Equipment fault diagnosis method and system based on artificial intelligence

Abstract

The invention provides an equipment fault diagnosis method and system based on artificial intelligence, and the method comprises the steps: evaluating the health index of equipment based on vibration intensity according to obtained fault related data; respectively inputting fault related data into the trained classification model and collective model to respectively obtain fault diagnosis results of the equipment; inputting the time sequence characteristics of the historical data into the trained time sequence model to obtain an equipment fault diagnosis result; fusing the fault diagnosis result obtained by the collective model, the fault diagnosis result obtained by the classification model and the fault diagnosis result obtained by the time sequence model to obtain a final fault diagnosis result; and meanwhile, intelligent diagnosis and management of faults are realized by constructing a multi-model fused artificial intelligence algorithm model, so that the equipment fault shutdown duration and the equipment fault shutdown rate are reduced, and the maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of fault diagnosis, and in particular relates to an equipment fault diagnosis method and system based on artificial intelligence. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Equipment plays a central role in production lines, and its stable operation is directly related to the operation of the entire line. Equipment failure can significantly disrupt production line operations, resulting in reduced production efficiency and potentially compromising process quality. Currently, despite significant improvements in production automation, its specific application in the field of equipment fault diagnosis and management is relatively limited: 1. Risk of critical equipment downtime: Tobacco production lines are highly dependent on the stable operation of various equipment types, with transmission equipment accounting for over 80% and playing a crucial role. Common failures of this type of equipment include bearing failure, shaft failure, fan failure, and lubrication system failure. Failure of these devices can negatively impact tobacco processing quality and cause delays in tobacco production and tobacco-making production schedules.

[0004] 2. Traditional fault detection methods are lagging: Traditional fault diagnosis methods, such as regular inspections and manual auscultation, often struggle to detect hidden anomalies and potential fault signals within equipment. These methods are not only inefficient but also fail to issue timely fault warnings. This prevents maintenance personnel from taking effective preventive or intervention measures before a fault actually breaks out and materially impacts production. This lag not only increases the risk of equipment damage but also significantly limits the stability and reliability of the production line.

[0005] 3. Lack of Intelligent Fault Diagnosis: In the current production environment, there is a lack of effective intelligent fault diagnosis and management systems to monitor equipment operating status in real time and provide accurate early warnings. Existing technologies mostly remain at the level of simple data collection and analysis, unable to deeply explore the underlying causes and patterns behind equipment failures. This results in maintenance personnel being unable to prevent failures through timely maintenance interventions, and unable to quickly locate the problem and implement effective repair measures after a failure occurs. Summary of the Invention

[0006] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides an equipment fault diagnosis method and system based on artificial intelligence, constructs an artificial intelligence algorithm model with multi-model fusion to realize intelligent diagnosis and management of faults, improves the accuracy of fault diagnosis, thereby reducing the equipment fault downtime and equipment fault downtime rate, and improving maintenance efficiency.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides an artificial intelligence-based equipment fault diagnosis method, comprising: Obtain data related to the current fault of the device, as well as historical data containing data related to the current fault; Based on the acquired fault-related data, the health index of the equipment is evaluated based on the vibration severity; Input the fault-related data into the trained classification model and collective model respectively to obtain the equipment fault diagnosis results; input the time series features of the historical data into the trained time series model to obtain the equipment fault diagnosis results by capturing the dependency relationship in the time series; Fusing the fault diagnosis result obtained by the collective model, the fault diagnosis result obtained by the classification model, and the fault diagnosis result obtained by the time series model to obtain a final fault diagnosis result; Implement equipment fault diagnosis management based on the assessed equipment health index and final fault diagnosis results.

[0008] In a second aspect, the present invention provides an artificial intelligence-based equipment fault diagnosis system, comprising: An acquisition module is configured to: acquire data related to a current fault of the device, and historical data including the data related to the current fault; An evaluation module configured to: evaluate a health index of the equipment based on vibration severity according to the acquired fault-related data; A fault diagnosis module is configured to: input the fault-related data into the trained classification model and collective model respectively to obtain device fault diagnosis results; input the time series features of the historical data into the trained time series model to obtain device fault diagnosis results by capturing the dependency relationship in the time series; a fault diagnosis fusion module configured to fuse the fault diagnosis result obtained by the collective model, the fault diagnosis result obtained by the classification model, and the fault diagnosis result obtained by the time series model to obtain a final fault diagnosis result; The interactive management module is configured to implement fault diagnosis management of the device based on the assessed health index of the device and the final fault diagnosis result.

[0009] In a third aspect, the present invention provides an electronic device comprising a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein the computer instructions, when executed by the processor, perform the method described in the first aspect.

[0010] In a fourth aspect, the present invention provides a computer-readable storage medium for storing computer instructions, wherein when the computer instructions are executed by a processor, the method described in the first aspect is performed.

[0011] One or more of the above technical solutions have the following beneficial effects: The present invention ensures the continuous and stable operation of the equipment by evaluating the health status of the equipment; at the same time, it realizes the intelligent diagnosis and management of faults by constructing an artificial intelligence algorithm model with multi-model fusion. The three models operate in parallel and complement each other. They work together to generate fault diagnosis results through their respective expertise, thereby improving the accuracy of fault diagnosis. According to the fault diagnosis results, the equipment fault downtime and equipment fault downtime rate are reduced, thereby improving maintenance efficiency.

[0012] The present invention establishes a complete quality risk early warning mechanism, which can timely diagnose and warn potential equipment risks, significantly reduce the possibility of quality risks occurring, and safeguard product quality.

[0013] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0015] Figure 1 This is a flow chart of a method for diagnosing equipment faults based on artificial intelligence in an embodiment of the present invention; Figure 2 This is a diagram of equipment failure analysis provided in an embodiment of the present invention; Figure 3 The equipment fault diagnosis algorithm model architecture provided in the embodiment of the present invention; Figure 4 The artificial intelligence-based equipment fault diagnosis and management system architecture provided in the embodiment of the present invention; Figure 5 Schematic diagram of the correspondence between the characteristic signal parameter system and the characteristic space in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0017] It should be noted that the terms used herein are for describing particular embodiments only and are not intended to limit the exemplary embodiments according to the present invention.

[0018] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0019] Example 1 This embodiment discloses an artificial intelligence-based device fault diagnosis method, including: Obtain data related to the current fault of the device, as well as historical data containing data related to the current fault; Based on the acquired fault-related data, the health index of the equipment is evaluated based on the vibration severity; Input the fault-related data into the trained classification model and collective model respectively to obtain the equipment fault diagnosis results; input the time series features of the historical data into the trained time series model to obtain the equipment fault diagnosis results by capturing the dependency relationship in the time series; The fault diagnosis results obtained by the collective model, the fault diagnosis results obtained by the classification model, and the fault diagnosis results obtained by the time series model are integrated to obtain the final fault diagnosis result; Implement equipment fault diagnosis management based on the assessed equipment health index and final fault diagnosis results.

[0020] This embodiment ensures the continuous and stable operation of the equipment by evaluating the health status of the equipment; at the same time, it realizes intelligent diagnosis and management of faults by building an artificial intelligence algorithm model that integrates multiple models, thereby reducing the equipment failure downtime and equipment failure downtime rate and improving maintenance efficiency.

[0021] In this embodiment, equipment failure types and their corresponding characteristic information are analyzed. Equipment failures primarily fall into two categories: mechanical failures and electrical failures. Mechanical failures specifically include bearing failures, gear failures, rotor failures, and lubrication system failures. These failures are often accompanied by phenomena such as excessive equipment temperature, abnormal noise, and increased vibration. Electrical failures involve insulation failures due to aging insulation materials, winding failures due to overheating or short circuits, arcing, power phase loss, and poor power quality. The characteristic information corresponding to each type of equipment failure encompasses signals such as temperature, sound, vibration, magnetic flux, and current.

[0022] Multi-element composite sensors are used to monitor the operation of the equipment at key locations, including the drive end and non-drive end of the motor, the front bearing and rear bearing of the reduction motor unit, the bearing seat of the vibrating conveyor unit, etc., to obtain data information related to equipment failure.

[0023] As an optional implementation method, the slim cigarette production line in the tobacco-making workshop includes a total of 41 devices and a total of 82 multi-component composite sensors are installed.

[0024] As an optional implementation, data transmission utilizes the silk-making workshop's already stable Industrial Ethernet network as the core architecture for the data transmission network. Furthermore, to ensure smooth and unimpeded data transmission, a polling mechanism is employed to transmit device data. This strategy cleverly mitigates the risk of data conflicts, ensuring efficient and stable data transmission.

[0025] As an optional implementation, after completing equipment-related data collection, the collected data is stored. This data storage is achieved through a three-database, multiple-backup strategy. Using PC workstations, sensor data is collected and written in real time. This data is first temporarily stored in the silk-making operation database. Subsequently, the central computer room's operation database efficiently accesses the silk-making operation database at preset intervals, batch-reads data, and performs necessary backup operations. Ultimately, this data is securely stored in the central computer room's historical database.

[0026] In this embodiment, feature extraction from acquired equipment fault-related data is achieved through Fourier transform and current fingerprint recognition technology. Specifically, Fourier transform is used to convert signal data from the original time or spatial domain to the frequency domain, thereby providing a new perspective to deeply understand the intrinsic properties and unique characteristics of the signal. Current fingerprint recognition technology is then used to extract specific characteristic information from the current signal, including three-phase current amplitude, load conditions, and induced current uniformity.

[0027] Among them, the current fingerprint recognition technology is used to extract specific characteristic information from the current signal, including three-phase electric amplitude, load conditions, induced current uniformity and other characteristic information, specifically: The three-phase electric amplitude can be calculated using the following formula: Ua = U max sin(ωt) Ub = U max sin(ωt + 120°) Uc = U max sin(ωt + 240°) Among them, Ua, Ub and Uc represent the voltage amplitudes of the three phases respectively, Umax represents the maximum voltage amplitude, ω represents the angular frequency, and its relationship with the frequency f is as follows: ω = 2πf , t represents time.

[0028] The current load characteristics can be calculated using the following formula:

[0029] in, Represents the frequency characteristics of the load current; It is the fundamental frequency of the power supply system, that is, the standard frequency of the power grid, usually 50Hz or 60Hz; is the harmonic or modulation frequency component in the load current; m is the modulation index or harmonic amplitude coefficient; is the power factor of the load.

[0030] When extracting the uniformity characteristics of the induced current, the calculation can be performed according to the following formula:

[0031]

[0032] in, Represents the unbalance rate or fluctuation rate characteristics of the induced current; is the fundamental frequency of the induced current under ideal or standard conditions; k is a coefficient related to the induced current distribution or load characteristics; s is another key parameter related to the uniformity of the induced current; Indicates the slip frequency.

[0033] This embodiment adopts a dual-wheel drive strategy of deep integration of data and mechanism to construct a health status evaluation model for equipment. The health status evaluation model of equipment not only relies on the calculation of vibration intensity to evaluate the health index, but also comprehensively considers and carefully judges various types of faults and their severity, and accordingly formulates a scientific and reasonable health index deduction mechanism.

[0034] In this embodiment, the health index is calculated by considering the equipment type, size and vibration speed according to the ISO10816 standard, and a clear vibration severity level classification is set for each transmission equipment.

[0035] When calculating vibration severity in the frequency domain, measurements are taken at multiple different points on the machine, such as shafts, bearings, or other parts of the mechanical structure, in two or three measurement directions, and at various measurement positions based on the frequency domain amplitude data of the displacement, velocity, or acceleration signals of the vibration to be measured. The vibration severity is the maximum broadband vibration value measured under specified operating conditions. That is, the maximum value of a set of vibration values ​​obtained from various measurement points and directions is taken as the relevant value of the vibration severity.

[0036] To measure the health status of a device, we use the health index (HD). The definition of the health index HD is as follows:

[0037] in, is the maximum value allowed for vibration intensity. is the threshold value when the vibration intensity is normal, is the actual value of vibration intensity.

[0038] The health index of this embodiment is intuitively presented in a quantitative form of 0 to 100, with each 20 points representing a health level, as shown in Table 1.

[0039] Table 1: Correspondence between health index and maintenance work

[0040] The fault handling mechanism also includes manually determining the fault level and setting up a corresponding point deduction system for different fault levels to quantify their impact, as shown in the table below.

[0041] Table 2 Health Index Deduction Mechanism

[0042] like Figure 3 As shown, this embodiment constructs a multi-model fusion artificial intelligence algorithm model to realize intelligent fault diagnosis, specifically including a classification model, a timing model and a collective model; the classification model, the timing model and the collective model operate in parallel, and at the same time they complement each other, and work together to generate fault diagnosis results through their respective expertise.

[0043] In this embodiment, the classification model has the ability to efficiently classify and deeply analyze equipment failure data. The classification model uses experimental data obtained from a high-simulation test bench and multi-dimensional features such as temperature, sound, vibration, magnetic flux and current collected on site as input information, and outputs the diagnostic results of the equipment failure through processing and analysis.

[0044] As an optional implementation, a random forest classification algorithm in machine learning is used to build a classification model.

[0045] Furthermore, the collective model enhances the comprehensiveness and accuracy of overall predictions or decisions by integrating diverse information from actual field equipment data. This model then performs a weighted fusion with the classification model to jointly output diagnostic results for equipment failures. Furthermore, the algorithmic model integrates the rich knowledge of maintenance experts—high-fidelity test bench experimental data—and closely interacts with users' real-time feedback mechanisms. This model continuously simulates and approximates the analytical thinking patterns of experts while incorporating newly acquired fault data, enabling adaptive improvement and continuous optimization of algorithmic performance.

[0046] In the collective model, the FFT spectrum analysis-based method is used to accurately identify abnormal equipment signals in order to capture the fault characteristics of the equipment in the frequency domain; the sound SEE, or signal enhancement and evaluation method, focuses on detecting bearing anomalies and identifying potential problems through changes in sound signals; in the field of unsupervised learning, the PCA-Hotelling method is used to perform anomaly judgment on vibration data to reveal hidden patterns and abnormal points in the data; in addition, automatic decoder technology is introduced to further enhance the model's anomaly detection capabilities in complex data environments. The collective model obtains abnormal conditions in various aspects of the equipment by diagnosing various aspects of the equipment.

[0047] In this embodiment, based on the key benchmark of timestamp, various characteristic information such as temperature data, vibration data and current data are orderly integrated. On this basis, a characteristic signal parameter system and characteristic space are constructed, such as Figure 5 As shown in FIG, time domain features, frequency domain features, and time-frequency domain features are extracted from the integrated data as input of the time series model, and the fault type is output.

[0048] Specifically, extracting time domain features includes calculating mean square value, peak value, kurtosis and standard deviation; extracting frequency domain features includes spectral energy, spectral center of gravity, spectral variance and spectral kurtosis; extracting time-frequency domain features includes information entropy, wavelet energy spectrum and bearing frequency index.

[0049] The time series model focuses on capturing the temporal dynamic characteristics contained in equipment data, using the time series features in historical data as input to obtain the corresponding fault diagnosis results, and performing Stacking fusion with the output results of the classification model and the collective model. By utilizing the learning ability of the algorithm model itself, it learns multiple prediction results from other models, and uses the prediction results of each sub-model as the input of the fusion model determined by the Stacking fusion algorithm to make the final decision, which jointly acts on the final diagnostic output of the equipment fault.

[0050] The training data of this embodiment includes a vibration diagnosis library of more than 1,200 items, a noise anomaly library of more than 650 items, an ultrasonic model library of more than 230 items, and a magnetic flux anomaly library of more than 80 items, totaling more than 2,100 items.

[0051] This embodiment also includes: implementing interactive fault diagnosis and management for users, which is divided into two interactive levels: device overview and device details, and supports access in the form of large screens, computers, etc.; and implementing closed-loop management of equipment failures based on the equipment health status and the specific causes and solutions obtained from fault diagnosis.

[0052] Users can first get an overview of the overall situation through the device overview, and then click on any specific device to view the device's fault diagnosis results, detailed health status evaluation report and other related information in depth.

[0053] The design of all operation pages of human-computer interaction conforms to the standard WEB page operation style. The rich browser interface technology built with B / S popular technology provides a friendly user interface, humanized design, simple and easy to use, beautiful and easy to operate.

[0054] The equipment fault diagnosis and management system has highly integrated functions. It can not only provide detailed motor fault alarms, but also realize the visualization of the fault inside the motor, and provide accurate diagnostic conclusions and targeted maintenance suggestions. It has built a comprehensive solution from initial alarm to final maintenance, and realized closed-loop management of the entire life cycle of equipment failures.

[0055] This embodiment analyzes equipment fault types and their corresponding characteristic information. The characteristic information corresponding to various equipment faults covers signals such as temperature, sound, vibration, magnetic flux, and current. It obtains data information related to equipment faults and applies multi-element composite sensors to monitor the equipment's operation at key locations, such as motors, reduction motors, and vibrating conveyors. It stores data and extracts features from data resources. Data storage is achieved through a three-database, multiple-backup strategy, and feature extraction is achieved through Fourier transform and current fingerprint recognition technology. It establishes an equipment health status evaluation mechanism and model, using a two-wheel drive strategy that deeply integrates data and mechanisms to construct the equipment health status evaluation model. It constructs a multi-model fusion artificial intelligence algorithm model to achieve intelligent fault diagnosis. Specifically, it includes a classification model, a time series model, and a collective model. These three models operate in parallel and complement each other, contributing to the generation of final fault diagnosis results through their respective expertise. It implements an interactive fault diagnosis and management system for users, divided into two interactive levels: equipment overview and equipment details, and supports access via large screens and computers. It implements closed-loop management of equipment faults based on the equipment health status and the specific causes and solutions derived from fault diagnosis.

[0056] Example 2 The purpose of this embodiment is to provide an artificial intelligence-based equipment fault diagnosis system, including: An acquisition module is configured to: acquire data related to a current fault of the device, and historical data including the data related to the current fault; An evaluation module configured to: evaluate a health index of the equipment based on vibration severity according to the acquired fault-related data; A fault diagnosis module is configured to: input the fault-related data into the trained classification model and collective model respectively to obtain equipment fault diagnosis results; input the time series features of the historical data into the trained time series model to obtain equipment fault diagnosis results; a fault diagnosis fusion module configured to fuse the fault diagnosis result obtained by the collective model, the fault diagnosis result obtained by the classification model, and the fault diagnosis result obtained by the time series model to obtain a final fault diagnosis result; The interactive management module is configured to implement fault diagnosis management of the device based on the assessed health index of the device and the final fault diagnosis result.

[0057] like Figure 4 As shown, the artificial intelligence-based equipment fault diagnosis system provided in this embodiment includes a data acquisition module, a data processing module, an intelligent diagnosis module, a human-computer interaction module and system deployment.

[0058] Specifically, the acquisition module includes data acquisition and data transmission. Data acquisition is performed by wired sensors, and data transmission is performed by wired transmission.

[0059] Specifically, the data processing module includes data storage and feature extraction. The data storage adopts relational database storage. The feature extraction includes frequency domain feature extraction and current feature extraction.

[0060] Specifically, the intelligent diagnosis module includes health status evaluation and fault diagnosis. Health status evaluation includes health level classification, health index calculation and health index deduction mechanism; fault diagnosis includes classification model and time series model.

[0061] Specifically, the human-computer interaction module includes diagnostic result display and fault management; the diagnostic result display can be displayed through digital tables and graphical displays, and fault management adopts closed-loop control.

[0062] Specifically, system deployment includes server configuration, operating system deployment and operating environment construction. Server configuration adopts local server configuration, operating system deployment adopts Linux system deployment, and operating environment construction adopts Docker environment construction.

[0063] In further embodiments, there is also provided: An electronic device includes a memory and a processor, and computer instructions stored in the memory and executed by the processor. When the computer instructions are executed by the processor, the method described in Example 1 is performed. For the sake of brevity, no further details are given here.

[0064] It should be understood that in this embodiment, the processor may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0065] The memory may include a read-only memory and a random access memory, and provides instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0066] A computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the method described in embodiment 1 is performed.

[0067] The method in Example 1 can be directly implemented as being executed by a hardware processor, or by a combination of hardware and software modules within the processor. The software module can be located in a storage medium well-established in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not given here.

[0068] A computer program product includes a computer program, and when the computer program is executed by a processor, the method described in embodiment 1 is implemented.

[0069] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions contained in program modules, which are executed in a device on a real or virtual processor of a target to perform the process / method described above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided between program modules as needed. The machine-executable instructions for the program modules can be executed in local or distributed devices. In distributed devices, program modules can be located in local and remote storage media.

[0070] The computer program code for implementing the method of the present invention can be written in one or more programming languages. These computer program codes can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the computer or other programmable data processing device, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on a computer, partially on a computer, as an independent software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0071] In the context of the present invention, computer program code or related data can be carried by any appropriate carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals include electrical, optical, radio, acoustic, or other forms of propagated signals, such as carrier waves, infrared signals, and the like.

[0072] Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0073] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. An artificial intelligence-based equipment fault diagnosis method, characterized in that: include: Obtain data related to the current fault of the device, as well as historical data containing data related to the current fault; Based on the acquired fault-related data, the health index of the equipment is evaluated based on the vibration severity; Input the fault-related data into the trained classification model and collective model respectively to obtain the equipment fault diagnosis results; input the time series features of the historical data into the trained time series model to obtain the equipment fault diagnosis results by capturing the dependency relationship in the time series; Fusing the fault diagnosis result obtained by the collective model, the fault diagnosis result obtained by the classification model, and the fault diagnosis result obtained by the time series model to obtain a final fault diagnosis result; Implement equipment fault diagnosis management based on the assessed equipment health index and final fault diagnosis results.

2. The artificial intelligence-based equipment fault diagnosis method according to claim 1, characterized in that: The fault-related data includes at least temperature, sound, vibration, magnetic flux and current, and Fourier transform and current fingerprint recognition technology are used to extract features from the fault-related data.

3. The artificial intelligence-based equipment fault diagnosis method according to claim 1 or 2, characterized in that: Based on the acquired fault-related data, the health index of the equipment is evaluated based on the vibration severity, specifically: Determine the corresponding vibration severity levels for different equipment; Calculate the vibration severity of each frequency band within the specified frequency range based on the equipment's fault-related data; According to the calculated vibration intensity of each frequency band and the determined vibration intensity level, the corresponding health index of the equipment is obtained.

4. The method for diagnosing equipment faults based on artificial intelligence according to claim 1, wherein: In the collective model, the FFT spectrum analysis method is used to identify abnormal equipment signals to capture the fault characteristics of the equipment in the frequency domain; the sound analysis method is used to identify potential problems and detect bearing abnormalities through changes in sound signals; and the principal component analysis method is used to judge abnormalities in vibration data to reveal hidden patterns and abnormal points in the data.

5. The method for diagnosing equipment faults based on artificial intelligence according to claim 1, characterized in that: Also includes: A health index deduction mechanism is established for the fault levels corresponding to different fault types to quantify the impact of the faults.

6. The method for diagnosing equipment faults based on artificial intelligence according to claim 1, characterized in that: The fault diagnosis result obtained by the collective model, the fault diagnosis result obtained by the classification model, and the fault diagnosis result obtained by the time series model are integrated to obtain the final fault diagnosis result, which is specifically: Determine the fusion model based on the Stacking fusion algorithm; The fault diagnosis result obtained based on the collective model, the fault diagnosis result obtained by the classification model, and the fault diagnosis result obtained by the time series model are processed using the fusion model to obtain a final fault diagnosis result.

7. An artificial intelligence-based equipment fault diagnosis system, characterized in that: include: An acquisition module is configured to: acquire data related to a current fault of the device, and historical data including the data related to the current fault; An evaluation module configured to: evaluate a health index of the equipment based on vibration severity according to the acquired fault-related data; A fault diagnosis module is configured to: input the fault-related data into the trained classification model and collective model respectively to obtain device fault diagnosis results; input the time series features of the historical data into the trained time series model to obtain device fault diagnosis results by capturing the dependency relationship in the time series; a fault diagnosis fusion module configured to fuse the fault diagnosis result obtained by the collective model, the fault diagnosis result obtained by the classification model, and the fault diagnosis result obtained by the time series model to obtain a final fault diagnosis result; The interactive management module is configured to implement fault diagnosis management of the device based on the assessed health index of the device and the final fault diagnosis result.

8. An electronic device, characterized in that: The method comprises a memory and a processor, and computer instructions stored in the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method according to any one of claims 1 to 6 is completed.

9. A computer-readable storage medium, characterized in that Used to store computer instructions, which, when executed by a processor, complete the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program, which is used to implement the method according to any one of claims 1 to 6 when the computer program is executed by a processor.

Images