Fault diagnosis method, fault diagnosis device and terminal of mechanical system

Abstract

The invention discloses a fault diagnosis method, a fault diagnosis device and a terminal of a mechanical system, and relates to the technical field of fault diagnosis of mechanical systems. The fault diagnosis method of the mechanical system comprises the following steps: disassembling the mechanical system, dividing an energy path and an information path, and marking transmission paths and nodes of the energy path and the information path; establishing a transfer model of the energy path and the information path, and determining a deviation between a final output result of the energy path and the information path transfer and the reality according to the established model; and determining the fault type of the specific path of the mechanical system according to the obtained deviation value. Through cooperative judgment of the energy path and the information path, the mechanical system is disassembled into the transmission path and the information transmission node, the complex mechanical system is simplified, fault confirmation of the power mechanism and the transmission mechanism is performed by using the transmission model based on the energy path, and preliminary judgment is provided for subsequent accurate fault positioning of the mechanical system.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical system fault diagnosis, and in particular to a mechanical system fault diagnosis method, a fault diagnosis device and a terminal. Background Art

[0002] Complex equipment is often formed by the coupling of multiple components or functional units. During operation, there are many sources of failure, and local failure of components will affect the overall operation of the equipment. At the same time, component failures of such complex equipment often have unclear characteristics in the early stages and are difficult to detect and isolate, posing a hidden danger to the normal operation of the equipment.

[0003] For example, a fault diagnosis method, a fault diagnosis device, and a terminal for a mechanical system disclosed in Chinese patent publication number CN111310589A, wherein the diagnostic method adaptively decomposes the state signal of a preset detection position of the mechanical system within a preset detection period to obtain multiple eigenmode functions, wherein the center frequency of the corresponding signal of each eigenmode function is different, at least one target eigenmode function is selected from the multiple eigenmode functions, and based on the at least one target eigenmode function, the mechanical system is diagnosed for faults, and the fault state of the mechanical system is judged by the target eigenmode function, thereby reducing the interference of non-fault signals on fault signals and improving the accuracy of fault diagnosis; when no fault occurs in the mechanical system, the target eigenmode function and the reference signal have a large deviation, which can more accurately determine the state of the mechanical system and improve the accuracy of fault diagnosis.

[0004] Existing mechanical system fault diagnosis mainly relies on a single data source, namely the signal decomposition of the detection component. However, due to the existence of power transmission in mechanical systems, if power transmission and signal transmission are not considered in a coordinated manner, the fault location will be ambiguous. It is difficult to accurately locate the fault link through threshold alarms or spectrum analysis alone. Especially in mechanical systems with multiple power paths in parallel, it is impossible to distinguish between primary and secondary fault sources. In-depth analysis of signals relies on high-algorithm equipment, which is costly and difficult to deploy. Summary of the Invention

[0005] In view of the deficiencies in the prior art, the present invention provides a fault diagnosis method, a fault diagnosis device and a terminal for a mechanical system, which solve the problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a fault diagnosis method for a mechanical system, the method comprising the following specific steps:

[0007] S100: Disassemble the mechanical system, divide the energy path and the information path, and mark the transmission paths and nodes of the energy path and the information path;

[0008] S200, establishing a transmission model of the energy path and the information path, and determining the deviation between the final output result of the energy path and the information path transmission and the actual result based on the established model;

[0009] S300: Determine the specific fault type of the mechanical system according to the obtained deviation value, perform a path search of the energy path, and complete the determination of the mechanical system fault.

[0010] A further improvement of the technical solution of the present invention is that the transmission path of the energy path in S100 includes several transmission links, and the node annotations of the information path in S100 include sensor nodes, control nodes and execution nodes.

[0011] A further improvement of the technical solution of the present invention is that: when disassembling the mechanical system, the power system and the transmission system have the same angular velocity. The transmission parts of the transmission link - transmission shaft, gears, bearings and other components are recorded as the same transmission link. The reason for marking the same transmission node is that the energy loss law of components with the same angular velocity is consistent, so the first The angular velocity of each transmission link is set to , the input torque is , the input power is According to the law of conservation of energy, in S200, the transfer model based on the energy path includes a transfer model of a single transmission link:

[0012] ;

[0013] in, = For the Input power of each link, For the The output power of each link, is the friction loss power, is the viscous friction loss power;

[0014] in = , For the The friction coefficient of each link is given by The material of the contact surface of the transmission parts of each link is determined by the manufacturer or instruction manual of the transmission parts used before the mechanical system fault diagnosis; , For the The viscous friction coefficient of each link is determined by the lubrication conditions and fluid shear characteristics, and is provided by the manufacturer or instruction manual of the transmission parts used before mechanical system fault diagnosis.

[0015] In summary, the mechanical system consists of n links in series between the power system and the transmission system. Therefore, referring to the calculation method of the transmission chain, the total output power of the mechanical system is the cumulative product of the initial input power and the efficiency of each link;

[0016] The transfer model also includes a transfer model with several links connected in series:

[0017] ;

[0018] in, is the input power of the first link, that is, the output power of the initial power source, Represents the overall loss of the overall link, is the final output power of the mechanical system. Furthermore, in some single-link mechanical systems, it is directly given by .

[0019] A further improvement of the technical solution of the present invention is that in S200, the energy path-based transfer model, when the mechanical system is in a variable speed or variable load condition, further includes:

[0020] ;

[0021] This application takes into account that if the mechanical system is in a variable speed or variable load condition, The torque and angular velocity of each transmission link change with time, so it is calculated by integration, where: For the Links at the moment The instantaneous torque, For the The angular velocity of each link, is the integral variable, expressed as a time domain At a certain moment within For the The friction coefficient of each link, For the The viscous friction coefficient of each link is, where, For the Links at the moment The instantaneous torque, For the The angular velocity of each link, is the integral variable, expressed as a time domain At a certain moment within For the The friction coefficient of each link, For the The viscous friction coefficient of each link.

[0022] A further improvement of the technical solution of the present invention is that: in S200, based on the information path transmission model, the state matrix and input matrix of each node are determined through experiments or simulations, a dynamic equation is established, and the time domain and frequency domain of each node signal are output as features, and finally the contribution of each node to the total output is analyzed;

[0023] Among them, the time domain feature output of each node signal includes:

[0024] ;

[0025] in, is the effective signal power, which can be calculated by dividing the square of the signal voltage by the load resistance, that is, , The power of background noise, usually generated by thermal noise, electromagnetic interference or mechanical vibration, and SNR is the ratio of the signal power to the noise power at the node. When the vibration sensor monitors the bearing fault, the effective vibration signal May be affected by mechanical noise Contamination, low SNR makes it difficult to extract fault features;

[0026] The frequency domain feature output of each node signal includes:

[0027] ;

[0028] Where, The node is in normal state at the frequency point The transfer function amplitude, such as the sensor output signal, is the transfer function amplitude of the node at the same frequency point in its current state, and N is the total number of analyzed frequency points;

[0029] The contribution of each node to the total output in each node signal includes:

[0030] ;

[0031] Where, For path In frequency The contribution of For path The transfer function reflects the transfer characteristics of the signal from the excitation source to the sensor. For path The input excitation signal, such as motor vibration and gear meshing force, is represented by M, which is the total number of parallel transmission paths in the information path.

[0032] For example, in a generator set, the vibration of the gearbox, bearings, impeller and other components is transmitted to each sensor through the structure, that is, each node. Size, judge a >0.5, then the node sensor and the detected component are the main vibration sources, and the component is The model algorithm is used to verify whether the component is abnormal.

[0033] A further improvement of the technical solution of the present invention is that in S300, the path search based on the energy path includes the following specific steps:

[0034] S301, the power source of the mechanical system is taken as the path starting point S, such as a motor or engine, the terminal actuator of the mechanical system (such as a wheel or a robotic arm) is taken as the end point T, and the intermediate transmission links such as gears, shafts, clutches, etc. are taken as links , the energy transfer path, such as shaft → gear → shaft, is a directed edge E, and the weight is the power loss and delivery efficiency ;

[0035] S302. Search all paths from the starting point S to the end point T, calculate the total efficiency and total power loss, and if multiple paths completely overlap in a certain section, and the overlapping section is the shortest path, retain only that section once.

[0036] S303: Analyze the power transfer status step by step along the energy path to locate the fault link.

[0037] A further improvement of the technical solution of the present invention is that: in said S303, the fault determination in the fault location link includes: | Time, link Fault, where For the The output power of each link, For the Rated output power of each link, The power deviation threshold that defines the health status of the node.

[0038] The present invention further provides a fault diagnosis device for a mechanical system, the device comprising:

[0039] Disassembly module: Based on the 3D modeling diagram or physical disassembly data of the mechanical system, decompose the system into energy paths and information paths, and mark the nodes and transmission paths;

[0040] Acquisition module: real-time acquisition of multi-source data during the operation of the mechanical system, including physical parameters of the energy path and signal characteristics of the information path;

[0041] Modeling and analysis module: builds energy path and information path transmission models based on collected data, calculates output deviations and locates faults;

[0042] Fault diagnosis module; determines the fault type based on the deviation value and locates the fault link through path search.

[0043] The present invention also provides a terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the mechanical system fault diagnosis method as described above when executing the computer program.

[0044] Beneficial effects

[0045] Compared with the prior art, the beneficial effects of the present invention are: through the coordinated judgment of energy paths and information paths, the mechanical system is disassembled into transmission paths and information transmission nodes, the complex mechanical system is simplified, and the transmission model based on the energy path is used to confirm the faults of the power mechanism and the transmission mechanism, providing a preliminary judgment for the subsequent precise positioning of mechanical system faults. Secondly, through the transmission model based on the information path, after the fault occurs, the fault diagnosis is performed on the transmission node on the information path, providing a basis for the precise positioning of the mechanical system fault. Finally, when it is determined that there is a fault in the power system, the power transfer status is analyzed step by step along the energy path through path search, and the analysis output based on the information path transmission model is combined to achieve precise positioning of the fault. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of a fault diagnosis method for a mechanical system;

[0047] Figure 2 This is a flowchart of path search in a fault diagnosis method for a mechanical system. DETAILED DESCRIPTION

[0048] Various exemplary embodiments, features, and aspects of the present application will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise indicated.

[0049] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0050] In addition, numerous specific details are provided in the following specific examples to better illustrate the present application. Those skilled in the art will appreciate that the present application can be practiced without certain specific details. In some instances, methods, means, and components well known to those skilled in the art are not described in detail in order to highlight the main purpose of the present application.

[0051] The present invention provides a fault diagnosis method for a mechanical system, which specifically includes the following steps:

[0052] S100. Disassemble the mechanical system and divide the energy path and information path. For example, the energy path includes the torque and power in the power system and transmission system, and the information path includes the sensor signals and control instructions of the actuator, support system, and control system. After marking the transmission paths and nodes of the energy path and information path, establish a transmission model. Based on the established model, preliminarily determine the final output of the energy path and information path transmission during mechanical system fault diagnosis;

[0053] The energy path in S100 includes several transmission links, and the node annotations of the information path in S100 include sensor nodes, control nodes, and execution nodes.

[0054] S200, establishing a transmission model of the energy path and the information path, and determining the deviation between the final output result of the energy path and the information path transmission and the actual result based on the established model;

[0055] Among them, when disassembling the mechanical system, the power system and the transmission system have the same angular velocity The transmission parts of the transmission link - transmission shaft, gears, bearings and other components are recorded as the same transmission link. The reason for marking the same transmission node is that the energy loss law of components with the same angular velocity is consistent, so the first The angular velocity of each transmission link is set to , the input torque is , the input power is , according to the law of conservation of energy, The input power and loss of each transmission link are separated as follows:

[0056] ;

[0057] in, = For the Input power of each link, For the The output power of each link, is the friction loss power, is the viscous friction loss power;

[0058] in = , For the The friction coefficient of each link is given by The material of the contact surface of the transmission parts of each link is determined by the manufacturer or instruction manual of the transmission parts used before the mechanical system fault diagnosis; , For the The viscous friction coefficient of each link is determined by the lubrication conditions and fluid shear characteristics, and is provided by the manufacturer or instruction manual of the transmission parts used before mechanical system fault diagnosis.

[0059] In summary, a mechanical system consists of n links in series between a power system and a transmission system. Therefore, referring to the calculation method of the transmission chain, the total output power of the mechanical system is the cumulative product of the initial input power and the efficiency of each link. The transfer model of the final output of the energy path can be obtained, including:

[0060] ;

[0061] in, is the input power of the first link, that is, the output power of the initial power source, such as the motor, Represents the overall loss of the overall link, is the final output power of the mechanical system. Furthermore, in some single-link mechanical systems, it is directly given by ;

[0062] The transfer model of the final output of the energy path can also include:

[0063] ;

[0064] This application takes into account that if the mechanical system is in a variable speed or variable load condition, The torque and angular velocity of each transmission link change with time, so it is calculated by integration, where: For the Links at the moment The instantaneous torque, For the The angular velocity of each link, is the integral variable, expressed as a time domain At a certain moment within For the The friction coefficient of each link, For the The viscous friction coefficient of each link;

[0065] In specific applications, in the transmission system of CNC machine tools, if the gearbox is assumed to be the second link, =0.03, =0.0015, input torque =50, angular velocity , then the efficiency of a single link is:

[0066] =0.9244, =1000W·0.9244=924.4W. If the measured value deviates significantly from the theoretical value, it means that there is a fault in the transmission system and power system that transmits energy of this mechanical system, and there is abnormal loss.

[0067] If there is a fault in the energy path, the path search based on the energy path includes the following specific steps:

[0068] S301, the power source of the mechanical system is taken as the path starting point S, such as a motor or engine, the terminal actuator of the mechanical system (such as a wheel or a robotic arm) is taken as the end point T, and the intermediate transmission links such as gears, shafts, clutches, etc. are taken as links , the energy transfer path, such as shaft → gear → shaft, is a directed edge E, and the weight is the power loss and delivery efficiency ;

[0069] S302. Search all paths from the starting point S to the end point T, calculate the total efficiency and total power loss, and if multiple paths completely overlap in a certain section, and the overlapping section is the shortest path, retain only that section once.

[0070] S303: Analyze the power transfer status step by step along the energy path to locate the fault link.

[0071] In terms of specific implementation, assume that the path S→ → →T and S→ → In S→ The segments overlap, if S→ is the most efficient path, then delete the repeated S→ part.

[0072] S301. Define the energy path as engine (S) → clutch → transmission → final reducer → wheel (T).

[0073] S302, deleting redundant paths, such as the least efficient path among the transmission parallel shift paths;

[0074] S303, detect the output power of the gearbox node 15% below the rated value, exceeding the threshold =10%, the moment model calculation shows that the actual transmission ratio i=2.8, deviating from the nominal value 3.0, according to the sensor detection results, the vibration energy The abnormal increase in meshing frequency is judged to be a gear wear fault.

[0075] In S303, the fault determination in the fault location step includes: Time, link Fault, where For the The output power of each link, For the Rated output power of each link, To define the power deviation threshold of the node health status, the diagnosis of specific components of the energy transmission path can be realized.

[0076] If the energy path is normal and there is no fault in the power transmission, use the following method to diagnose the fault of the information path.

[0077] When disassembling a mechanical system, it is broken down into sensor nodes, such as temperature and pressure sensors, control nodes, such as PLC controllers, and execution nodes, such as motors and hydraulic valves. The state matrix and input matrix of each node are determined through experiments or simulations. The time domain and frequency domain of each node signal are analyzed as features. Finally, the contribution of each sub-path to the total output is analyzed.

[0078] Based on the information path transmission model, the state matrix and input matrix of each node are determined through experiments or simulations, and dynamic equations are established. The time domain and frequency domain of each node signal are output as features, and finally the contribution of each node to the total output is analyzed.

[0079] Among them, the time domain feature output of each node signal includes:

[0080] ;

[0081] in, is the effective signal power, which can be calculated by dividing the square of the signal voltage by the load resistance, that is, , The power of background noise, usually generated by thermal noise, electromagnetic interference or mechanical vibration, and SNR is the ratio of the signal power to the noise power at the node. When the vibration sensor monitors the bearing fault, the effective vibration signal May be affected by mechanical noise Pollution and low SNR make it difficult to extract fault features. For example, in motor control, the position feedback output by the encoder, namely the sine / cosine signal, is susceptible to electromagnetic interference and circuit noise. For such mechanical system faults, the above model is used to perform time domain feature analysis on the signal at this node, as follows:

[0082] Input signal voltage to this node and noise voltage , signal voltage and noise voltage It can be collected by an oscilloscope, or based on historical maintenance data or given by the manufacturer and manual, and then the power can be calculated. , , assuming that =1V, =0.01V, R=50Ω, If the SNR is 40dB, the motor position error is less than 0.1%. However, in actual operation of the mechanical system, if the SNR drops to 20, the error exceeds 1%, causing motor control instability. The reason for the SNR reduction may be sensor aging or motor bearing wear exacerbating noise. Regardless of the cause, the precise cause and location are provided for actual fault diagnosis.

[0083] The frequency domain feature output of each node signal includes:

[0084] ;

[0085] Where, The node is in normal state at the frequency point The transfer function amplitude, such as the sensor output signal, is the transfer function amplitude of the node at the same frequency point in its current state, and N is the total number of analyzed frequency points;

[0086] In the fault diagnosis of the motor gearbox, the transfer function is known to be , the current After inputting into the model, assuming that the frequency point N is 100, then

[0087] =4.2; exceeds the set threshold of 3, and is judged as a gearbox abnormality;

[0088] The contribution of each node to the total output in each node signal includes:

[0089] ;

[0090] Where, For path In frequency The contribution of For path The transfer function reflects the transfer characteristics of the signal from the excitation source to the sensor. For path The input excitation signal, such as motor vibration and gear meshing force, is represented by M, which is the total number of parallel transmission paths in the information path.

[0091] For example, in a generator set, the vibration of the gearbox, bearings, impeller and other components is transmitted to each sensor through the structure, that is, each node. Size, judge a >0.5, then the node sensor and the detected component are the main vibration sources, and the component is Model algorithm to verify whether the component is abnormal;

[0092] Assume that the spindle of a CNC machine tool vibrates excessively during machining and the source of the vibration needs to be located. The system consists of three main transmission paths: spindle bearing, gearbox, and motor rotor. An accelerometer is used to measure the vibration signal, and the transfer function of each path is calibrated using the hammer excitation method. 、 、 , at the fault frequency =500Hz, calculate the contribution of each path: = =0.72, =0.18, ,In summary, the spindle bearing of path 1,contributes 72%, which is determined to be bearing wear or loose assembly.

[0093] The present invention also provides a fault diagnosis device for a mechanical system, the device comprising:

[0094] Disassembly module: Based on the 3D modeling diagram or physical disassembly data of the mechanical system, decompose the system into energy paths and information paths, and mark the nodes and transmission paths;

[0095] Acquisition module: real-time acquisition of multi-source data during the operation of the mechanical system, including physical parameters of the energy path and signal characteristics of the information path;

[0096] Modeling and analysis module: builds energy path and information path transmission models based on collected data, calculates output deviations and locates faults;

[0097] Fault diagnosis module; determines the fault type based on the deviation value and locates the fault link through path search.

[0098] The present invention also provides a terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the mechanical system fault diagnosis method as recited in any one of claims 1 to 7 when executing the computer program.

[0099] In a specific implementation, the present application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program. When executed by the data processing unit, the computer program can execute the invention of a mechanical system fault diagnosis method, fault diagnosis device, and terminal provided by the present invention, as well as some or all of the steps in each embodiment. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0100] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented via computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a computer program, or software product. This computer program software product can be stored in a storage medium and include instructions for enabling a device including a data processing unit (such as a personal computer, server, single-chip microcomputer, MCU, or network device) to execute the methods of various embodiments of the present invention or certain portions of these embodiments.

[0101] The present invention provides a mechanical system fault diagnosis method, fault diagnosis device, and terminal. While there are numerous methods and approaches for implementing this technical solution, the above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art may make various improvements and modifications without departing from the principles of the present invention, and such improvements and modifications are also within the scope of protection of the present invention. Any components not specified in this embodiment may be implemented using existing technologies.

Claims

1. A fault diagnosis method for a mechanical system, characterized in that: The method comprises the following specific steps: S100: Disassemble the mechanical system into energy paths and information paths, and label the transmission paths and nodes of the energy paths and information paths; S200: Establishing a transmission model of the energy path and the information path, and determining the deviation between the final output result transmitted by the energy path and the information path and the actual result based on the established model; S300: Determine the specific fault type of the mechanical system according to the obtained deviation value, perform a path search of the energy path, and complete the determination of the mechanical system fault.

2. A mechanical system fault diagnosis method according to claim 1, characterized in that: The energy path in S100 includes several transmission links, and the node annotations of the information path in S100 include sensor nodes, control nodes, and execution nodes.

3. The fault diagnosis method of a mechanical system according to claim 1, characterized in that: In S200, the energy path-based transfer model includes a single transmission link transfer model: ； in, = For the Input power of each link, For the The output power of each link, is the friction loss power, is the viscous friction loss power, assuming that The angular velocity of each transmission link is set to , the input torque is , the input power is ,but = , For the The friction coefficient of each link, , For the The viscous friction coefficient of each link; The transfer model also includes the transfer of several links in series, specifically: ； in, is the input power of the first link, that is, the output power of the initial power source, Represents the overall loss of the overall link, is the final output power of the mechanical system.

4. A mechanical system fault diagnosis method according to claim 1, characterized in that: In S200, the energy path transfer model, when the mechanical system is in a variable speed or variable load condition, further includes: ； Where, For the Links at the moment The instantaneous torque, For the The angular velocity of each link, is the integral variable, expressed as a time domain At a certain moment within For the The friction coefficient of each link, For the The viscous friction coefficient of each link.

5. The fault diagnosis method for a mechanical system according to claim 1, characterized in that: In S200 , based on the information path transfer model, the state matrix and input matrix of each node are determined through experiments or simulations, a dynamic equation is established, and the time domain and frequency domain of each node signal are output as features, and finally the contribution of each node to the total output is analyzed; Among them, the time domain feature output of each node signal includes: ； in, is the effective signal power, is the power of background noise, SNR is the ratio of node signal power to noise power; Frequency domain feature output of each node signal, including: ; Where, The node is in normal state at the frequency point The transfer function amplitude of is the transfer function amplitude of the node at the same frequency point in its current state, and N is the total number of analyzed frequency points; The comprehensive contribution of each node to the total output in each node signal, including: ； Where, For path In frequency The contribution of For path The transfer function reflects the transfer characteristics of the signal from the excitation source to the sensor. For path is the input excitation signal, and M is the total number of parallel transmission paths in the information path.

6. A mechanical system fault diagnosis method according to claim 1, characterized in that: The path search in S300 includes the following specific steps: S301, the power source of the mechanical system is taken as the path starting point S, the terminal actuator in the mechanical system is taken as the end point T, and the intermediate transmission link is taken as the link , the energy transfer path is the directed edge E, and the weight is the power loss and delivery efficiency ; S302. Search all paths from the starting point S to the end point T, calculate the total efficiency and total power loss, and if multiple paths completely overlap in a certain section, and the overlapping section is the shortest path, only retain the section once. S303: Analyze the power transfer status step by step along the energy path to locate the fault.

7. A mechanical system fault diagnosis method according to claim 6, characterized in that: In said S303, the fault determination in the fault location step includes: | Time, link Fault, where For the The output power of each link, For the Rated output power of each link, The power deviation threshold that defines the health status of the node.

8. A fault diagnosis device for a mechanical system, applied to the fault diagnosis method for a mechanical system according to any one of claims 1 to 7, characterized in that: The device comprises: Disassembly module: Based on the 3D modeling diagram or physical disassembly data of the mechanical system, decompose the system into energy paths and information paths, and mark the nodes and transmission paths; Acquisition module: real-time acquisition of multi-source data during the operation of the mechanical system, including physical parameters of the energy path and signal characteristics of the information path; Modeling and analysis module: builds energy path and information path transmission models based on collected data, calculates output deviations and locates faults; Fault diagnosis module; determines the fault type based on the deviation value and locates the fault link through path search.

9. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the mechanical system fault diagnosis method according to any one of claims 1 to 7 are implemented.

Images