning to make the diagnostic results more accurate.

[0055] The fault diagnosis rule base first obtains the characteristic parameters of the target device type and then is established based on the characteristic parameters;

[0056] The symptom fact base is established based on historical failure data;

[0057] The equipment database is established based on the equipment structure, bearing type, model and speed, and the diagnostic logic mode is preset in the equipment database;

[0058] Fault reasoning diagnosis of equipment signals collected by monitoring is to first determine the diagnostic logic mode according to the hardware information of the target equipment, and then combine the fault diagnosis rule base and symptom fact base and perform fault reasoning on the fault signals obtained by monitoring based on the fuzzy fault diagnosis principle.

[0059] The specific process of fault reasoning diagnosis is as follows:

[0060] Input feature parameters, match cases according to the feature parameters, and output the fault cause when the case matches completely within the range;

[0061] If there is no matching fault case in the database, rule matching will be started until the best result is obtained;

[0062] If there is no matching case or rule, the fault event will be memorized and stored in the database, and a prompt dialog box will be displayed to select manual diagnosis or AI intelligent diagnosis.

[0063] The process of AI intelligent diagnosis is as follows:

[0064] Edit the basic component information of the target device;

[0065] Select the shaft system component in the basic components and set the basic parameters of the shaft system;

[0066] Edit the corresponding shaft system diagnostic model according to the set shaft system basic parameters;

[0067] Select the shaft system diagnosis model corresponding to the monitoring point that needs to be diagnosed;

[0068] The shaft system diagnosis model performs AI intelligent diagnosis on the corresponding monitoring signals to obtain the corresponding fault reasoning conclusions.

[0069] The above-mentioned embodiments only express the preferred implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for ordinary technicians in this field, several modifications, improvements and substitutions can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. A fault diagnosis system based on AI data drive, characterized in that: include: The sensing module is installed on the target device to collect the operating signal data of the target device and transmit the signal data to the monitoring station through the signal cable. The monitoring station completes local data processing and then transmits the data to the server through the network cable and optical fiber. The analysis module is used to display, store, distribute and model the acquired equipment signal data. On-site equipment maintenance and management personnel can remotely access the intelligent monitoring and diagnosis system to obtain the real-time operating status of the equipment. The application module is used for remote data analysis and fault diagnosis of all monitoring equipment, remote monitoring equipment operation signal data, and the data can be connected to the large screen to display the current status and statistical information of the equipment in real time.

2. The AI ​​data-driven fault diagnosis system according to claim 1, characterized in that: The perception module is provided with an online monitoring module, the analysis module is provided with a fault diagnosis module, and the application module is provided with a login registration module, a system setting module and a data query module; The online monitoring module is used to collect monitoring point signals, and to perform operation analysis based on the time domain, frequency domain, trend, phase, time 3D, speed 3D, axis trajectory, Bode diagram and Nyquist diagram of the monitoring point signals to obtain the operating status of the target equipment. At the same time, corresponding alarm trigger thresholds are set according to different speeds and powers, and different alarm status color labels are set according to different alarm trigger thresholds; Fault diagnosis module, used to establish fault diagnosis rule base, symptom fact base and equipment database, and perform fault reasoning diagnosis on the monitored and collected equipment signals; The login and registration module is used to register and log in the user's account, so that the user can establish a connection with the server through the client; The system settings module is used to complete user management, organization, department, role and responsibility definition, and define and manage user viewing permissions, data permissions, and operation permissions; The data query module is used to display data, query equipment status data and abnormal data, provide data overview, and support data export.

3. The AI ​​data-driven fault diagnosis system according to claim 2, characterized in that: The online monitoring module is to install a monitoring sensor on the target device and use the monitoring sensor as a monitoring point. The monitoring point signal is uploaded to the main board through a signal line. After the main board compresses and filters the collected signal data, the signal data is uploaded to the server through optical fiber. The host computer system establishes a three-dimensional model of the device and the monitoring point, binds the uploaded signal data one by one, sets the alarm to observe the operating data of the device, and realizes real-time monitoring of the setting operating status.

4. The AI ​​data-driven fault diagnosis system according to claim 2, characterized in that: The fault diagnosis module collects device signal data, and the back end compares it according to preset rules, performs core algorithm calculations, obtains diagnostic results and possible values, and then summarizes large amounts of data and uses machine learning to make the diagnostic results more accurate.

5. The AI ​​data-driven fault diagnosis system according to claim 4 is characterized in that: The fault diagnosis rule base first obtains characteristic parameters of the target device type and then is established based on the characteristic parameters; The symptom fact base is established based on historical failure data; The equipment database is established based on the equipment structure, bearing type, model and speed, and the diagnostic logic mode is preset in the equipment database; Fault reasoning diagnosis of equipment signals collected by monitoring is to first determine the diagnostic logic mode according to the hardware information of the target equipment, and then combine the fault diagnosis rule base and symptom fact base and perform fault reasoning on the fault signals obtained by monitoring based on the fuzzy fault diagnosis principle.

6. The AI ​​data-driven fault diagnosis system according to claim 5, characterized in that: The specific process of the fault reasoning diagnosis is as follows: Input feature parameters, match cases according to the feature parameters, and output the fault cause when the case matches completely within the range; If there is no matching fault case in the database, rule matching will be started until the best result is obtained; If there is no matching case or rule, the fault event will be memorized and stored in the database, and a prompt dialog box will be displayed to select manual diagnosis or AI intelligent diagnosis.

7. The AI ​​data-driven fault diagnosis system according to claim 6, characterized in that: The process of AI intelligent diagnosis is as follows: Edit the basic component information of the target device; Select the shaft system component in the basic components and set the basic parameters of the shaft system; Edit the corresponding shaft system diagnostic model according to the set shaft system basic parameters; Select the shaft system diagnosis model corresponding to the monitoring point that needs to be diagnosed; The shaft system diagnosis model performs AI intelligent diagnosis on the corresponding monitoring signals to obtain the corresponding fault reasoning conclusions.

8. The AI ​​data-driven fault diagnosis system according to claim 2, characterized in that: The login registration module performs user login by calling the interface for obtaining token, transmitting the summary of the user name and password to the server, and the server compares the user information in the database. If a match is found, the token bound to the user is returned. When the server-side interface is called, the IP address and user-agent information of the initiating request are compared to realize single sign-on in multiple services, and one token can access each service.

9. The AI ​​data-driven fault diagnosis system according to claim 2, characterized in that: The system setting module adopts a role-based permission access control strategy, creates various roles according to work requirements, associates permissions with roles, and users are assigned to corresponding roles based on responsibilities and qualifications, so that users become members of the target role and obtain corresponding role permissions. Roles can be granted new permissions based on new requirements and system mergers. The role-based permission access control strategy supports the principle of least privilege, the principle of separation of duties, and the principle of data abstraction.

10. The AI ​​data-driven fault diagnosis system according to claim 2, characterized in that: The data query module transmits query conditions by sending requests, queries the database for data that meets the conditions, and then assembles the data through association and responds to the front end to complete the display, search and viewing of the data. It also supports a variety of fuzzy search matching, matching by time end, keyword matching search, and optimizes the database storage structure by adding database query indexes.

Images