Equipment abnormity diagnosis method and device

Abstract

The invention provides an equipment abnormality diagnosis method and device, which are used for obtaining a frequency constraint range based on parameters of equipment, matching a rotation related frequency of the equipment with the frequency constraint range, and determining whether the equipment is abnormal according to a matching result, so that more timely and more accurate equipment fault diagnosis can be realized. The method comprises the following steps: firstly, acquiring a rotation related frequency of the electromechanical transmission equipment; besides, the frequency constraint range of the electromechanical transmission equipment is further obtained, the frequency constraint range is obtained according to parameters of the electromechanical transmission equipment, and the frequency constraint range can be understood as the range of rotation related frequencies possibly generated when the electromechanical transmission equipment is in the fault working state; and matching the rotation-related frequency with the frequency constraint range, and determining whether the electromechanical transmission equipment has a fault according to a matching result of the rotation-related frequency and the frequency constraint range.

Description

Technical Field

[0001] The present application relates to the field of machinery, and in particular to a method and device for diagnosing equipment abnormality. Background Art

[0002] Rolling bearings, gears and motors, as the main transmission components, are widely used in key transmission equipment such as the three-electric systems of smart cars, mining belt conveyors, chemical centrifugal pumps, and high-speed rail running gear. The healthy operation of transmission equipment is crucial to ensuring the safety of the entire machine.

[0003] At present, the health monitoring method of rolling bearings, gears and other transmission components in the industry is usually based on vibration information analysis, and vibration sensors are mostly installed outside the transmission component housing. Therefore, the vibration information captured by the vibration sensor includes both the actual vibration of rolling bearings, gears and other transmission components, as well as the interference vibration transmitted through the transmission component housing or transmission shaft. Therefore, the noise in the collected vibration signal is usually large and the signal-to-noise ratio is small, which brings huge challenges to the predictive maintenance of key transmission components. Therefore, how to achieve more accurate abnormality detection of transmission equipment has become an urgent problem to be solved. Summary of the invention

[0004] The present application provides a method and device for diagnosing equipment abnormality, which are used to obtain a frequency constraint range based on equipment parameters, and match the rotation-related frequency of the equipment with the frequency constraint range, and determine whether the equipment is abnormal based on the matching result, so as to achieve more timely and accurate fault diagnosis.

[0005] In view of this, in a first aspect, the present application provides a method for diagnosing equipment abnormalities, comprising: first, obtaining a rotation-related frequency of an electromechanical transmission device, the rotation-related frequency including a frequency generated by the electromechanical transmission device when the electromechanical transmission device rotates; in addition, obtaining a frequency constraint range of the electromechanical transmission device, the frequency constraint range being obtained based on parameters of the electromechanical transmission device, and the frequency constraint range may include a range of possible rotation frequencies of the electromechanical transmission device when the electromechanical transmission device is in a faulty state; subsequently, identifying the rotation-related frequency through multi-frequency weak constraints, matching the rotation-related frequency and the frequency constraint range, and determining whether the electromechanical transmission device has a fault based on the matching result of the rotation-related frequency and the frequency constraint range.

[0006] In the present application, the frequency constraint range can be obtained based on the parameters of the electromechanical transmission device, and the rotation-related frequency can be matched with the frequency constraint range, so that it can be determined in time whether the electromechanical transmission device is abnormal.

[0007] In a possible implementation, the aforementioned rotation-related frequency may be a rotation frequency of the electromechanical transmission device directly received, or may be calculated based on the collected information; the rotation-related frequency of the electromechanical transmission device such as the aforementioned may include: obtaining vibration information, which may include the collected information generated by the vibration of the electromechanical transmission device during operation; then, obtaining the rotation-related frequency of the electromechanical transmission device based on the vibration information, which rotation-related frequency includes the frequency generated by the electromechanical transmission device when it rotates.

[0008] In a possible implementation, if the electromechanical transmission device includes a shaft system component, the rotation related frequency includes the rotation frequency of the shaft system component, and if the electromechanical transmission device includes a bearing, the rotation related frequency may also include a fault characteristic frequency of the bearing. Therefore, when the electromechanical transmission device includes different components, the calculated rotation related frequency may be used to represent the rotation frequency or the fault characteristic frequency when a rotation fault occurs, and thus may be applicable to a variety of scenarios.

[0009] In a possible implementation, the aforementioned acquisition of the rotation-related frequency of the electromechanical vibration device based on the vibration information may include: decomposing the vibration information to obtain at least one component, wherein each component can be used to represent the vibration condition of one or more components in the electromechanical transmission device; and then, calculating the rotation-related frequency corresponding to each component based on the at least one component. Therefore, in the implementation of the present application, the vibration signal can be decomposed to obtain the component representing the operating state of each component, thereby obtaining the rotation-related frequency of each component, and reducing the interference between signals by decomposing the signal, thereby improving the accuracy of the calculated rotation-related frequency.

[0010] In a possible implementation, the aforementioned calculation of the rotation-related frequency based on at least one component includes: obtaining at least one of the envelope signal, envelope spectrum signal or envelope harmonic noise spectrum of each component; calculating a frequency set of each component based on at least one of the envelope signal, envelope spectrum signal or envelope harmonic noise spectrum of each component, the frequency set including at least one of harmonic frequency, correlation frequency or modulation frequency, the correlation frequency being the frequency corresponding to the maximum value of the correlation coefficient in each component; obtaining the rotation-related frequency based on at least one of the frequency sets of each component. Therefore, in the implementation of the present application, multiple frequencies of each component can be calculated, so that the rotation-related frequency can be more accurately calculated through the constraint relationship between multiple frequencies.

[0011] In a possible implementation, the aforementioned obtaining of the rotation-related frequency according to at least one of the frequency sets of each component may include: if the frequencies in the frequency set satisfy a preset relationship, then using at least one of the harmonic frequency, the related frequency or the modulation frequency as the rotation-related frequency.

[0012] Therefore, in the implementation manner of the present application, multiple frequencies of each component may be calculated, and the rotation-related frequency corresponding to each component may be accurately determined based on the constraint relationship between the multiple frequencies.

[0013] In a possible implementation, the aforementioned decomposing the vibration signal to obtain at least one component may include: performing standardization processing on the vibration information to obtain a standardized signal; and decomposing the standardized signal to obtain at least one component.

[0014] Therefore, in the implementation manner of the present application, the vibration information can be decomposed after being standardized, thereby reducing the interference information included in the vibration signal and improving the accuracy of the rotation-related frequencies obtained by subsequent calculation.

[0015] In a possible implementation, the aforementioned determination of whether an electromechanical transmission device has a fault based on the matching results of the rotation-related frequency and the frequency constraint range may include: if the rotation-related frequency is within the frequency constraint range, determining whether the electromechanical transmission device has a fault based on an indicator of at least one dimension of the component corresponding to the rotation-related frequency; if all rotation-related frequencies are not within the frequency constraint range, determining that the electromechanical transmission device has not a fault based on the vibration information.

[0016] Therefore, in an implementation manner of the present application, a diagnostic fault-tolerant frequency range is set. When the rotation-related frequency matches the frequency constraint range, it is possible to further determine whether the device is faulty based on an indicator of at least one dimension of the component corresponding to the rotation-related frequency, thereby accurately identifying whether the device is faulty.

[0017] In a possible implementation, if the rotation-related frequency is within the frequency constraint range, then it is determined that the electromechanical transmission equipment has a fault, including: if the rotation-related frequency is within the frequency constraint range, obtaining an indicator of at least one dimension of the component corresponding to the rotation-related frequency; if the indicator of at least one dimension meets a preset condition, then it is determined that the electromechanical transmission equipment has a fault.

[0018] Therefore, in the implementation manner of the present application, indicators of multiple dimensions of abnormal components can be calculated, so as to jointly identify whether there is a fault from multiple dimensions and obtain a more accurate identification result.

[0019] In a possible implementation, the index of at least one dimension mentioned above includes one or more of the following: an impact index, an energy index or a harmonic noise index, the impact index includes at least one of kurtosis, information entropy, sparse measure or Gini index, the energy index represents the proportion of the component corresponding to the rotation-related frequency, and the harmonic noise index is determined according to the rotation-related frequency and its multiples. Therefore, in the implementation of the present application, whether the device is faulty can be comprehensively measured from dimensions such as signal impact, signal energy or harmonic noise index, so as to obtain accurate identification results.

[0020] In a possible implementation, if it is determined that the electromechanical transmission device has a fault, the aforementioned method may further include: determining the fault type of the fault according to the rotation-related frequency and the preset mapping relationship. Therefore, in the implementation of the present application, if it is determined that the device has a fault, the fault type of the device may also be determined based on the preset mapping relationship between the rotation-related frequency and the fault, thereby obtaining a more comprehensive diagnostic result.

[0021] In a possible implementation, the aforementioned method may further include: obtaining first information and second information, the first information being obtained based on data calculated in the aforementioned step of obtaining the rotation-related frequency of the electromechanical vibration device, and the second information being obtained based on data calculated in the step of determining whether the electromechanical transmission device has a fault based on a matching result between the rotation-related frequency and the frequency constraint range; saving the first information and the second information in archived data, and the data saved in the archived data is used to monitor the status of the electromechanical transmission device.

[0022] Therefore, in the implementation of the present application, the data calculated during the abnormal diagnosis process can be archived, that is, the data generated during the abnormal diagnosis process can be saved, so that the state change of the device can be traced back based on the saved data to realize the monitoring of the health status of the device.

[0023] In a possible implementation, the parameters of the electromechanical transmission device may include one or more of the following: bearing model, gearbox model, number of teeth of each gear in the gearbox, fault characteristic order of the inner ring, outer ring, cage or rolling element of the bearing. Therefore, in the implementation of the present application, the frequency constraint range can be calculated based on the specific structure of the electromechanical transmission device, so as to obtain a more accurate fault tolerance range, thereby achieving more accurate fault identification.

[0024] In a possible implementation, a vibration sensor is provided in the electromechanical transmission device, and the aforementioned acquisition of vibration information may include: obtaining vibration information through information collected by the vibration sensor. In the implementation of the present application, a vibration sensor may be provided in the electromechanical transmission device, and the vibration sensor may be provided usually near a component that needs to be diagnosed for a fault, so that vibration information of the component that needs to be diagnosed for a fault during operation can be collected, thereby achieving fault diagnosis of the component.

[0025] In a second aspect, the present application provides a device for diagnosing equipment abnormality, comprising:

[0026] A rotation-related frequency calculation module, used to obtain the rotation-related frequency of the electromechanical transmission equipment;

[0027] A constraint calculation module is used to obtain a frequency constraint range of the electromechanical transmission device, where the frequency constraint range is obtained based on parameters of the electromechanical transmission device;

[0028] The diagnosis module is used to determine whether the electromechanical transmission device has a fault according to the matching result of the rotation-related frequency and the frequency constraint range.

[0029] In a possible implementation, the rotation-related frequency calculation module is specifically used to: obtain vibration information, the vibration information includes information generated by the collected vibration of the electromechanical transmission device, and obtain the rotation-related frequency of the electromechanical transmission device according to the vibration information.

[0030] In a possible implementation, the rotation-related frequency calculation module is specifically used to: decompose the vibration information to obtain at least one component, wherein each component represents the vibration condition of one or more components in the electromechanical transmission device; and calculate the rotation-related frequency based on the at least one component.

[0031] In a possible implementation, if the electromechanical transmission device includes a shaft system component, the rotation-related frequency includes the rotation frequency of the shaft system component; if the electromechanical transmission device includes a bearing, the rotation-related frequency includes a fault characteristic frequency of the bearing.

[0032] In a possible implementation, the rotation-related frequency calculation module is specifically used to include: obtaining at least one of the envelope signal, envelope spectrum signal or envelope harmonic noise spectrum of each component; calculating a frequency set of each component based on at least one of the envelope signal, envelope spectrum signal or envelope harmonic noise spectrum of each component, the frequency set including at least one of the harmonic frequency, the correlation frequency or the modulation frequency, the correlation frequency being the frequency corresponding to the maximum value of the correlation coefficient in each component; obtaining the rotation-related frequency based on at least one of the frequency sets of each component.

[0033] In a possible implementation, the rotation-related frequency calculation module is specifically configured to: if the frequencies in the frequency set satisfy a preset relationship, use at least one of the harmonic frequency, the related frequency or the modulation frequency as the rotation-related frequency.

[0034] In a possible implementation manner, the rotation-related frequency calculation module is specifically used to: perform standardization processing on the vibration information to obtain a standardized signal; and decompose the standardized signal to obtain at least one component.

[0035] In one possible implementation, the diagnostic module is specifically used to: if the rotation-related frequency is within the frequency constraint range, determine whether the electromechanical transmission device has a fault based on an indicator of at least one dimension corresponding to the rotation-related frequency; if the rotation-related frequency is not within the frequency constraint range, determine that the electromechanical transmission device has not a fault based on vibration information.

[0036] In a possible implementation, the diagnostic module is specifically used to: if the rotation-related frequency is within the frequency constraint range, obtain an indicator of at least one dimension corresponding to each component; if the indicator of at least one dimension meets a preset condition, determine that the electromechanical transmission equipment has a fault.

[0037] In one possible implementation, the indicators of at least one dimension include one or more of the following: an impact indicator, an energy indicator, or a harmonic noise indicator. The impact indicator includes at least one of kurtosis, information entropy, sparsity measure, or Gini index. The energy indicator represents the proportion of the component corresponding to the rotation-related frequency. The harmonic noise indicator is determined based on the rotation-related frequency and its multiples.

[0038] In a possible implementation manner, the diagnosis module is further used to determine the fault type of the fault according to the rotation-related frequency and a preset mapping relationship.

[0039] In a possible embodiment, the above-mentioned device also includes an archiving module, which is used to: obtain first information obtained from data calculated in the rotation-related frequency step of the electromechanical vibration equipment, and second information obtained from data calculated in the step of determining whether the electromechanical transmission equipment has a fault based on the matching result of the rotation-related frequency and the frequency constraint range; save the first information and the second information in the archived data, and the data saved in the archived data is used to monitor the status of the electromechanical transmission equipment.

[0040] In a possible implementation, the parameters of the electromechanical transmission equipment include one or more of the following: bearing model, gearbox model, number of teeth of each gear stage in the gearbox, fault characteristic order of the bearing inner ring, outer ring, cage or rolling element.

[0041] In a possible implementation manner, a vibration sensor is provided in the electromechanical transmission device, and the acquisition module is specifically used to obtain vibration information through information collected by the vibration sensor.

[0042] In a third aspect, the present application provides an equipment abnormality diagnosis device, which includes: a processor, a memory, an input and output device, and a bus; the memory stores computer instructions; when the processor executes the computer instructions in the memory, the memory stores computer instructions; when the processor executes the computer instructions in the memory, it is used to implement any one of the implementation methods of the first aspect.

[0043] In a possible implementation, the device for diagnosing equipment abnormality provided in the present application may also include a vibration sensor for collecting information on vibrations generated when the mechanical equipment is working.

[0044] In a fourth aspect, an embodiment of the present application provides a chip system, which includes a processor and an input / output port, wherein the processor is used to implement the processing functions involved in the device abnormality diagnosis method described in the first aspect above, and the input / output port is used to implement the transceiver functions involved in the device abnormality diagnosis method described in the first aspect above.

[0045] In one possible design, the chip system also includes a memory, which is used to store program instructions and data for implementing the functions involved in the method described in the first aspect or any embodiment of the first aspect.

[0046] The chip system may be composed of chips, or may include chips and other discrete devices.

[0047] In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium. The computer-readable storage medium stores computer instructions; when the computer instructions are executed on a computer, the computer executes the method described in the first aspect or any possible implementation of any one of the embodiments of the first aspect.

[0048] In a sixth aspect, an embodiment of the present application provides a computer program product. The computer program product includes a computer program or instructions, and when the computer program or instructions are executed on a computer, the computer executes the method described in the first aspect or any possible implementation of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of the structure of an electromechanical transmission device provided in this application;

[0050] Figure 2 A schematic diagram of a process flow of a device diagnosis method provided in this application;

[0051] Figure 3 A schematic diagram of another device diagnostic method provided by the present application;

[0052] Figure 4 A schematic diagram of another device diagnostic method provided by the present application;

[0053] Figure 5 A schematic diagram of a vibration signal in the method provided by the present application;

[0054] Figure 6 A schematic diagram of a standardized vibration signal in the method provided by the present application;

[0055] Figure 7 A schematic diagram of a component in the method provided by this application;

[0056] Figure 8 A schematic diagram of a component and the corresponding envelope spectrum and its envelope-related harmonic noise spectrum in the method provided by the present application;

[0057] Fig. 9 A schematic diagram of another component and the corresponding envelope spectrum and its envelope-related harmonic noise spectrum in the method provided by the present application;

[0058] Fig.10 A schematic diagram of indicators in various dimensions in the method provided in this application;

[0059] Fig.11 This is another schematic diagram of indicators in various dimensions in the method provided in this application;

[0060] Fig.12 A schematic diagram of the structure of a device diagnostic device provided in this application;

[0061] Fig.13 A schematic diagram of the structure of another equipment diagnostic device provided in this application. DETAILED DESCRIPTION

[0062] The following will describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0063] The method provided in this application can be applied to electromechanical transmission equipment. Electromechanical transmission equipment is used in a large number of scenarios, such as smart car three-electric system, mining belt conveyor, chemical centrifugal pump, high-speed rail running gear and other key transmission equipment. The healthy operation of the transmission equipment is crucial to ensure the safety of the whole machine. Usually, a vibration sensor can be installed in the electromechanical transmission equipment, and the vibration sensor is mostly installed outside the transmission component housing. Therefore, the vibration information captured by the vibration sensor includes both the actual vibration of the transmission components such as rolling bearings and gears, and the interference vibration transmitted through the transmission component housing or the transmission shaft. The actual vibration information of the transmission components such as rolling bearings and gears is attenuated to varying degrees during the transmission process, especially in large and heavy-duty transmission systems. The vibration information of the transmission components such as rolling bearings and gears has a long attenuation path and strong interference vibration. The vibration intensity often fails to effectively identify faults, which brings huge challenges to the predictive maintenance of key transmission components. In the process of digital transformation of industrial equipment, it is found that it is difficult to present the health information of key core equipment in real time based on traditional methods such as vibration intensity, which greatly affects the progress of the industrial Internet.

[0064] In order to monitor the health status of equipment through complex vibration signals, a feature extraction method is proposed to monitor the health of equipment. It can usually be divided into strict periodic constraints and non-periodic constraints. Among them, the fault feature extraction method without periodic constraints, such as the fast-kurtogram method and the med method, although it has less prior input and is flexible in application, is easily affected by interference components such as random shocks. In practical applications, faults are frequently missed. For this reason, a method based on periodic constraints is proposed to replace the method without periodic constraints, such as the MCKD method and the cyclostationary deconvolution method. This type of method aims to effectively suppress the influence of random shock components by adding strict periodic constraints. However, in practical applications, the fault characteristic period of key transmission equipment is difficult to obtain a priori, and the current feature enhancement method based on periodic constraints only has a fault tolerance range of 3-5 frequency resolutions near the theoretical fault characteristic frequency. The fault characteristic period of actual equipment often exceeds this fault tolerance range. Therefore, the fault monitoring method based on strict periodic constraints has not achieved a robust diagnostic effect in practical applications, and huge production losses caused by missed faults occur for a long time.

[0065] Therefore, there is a stronger demand in practical applications for how to effectively extract the actual fault characteristic cycle and carry out robust health status monitoring based on the actual fault characteristic cycle constraints; a cycle estimation method based on the envelope harmonic noise ratio is also proposed. This method uses the signal with the strongest harmonic noise ratio of the periodic signal as the fault characteristic cycle. Unfortunately, this estimation method shows insufficient cycle estimation accuracy in practical applications. First, its noise definition cannot effectively reflect the actual noise level. The cycle corresponding to the maximum point of the relevant signal of the envelope signal is not always the fault cycle. Therefore, the cycle estimated based on the envelope harmonic noise ratio deviates seriously from the actual fault characteristic cycle, resulting in frequent fault misdiagnosis in the health monitoring method based on this method. How to effectively obtain the potential fault characteristic cycle in the actual signal and carry out high-precision monitoring of the health status of key transmission equipment based on this is of great significance to improving production efficiency, reducing production risks, and accelerating the reliable predictive maintenance process of the industrial Internet.

[0066] In addition, efficient and highly reliable vibration sensors are widely installed in a variety of transmission equipment. With the continuous operation of the equipment, various equipment manufacturers have accumulated massive data. However, due to the accuracy of the abnormal identification method, there are obvious deficiencies in the adaptive mining and management of high-value abnormal data. The data mining method based on expert experience has a high threshold and low efficiency, which is difficult to meet the mining work of massive data. Therefore, even if massive data is accumulated, it is still difficult to apply big data technology to carry out equipment health status monitoring. How to carry out value mining and management of massive data based on high-precision abnormal identification methods has received more and more attention from equipment manufacturers. At present, there is an urgent need for a high-value data mining and management system based on high-precision abnormal identification.

[0067] For example, the rolling bearing fault diagnosis method based on fault feature frequency multiple energy judgment first uses wavelet decomposition to decompose the signal, then uses kurtosis as the objective function, selects four components with the maximum kurtosis for reconstruction, and then takes the envelope spectrum of the reconstruction. Through strict fault feature frequency constraints, the spectrum energy ratio at the fault feature frequency multiple is obtained, and the bearing state monitoring is carried out according to the difference in the spectrum energy ratio at the fault feature frequency multiple of the normal bearing and the faulty bearing. This method belongs to a strict periodic constraint method, which requires a priori input of the accurate fault feature frequency, otherwise, the error of the fault feature frequency will increase significantly at its frequency multiple. If the priori input is inaccurate, the energy ratio based on the fault feature frequency multiple cannot reflect the actual fault information, so it is easy to cause missed diagnosis; moreover, in an environment with frequent impact interference, the reconstruction component is selected based on the size of kurtosis. When the fault impact is lower than the environmental impact, it will lead to missed diagnosis due to the exclusion of fault information. Therefore, this method has high requirements for application scenarios and has limited fault diagnosis accuracy in complex scenarios.

[0068] For another example, the bearing condition monitoring and fault diagnosis method under multiple working conditions first collects the vibration information of the bearing through the torque sensor, and calculates the multi-domain features of the vibration signal in the time domain, frequency domain and time-frequency domain, and divides the multi-domain features into the test set and the training set in proportion. Secondly, the multi-domain features are reduced in dimension and fused through the supervised linear discriminant. Finally, the support vector machine model is optimized with the help of the fusion features of the training set, and the optimized model is used for the bearing condition monitoring and fault diagnosis of the test set. This method aims to select a suitable feature set through massive historical data for intelligent diagnosis. This method first requires a lot of historical data and a large amount of faulty bearing sample data, which greatly increases the application threshold of this method. Moreover, when this model is used on different equipment in different service environments, the effect is significantly reduced, and it is difficult to effectively extract fault features, and fault misdiagnosis and missed diagnosis occur many times.

[0069] For example, there is an existing automatic diagnosis method and system for gearbox gear faults. This method first obtains the gearbox vibration acceleration signal and the meshing frequency of the target gear. The meshing frequency of the target gear needs to obtain the rotational frequency, which is a strict periodic constraint. It is often necessary to use a speed sensor to obtain accurate rotational frequency information. Once the rotational frequency input error is large, the meshing frequency error will be magnified. A series of characteristic indicators are obtained based on this, such as modulation frequency, meshing frequency energy ratio, sideband energy rate, sideband energy rate ratio and other key indicators will be seriously inconsistent with the actual situation, resulting in missed diagnosis of faults.

[0070] Another example is the harmonic identification method based on fast spectral kurtosis. This method obtains vibration signals of equipment such as bearings and gears, and filters the signals with the help of a 1 / 3 binary tree filter group to obtain multiple filtered signals, and obtains its envelope spectrum for each filtered signal. Each envelope spectrum is divided into intervals, and the frequency with the largest amplitude in each envelope spectrum sub-interval is used as the basic frequency. Based on the basic frequency, a harmonic search is performed in the corresponding sub-interval to determine the number of harmonics and the amplitude intensity of the corresponding harmonics. The harmonic intensity index of each sub-interval is calculated based on this, and the basic frequency of the first n sub-intervals is determined as the rotation-related frequency according to the size of the harmonic intensity index. This method aims to identify whether there are harmonic components of the fault modulation frequency in the vibration signal of the bearing gear, and obtain key constraints for further fault diagnosis. This method only requires the input of the vibration signal, but it is limited by the performance of the filter group, which may lead to the failure of harmonic extraction due to the unreasonable setting of the center frequency and bandwidth of the filter group. Secondly, the sub-interval division method based on the envelope spectrum of the filtered signal is highly random and difficult to meet the needs of robust extraction of harmonic components. Moreover, obtaining the basic frequency based on the harmonic intensity index is prone to missed identification of fault harmonic frequencies.

[0071] Therefore, the method based on strict periodic constraints often requires the input of accurate fault characteristic frequency, which requires the use of speed measurement devices, which will significantly increase the cost of the test plan and the workload of sensor installation. In particular, many transmission components do not have the installation conditions for speed sensors. Once the speed sensor is missing, the prior input is accurate or the estimation accuracy is insufficient, resulting in missed fault identification. Although the method without periodic constraints is flexible in application, the accuracy of fault feature extraction faces major challenges in scenarios with complex interference components and severe attenuation of fault features. In actual industrial sites, key transmission components are often in a relatively harsh service environment. The risk of missed fault diagnosis for key equipment is unbearable. Due to the huge losses caused by the failure, it is necessary to develop a more accurate fault identification method. With the continuous advancement of the industrial Internet, vibration data of many transmission equipment are being collected in large quantities. However, due to the lack of fault identification accuracy, the existing massive data lacks pure data labels. Relying on artificial label data is not only inefficient, but also has large errors. The model trained based on this data training set is prone to frequent misdiagnosis and missed diagnosis on site, which brings great trouble to operation and maintenance work, and the value of massive data cannot be fully explored.

[0072] Therefore, the present application provides a method for diagnosing equipment abnormality, which is used to calculate the vibration frequency of the equipment based on the collected vibration signal, and calculate its vibration frequency range based on the information of the equipment, and match the vibration frequency with the vibration frequency range, so as to accurately diagnose whether the equipment is abnormal.

[0073] First, the method provided in the present application can be applied to electromechanical transmission equipment, which may include bearings, gear boxes, etc. Vibration sensors can be installed at the positions of the components that need to be monitored in the electromechanical transmission equipment, such as pressure sensors, gyroscope sensors, air pressure sensors, magnetic sensors, or acceleration sensors. Usually, the electromechanical transmission equipment rotates during operation, and the rotation will cause the equipment itself to vibrate. The vibration signal generated by the electromechanical transmission equipment during rotation can be collected by the set acceleration sensor, so that the health status of the electromechanical rotating equipment can be monitored based on the vibration signal later. Specifically, the acceleration sensor can detect the magnitude of the acceleration of the equipment in each direction (generally three axes).

[0074] In addition, when the shaft system in various transmission systems such as chemical equipment pumps and coal mine machinery belt conveyors fails, similar fault cycle impact characteristics may also occur. Therefore, through the method provided by this application, fault diagnosis can also be performed on shaft system faults in rotating equipment such as chemical equipment pumps and coal mine machinery belt conveyors.

[0075] For example, the structure of the electromechanical transmission device can be as follows Figure 1As shown, the electromechanical transmission device may specifically include a motor 101, a planetary gearbox 102, a fixed-axis gearbox 103, a magnetic powder brake 104, a vibration sensor 105, and the like.

[0076] The motor 101 can be arranged at one end of the electromechanical transmission device, and its output end is connected to the gear box, such as the output end can be directly connected to one end of the planetary gear box 102, the other end of the planetary gear box 102 is connected to one end of the fixed axis gear box 103, and the other end of the fixed axis gear box 103 is connected to the output end of the magnetic powder brake 104. The vibration sensor 105 can be arranged at the component attachment that needs to be abnormally diagnosed, such as Figure 1 The top of the planetary gearbox housing is shown in FIG.

[0077] The motor 101 can be used to transfer energy, for example, a powered coil (ie, a stator winding) can be used to generate a rotating magnetic field and act on the rotor to form a magneto-electrodynamic rotating torque, thereby achieving energy transfer.

[0078] Generally, when a gear train is running, if the axis of each gear is fixed relative to the frame, the gear train is called a fixed axis gear train. On the contrary, if the axis of the gear moves relative to a plane or axis, it is called a planetary gear train. Fixed axis gear train is one of the two major types of gear trains.

[0079] The planetary gearbox 102, which may also be called a planetary reduction box or a gearbox, has a structure in which a plurality of planetary gears rotate around a sun gear. It also reduces the transmission speed ratio while proportionally increasing the motor torque.

[0080] The fixed axis gearbox 103 means that the rotation axes of all gears in the transmission have fixed positions. Its function is to transmit over a long distance, obtain a larger transmission ratio, and change the direction of the driven shaft to obtain a variety of transmission ratios.

[0081] The magnetic powder brake 104 is a device that has the function of slowing down, stopping or keeping a moving part (or moving machinery) in a stopped state. It is a part that stops or slows down a moving part in a machine.

[0082] The vibration sensor 105 can be used to collect vibration signals, and can specifically include a pressure sensor, a gyroscope sensor, an air pressure sensor, a magnetic sensor, or an acceleration sensor, etc. It can be used to detect small displacements or speed changes caused by vibration on the surface of an object, and generate vibration signals.

[0083] Bearings are also installed inside the transmission box. Bearings are an important component in electromechanical transmission equipment. Their main function is to support the mechanical rotating body, reduce the friction coefficient during its movement, and ensure its rotation accuracy. For example, they can be installed in a fixed-axis gearbox to fix the axis.

[0084] It should be noted that the number of vibration sensors 105 can be one or more, and their installation positions can be determined based on actual needs. For example, if it is necessary to focus on abnormal monitoring of the gearbox, a vibration sensor can be installed near the gearbox; if it is necessary to focus on abnormal monitoring of the bearing, a vibration sensor can be installed near the bearing. The specific installation position of the vibration sensor in this application is not limited.

[0085] The method provided in the present application can be specifically deployed in a variety of devices, such as electromechanical transmission equipment, terminals or servers and other devices used to diagnose faults in electromechanical transmission equipment, and cloud platforms, etc., to perform remote fault diagnosis through the cloud platform, etc. The deployed equipment can be selected according to the actual application scenario.

[0086] The method flow provided in this application is introduced below in conjunction with the structure of the aforementioned electromechanical transmission equipment.

[0087] See also Figure 2 , a flow chart of a device abnormality diagnosis method provided in the present application is as follows.

[0088] 201. Obtain the rotation-related frequency of the electromechanical transmission equipment.

[0089] The rotation-related frequency may be the rotation frequency of the electromechanical transmission device directly received, or may be calculated based on the collected information. For example, the rotation-related frequency may be obtained by directly outputting the rotation frequency of the electromechanical transmission device through a vibration sensor, or the rotation-related frequency may be calculated based on the information collected by the vibration sensor. The present application is illustratively introduced by taking the example of calculating the rotation-related frequency from the information collected by the vibration sensor.

[0090] Generally, when the electromechanical transmission device is working, vibration will be generated, and the vibration information may include information generated by the electromechanical transmission device during vibration collected by the vibration sensor. For example, if the vibration sensor is an acceleration sensor, the acceleration sensor may be used to collect signals when the electromechanical transmission device is vibrating, and thus the vibration information of the electromechanical transmission device may be obtained through the data collected by the acceleration sensor, or the vibration information may include the signal collected by the acceleration sensor.

[0091] It should be noted that, usually in the process of monitoring the health status of electromechanical transmission equipment, the vibration information mentioned in this application can be vibration information collected in any unit time. For example, if the vibration signal is collected in units of 1ms, the vibration information may include one or more vibration signals per unit time.

[0092] Usually, the electromechanical transmission device rotates when working, and the electromechanical transmission device vibrates under the influence of the rotation. Usually, if the electromechanical transmission device rotates regularly, the vibration it generates also conforms to a certain rule. If the electromechanical transmission device is abnormal, its vibration may also be affected. Therefore, the vibration information of the electromechanical transmission device usually collected can reflect the rotation of the electromechanical transmission device. The present application can further determine whether the electromechanical transmission device is abnormal through the collected vibration information.

[0093] After the vibration information is obtained, the rotation frequency of the electromechanical transmission device during rotation can be calculated based on the vibration information. It should be understood that when diagnosing abnormalities in electromechanical equipment, the status of each component of the electromechanical transmission device can usually be diagnosed. Therefore, the rotation-related frequency mentioned in this application may include the rotation frequency of each component in the electromechanical transmission device, and the rotation frequency of each component may be the same or different. It may also include the frequency of each component of the electromechanical device passing through the fault position, that is, the rotation-related frequency may be the rotation frequency or the fault characteristic frequency.

[0094] In a possible implementation, if the electromechanical transmission device includes a shaft system component, the rotation-related frequency includes the rotation frequency of the shaft system component, and if the electromechanical transmission device includes a bearing, the rotation-related frequency includes the fault characteristic frequency of the bearing. Therefore, when the electromechanical transmission device includes different components, the calculated rotation-related frequency can be used to represent the frequency of rotation or the fault characteristic frequency during rotation fault, so that it can be applied to a variety of scenarios. Usually, for bearings, the characteristic frequency of the bearing fault is usually the vibration frequency generated by the bearing fault, and the bearing fault frequency is usually a non-integer multiple of the rotation speed frequency.

[0095] Optionally, the vibration information can be decomposed into signals, and the rotation-related frequency of the electromechanical transmission device can be calculated based on the decomposed signals. Specifically, the vibration information can be decomposed to obtain at least one component, and all decomposed mode components are called decomposition domain signals, each component can represent the vibration of one or more parts in the electromechanical transmission device, and then the rotation-related frequency of the electromechanical transmission device can be calculated based on the at least one component.

[0096] Optionally, when calculating the rotation-related frequency, one or more envelope signals, envelope spectrum signals or envelope harmonic noise spectrum signals of each component can be calculated, and then the frequency set corresponding to each component is calculated based on the calculated one or more signals. The frequency set may include one or more harmonic frequencies, correlation frequencies or modulation frequencies. The correlation frequency is the frequency corresponding to the maximum value of the correlation coefficient. Then, the rotation-related frequency corresponding to each component can be determined based on the harmonic frequency, correlation frequency and modulation frequency of each component.

[0097] Optionally, when determining the rotation-related frequency corresponding to each component according to the harmonic frequency, the related frequency and the modulation frequency, if the preset relationship is satisfied between the harmonic frequency, the related frequency and the modulation frequency, such as the multiples between the harmonic frequency, the related frequency and the modulation frequency are within a preset range, the harmonic frequency, the related frequency and / or the modulation frequency can be used as the rotation-related frequency. Therefore, in the embodiment of the present application, the rotation-related frequency can be calculated by calculating multiple frequencies such as the harmonic frequency, the related frequency or the modulation frequency, so that the rotation-related frequency can be accurately calculated based on the constraints between the harmonic and the modulation wave.

[0098] In some possible scenarios, the vibration information can be standardized to obtain a standardized signal. Usually, in order to facilitate calculation and eliminate the differences in individual service environments, the vibration signal can be standardized. When performing standardization, a standardization algorithm such as the Z-score standard algorithm or deviation standardization can be used. The appropriate standardization algorithm can be selected according to the actual application scenario.

[0099] Optionally, in order to avoid data loss, the vibration information can be saved before being standardized, or basic time domain features can be extracted from the vibration information and saved. For example, the mean square amplitude information (rms), frequency band energy and other information of the vibration signal can be calculated and stored, so that the health status of the equipment can be monitored based on the information carried by the collected raw data, so as to timely know the service life of the equipment.

[0100] Optionally, in the implementation manner of the present application, a sample digital file, or a storage file or a sample file, etc., may be established for the electromechanical transmission device or each component in the electromechanical transmission device, which may be determined specifically according to the actual application scenario. The sample digital file may be used to store data of the electromechanical transmission device or each component in the electromechanical transmission device, so as to monitor the state change of the electromechanical transmission device, timely update the service life of the device, or perform timely maintenance on the device, etc.

[0101] 202. Obtain a frequency constraint range of the electromechanical transmission equipment.

[0102] The frequency constraint range may be obtained or calculated based on the parameters of the electromechanical transmission device, and in the implementation manner of the present application, may be used to indicate the range in which the rotation-related frequency of the electromechanical transmission device may be when a fault occurs.

[0103] Specifically, when calculating the frequency constraint range, the calculation can be based on one or more of the following parameters: bearing model, gearbox model, number of teeth of each gear stage in the gearbox, fault characteristic order of the bearing inner ring, outer ring, retainer or rolling element, etc. Usually these parameters are fixed or fixed within a certain period, so the possible rotation-related frequencies of the equipment during normal operation can be known based on these parameters.

[0104] In some scenarios, the frequency constraint range can be understood as the range of rotation-related frequencies that may be generated by the electromechanical transmission device when a fault occurs. If the electromechanical transmission device has a parameter indicating its rotation frequency when it fails, when determining the frequency constraint range, the frequency constraint range can usually cover the characteristic frequencies of each fault order when the fault is indicated by the parameter, so that the frequency constraint range can only cover the rotation-related frequencies that may be generated by the electromechanical transmission device when it fails.

[0105] Of course, in some scenarios, the complement of the rotation-related frequencies that may be generated by the electromechanical transmission device may also be taken, that is, the frequency constraint range may also be set to the vibration frequency that may be generated when the electromechanical transmission device is working normally. In this scenario, if the electromechanical transmission device is working normally, its rotation-related frequencies are within the frequency constraint range. The specific setting method can be determined in the actual application scenario, and this application does not limit this.

[0106] It should be noted that, in this application, the frequency constraint range is exemplarily described as the possible rotation-related frequency of the electromechanical transmission device when it fails. The frequency constraint range mentioned below indicates the range of possible rotation-related frequency of the electromechanical transmission device when it fails.

[0107] 203. Determine whether the electromechanical transmission device has a fault based on the matching result of the rotation-related frequency and the frequency constraint range. If they do not match, determine that the electromechanical transmission device has no fault. If they match, further fault identification is required.

[0108] After calculating the rotation-related frequency and frequency constraint range of the electromechanical transmission device, the rotation-related frequency and the frequency constraint range can be matched, such as determining whether the rotation-related frequency is within the frequency constraint range, and determining whether the electromechanical transmission device has a fault based on the matching result.

[0109] In the implementation manner of the present application, the frequency constraint range that may be generated by the electromechanical transmission device during rotation can be determined based on the parameters of the electromechanical transmission device, and the actual rotation-related frequency can be obtained based on the vibration information of the electromechanical transmission device actually collected. The actual rotation-related frequency is compared with the possible frequency constraint range to determine whether the electromechanical transmission device is faulty.

[0110] In a possible implementation, it can be specifically determined whether the rotation-related frequency is within the frequency constraint range. If the rotation-related frequency is not within the frequency constraint range, it indicates that the electromechanical transmission device is in a normal working state, that is, no fault occurs. If the rotation-related frequency is within the frequency constraint range, it indicates that the electromechanical transmission device may have a fault based on the collected vibration information. At this time, it can be further confirmed whether the electromechanical transmission device has a fault or the type of fault that has occurred. Therefore, in the implementation of the present application, the actual rotation-related frequency of the electromechanical transmission device is matched with the fault tolerance range, combined with the characteristic indicators, so as to accurately determine whether the electromechanical transmission device has a fault.

[0111] In a possible implementation, if the rotation-related frequency is within the frequency constraint range, an indicator of at least one dimension corresponding to the vibration information can be obtained. If the indicator of at least one dimension meets a preset condition, it is determined that the electromechanical transmission device has a fault.

[0112] Optionally, the index of at least one dimension may include but is not limited to one or more of the following: an impact index, an energy index or a harmonic noise index. The impact index may specifically include one or more of kurtosis, information entropy, sparseness measure or Gini index, etc. The energy index may be used to indicate the proportion of components corresponding to rotation-related frequencies, and the harmonic noise index is determined according to the target and its multiples. Generally, the indexes of each dimension may be used to measure the degree of abnormality of the components represented by each component identified based on vibration information.

[0113] When only one of the dimensions is calculated, the indicator of the one dimension can be compared with the threshold to determine whether the electromechanical transmission device has a fault. When multiple dimensions are calculated, a joint judgment can be made based on the indicators of the multiple dimensions to determine whether the electromechanical transmission device has a fault and the severity of the fault.

[0114] Optionally, the impact index can be used to identify whether each component has a fault, and the fault severity can be evaluated by combining the impact index, harmonic noise index or energy index, etc. Therefore, the present application can analyze the components corresponding to each component from multiple dimensions, so as to more accurately identify whether each component has a fault and the fault severity when it fails, etc. through joint analysis.

[0115] In one possible scenario, if it is determined based on vibration information that an electromechanical transmission device has a fault, the fault type can be further identified based on the rotation-related frequency. For example, the fault type of the fault can be determined based on the rotation-related frequency and a preset mapping relationship, and the preset mapping relationship may include a mapping relationship between a plurality of pre-set frequencies and fault types. Therefore, the present application can not only accurately identify whether an electromechanical transmission device has a fault, but can also further identify the fault type when it is determined that a fault has occurred, so that the user can accurately know the working status of the electromechanical transmission device, and thus perform corresponding processing on the fault of the electromechanical transmission device, thereby improving the working efficiency of the electromechanical transmission device.

[0116] Optionally, first information can be obtained based on data calculated in the step of determining the rotation-related frequency of the electromechanical vibration device, and second information can be obtained based on data calculated in the step of determining whether the electromechanical transmission device has a fault based on the matching result of the rotation-related frequency and the frequency constraint range; the first information and the second information can be saved in the archived data, and the data saved in the archived data is used to monitor the status of the electromechanical transmission device.

[0117] The above describes the process of the device abnormality diagnosis method provided by the present application. The following describes the method process provided by the present application in more detail in combination with specific application scenarios.

[0118] First, the device abnormality diagnosis method provided by this application can be divided into multiple stages, such as Figure 3 As shown, it can be specifically divided into: information input 301, frequency identification 302, health monitoring 303 and data archiving 304.

[0119] Among them, in the information input stage, vibration signals, weak period constraint range, and structural parameters are collected through sensors.

[0120] In the frequency identification stage, the envelope-related harmonic noise spectrum, the strongest modulation frequency, the strongest related frequency, the strongest harmonic frequency, etc. are calculated based on the vibration signal.

[0121] In the health monitoring stage, characteristic indicators are calculated according to the data obtained in the frequency identification stage, and abnormality identification and fault diagnosis are performed based on the characteristic indicators, that is, the fault type is identified and the fault is located.

[0122] In the data health stage, various collected data are saved, such as key result diagrams, cycle characteristics, impact characteristics, energy characteristics, intensity characteristics and other multi-dimensional indicators.

[0123] The following introduces each stage in combination with specific application scenarios.

[0124] See also Figure 4 , a flow chart of another device abnormality diagnosis method provided by the present application is described as follows.

[0125] 401. Collect vibration signals.

[0126] Usually, a sensor can be set near the component that needs to be diagnosed abnormally, and of course, sensors can also be set at various positions of the electromechanical transmission equipment. The sensor can be a pressure sensor, a gyroscope sensor, an air pressure sensor, a magnetic sensor, or an acceleration sensor, etc. This application exemplarily introduces an acceleration sensor as an example, which can be determined according to the actual application scenario, and will not be repeated below.

[0127] For example, a vibration acceleration sensor can be installed on a rolling bearing seat or a gearbox housing near the target bearing or gear, and the vibration acceleration sensor can be used to collect vibration information of the equipment and output a vibration signal. The vibration signal can be Figure 5 shown.

[0128] 402. Extract time domain features.

[0129] Usually, when performing a signal vibration signal When performing standardization operations, many basic time domain features may be lost, such as signal mean square amplitude information rms, frequency band energy, etc. Therefore, before performing signal standardization, the basic time domain features of the signal can be calculated and stored in a sample digital file. The sample digital file can be a storage space established for electromechanical transmission equipment and used to store the data of the monitored electromechanical transmission equipment.

[0130] It should be understood that step 402 is an optional step, that is, the time domain features can be stored, or only real-time abnormality diagnosis can be performed without storage. Whether to execute step 402 can be determined based on the actual application scenario, and this application does not limit this.

[0131] 403. Signal standardization.

[0132] Signal standardization can also be called signal normalization. Specifically, the Z-score standardization algorithm or deviation standardization algorithm can be used to normalize the vibration signal. Perform standardization to eliminate the differences in individual service environments and obtain a standardized vibration signal x(t). For example, the standardized vibration signal can be as follows: Figure 6 shown.

[0133] Therefore, in the implementation manner of the present application, the influence of environmental noise on the collected vibration signal can be reduced by signal standardization, thereby reducing the interference with the subsequent identification of rotation-related frequencies and improving the accuracy of the calculated rotation-related frequencies.

[0134] It should be understood that step 403 is an optional step, that is, signal standardization may not be performed. The specific selection can be made according to the actual application scenario, and this application does not limit this.

[0135] 404. Signal decomposition.

[0136] Generally, when an electromechanical transmission device is in operation, multiple components rotate, so the collected vibration signal may include vibration information generated by the operation of multiple components. Therefore, the vibration signal can be decomposed to obtain multiple mode components, and each mode component can be understood as representing one of the vibration conditions in the electromechanical transmission device.

[0137] The standardized vibration signal x(t) can be decomposed by a signal decomposition algorithm, such as singular value decomposition, empirical mode decomposition or wavelet decomposition. All mode components after decomposition are called decomposition domain signals. This application exemplarily introduces the decomposition of the standardized vibration signal x(t) by using a singular value decomposition algorithm. The singular value decomposition algorithm and the components obtained by decomposition mentioned below can also be replaced by other decomposition algorithms and corresponding components, which will not be repeated below.

[0138] For example, the standardized signal x(t) can be decomposed into several mode components x by singular value decomposition. i (t) and the corresponding singular values ​​σ i , each component can usually represent the vibration of different parts, and the decomposed components can be as follows Figure 7 shown.

[0139] 405. Calculate the envelope-correlated harmonic noise spectrum.

[0140] After the signal is decomposed to obtain multiple mode components, each mode component can be envelope demodulated to obtain the envelope signal xenv of each mode component. i (t), envelope spectrum signal Xenv i (f) and the envelope-correlated harmonic noise spectrum Xauto i (f), Envelope-correlated harmonic noise spectrum Xauto i (f) is expressed as:

[0141]

[0142] Among them, N n,k (f) is the envelope spectrum Xenv i (f) Spectral noise level at frequency f, n, k represent the number of iterations and the number of fitting points, respectively, N n,k The calculation method of (f) can be expressed as:

[0143] make Find N 0,k All maximum points of (f);

[0144] N n k (f) represents the noise estimation result of the nth iteration envelope spectrum at frequency f, which can be specifically expressed as:

[0145]

[0146] e n,k (f) represents the iteration error, when max(e n,k When (f))<ε, the iteration terminates, e n,k (f) The calculation result is expressed as:

[0147] e n,k (f) = N n,k (f)-N n-1,k (f)

[0148] where ε can be set by setting a threshold ε0 or ε=min(N 0,k (f)) OK.

[0149] 406. Multi-frequency judgment identifies characteristic frequencies.

[0150] Usually, after calculating the envelope correlated harmonic noise, multiple frequencies of the components, such as modulation frequency, correlation frequency or harmonic frequency, can be identified through the envelope correlated harmonic noise spectrum, and one of them can be used as the rotation-related frequency of the device, referred to as the rotation-related frequency.

[0151] The present application exemplarily introduces the calculation of the modulation frequency, related frequency and harmonic frequency of the component. In some application scenarios, it can also be replaced by calculating only one or more of them. The specific selection can be based on the actual application scenario, and the present application does not limit this.

[0152] For example, based on the envelope-correlated harmonic noise spectrum of the i-th component signal, the strongest harmonic frequency f is calculated i0 , f i0 The calculation process can be expressed as:

[0153]

[0154] Calculate the strongest modulation frequency f i1 , f i1 The calculation process can be expressed as:

[0155]

[0156] Calculate the strongest correlation frequency f i2 , f i2 The calculation process is as follows:

[0157]

[0158] Among them, Argmax τ (·) represents the maximum likelihood estimation operator.

[0159] Usually, f i0 、f i1 With f i2 There is a frequency doubling relationship between them, so when f i0 、f i1 、f i2 When the following relationship is satisfied between them, f that satisfies the following relationship can be i0 、f i1 、f i2 As the characteristic frequency corresponding to the current mode component:

[0160] min({|m*f i0 -nf i1 |,|m*f i2 -nf i1 |}) <f ε

[0161] Where min(·) indicates the minimum value operation, m and n indicate the frequency multiplication, usually m and n are 1 to 5, and f ε Indicates the frequency constraint range, generally f ε =c*f s / N, c is a constant, usually 1 to 5, N represents the data length, f s / N represents the frequency resolution.

[0162] When f i0 、f i1 、f i2 When the above relationship is not satisfied, the corresponding f i0 、f i1 、f i2 The rotation-related frequency is not regarded as the rotation-related frequency, and the rotation-related frequency is assigned a negative frequency. However, the actual rotation-related frequency cannot be a negative number. Therefore, it is possible to avoid misleading subsequent abnormal diagnosis due to the calculated erroneous frequency value.

[0163] Therefore, the present application provides accurate identification of rotation-related frequencies through envelope-related harmonic noise spectra, which can ensure the accuracy of fault information mining, identify fault characteristic frequencies, and conduct high-precision fault diagnosis.

[0164] 407. Calculate the frequency constraint range by combining gear parameters, bearing parameters or speed range.

[0165] You can get the model of the bearing and gearbox, as well as the number of teeth or transmission ratio of each gear pair in the gearbox.i , where i = 1, 2, 3…, represents the i-th level of deceleration, and the fault-tolerant speed range of the equipment is obtained based on the service parameters of the equipment, such as the rated speed or the actual gear position. If expressed as [f min ,f max ], ensuring that the fault-tolerant speed range completely covers the actual speed, and setting the fault-tolerant speed range narrow enough as much as possible. The frequency constraint range can be understood as the rotation-related frequency range when the electromechanical transmission equipment may fail. The frequency constraint range can be obtained based on the fault-tolerant speed range of the equipment. For example, the fault-tolerant speed range can be directly used as the frequency constraint range, or a certain range can be added as the frequency constraint range based on the fault-tolerant speed range, etc. The specific range can be determined according to the actual application scenario.

[0166] For example, for the bearing abnormality monitoring scenario, the fault feature orders of the inner ring, outer ring, cage, and rolling element of the bearing are queried according to the bearing model and recorded as inn_ord, out_ord, cag_ord, and bal_ord respectively. The theoretical fault feature order set TOS = {inn ord ,out ord , cag ord ,bal_ord}, ​​according to the actual fault monitoring needs, the actual fault feature order set AOS is the theoretical fault feature order set TOS or its non-empty subset, then the weak period constraint range (that is, the possible range of the fault feature frequency) can be expressed as: [tfmin, tfmax] = [min(AOS)*fmin, max(AOS)*fmax].

[0167] For example, in the gearbox abnormality monitoring scenario, based on the transmission ratio ri, the speed range of each level [fmini, fmaxi] is calculated, and the weak periodic constraint range can be expressed as: [tfmin, tfmax] = [min{fmini}, max(fmaxi)].

[0168] For example, in the abnormal monitoring scenario of a motor, the current of the motor can usually be tested, and the vibration usually comes from the internal bearings. Therefore, only the current value can be compared with the pre-set current range to identify whether the device is faulty.

[0169] 408 . Determine whether the characteristic frequency matches the frequency constraint range. If not, execute step 409 . If yes, execute step 410 .

[0170] In the implementation mode of the present application, the frequency constraint range is calculated based on the prior input bearing, gearbox parameters and reasonable speed range, and then the rotation-related frequency can be determined by determining whether the rotation-related frequency is included in the frequency constraint range to determine whether the rotation-related frequency matches the frequency constraint range, which can be f i0 、fi1 If any one of them is included in the frequency constraint range, it is determined that the rotation-related frequency matches the frequency constraint range, and all matching decomposition domain signals are called target domain signals.

[0171] For example, the fourth mode component, envelope spectrum and its envelope-related harmonic noise spectrum can be expressed as Figure 8 As shown, the strongest modulation frequency is equal to the strongest harmonic frequency, i.e., f i1 =f i0 =50.0122, if f i0 or i1 If f is included in the frequency constraint range, then the frequency is the rotation-related frequency, and the corresponding decomposition domain signal is the target domain signal. i0 、f i1 If both are not included in the frequency constraint range, it is determined that the rotation-related frequency does not match the frequency constraint range, and the corresponding decomposition domain signal is not the target domain signal.

[0172] For another example, the seventh mode component, envelope spectrum and its envelope-related harmonic noise spectrum can be expressed as Fig. 9 As shown, the strongest modulation frequency f i1 =64.0781, the strongest correlation frequency f i2 =10.9402, the strongest harmonic frequency is f i0 =7.8144, which does not satisfy the characteristic frequency disclosure. This means that the frequency is not a rotation-related frequency, and the corresponding decomposition domain signal is not a target domain signal.

[0173] When the vibration signal All corresponding mode components x i If the rotation-related frequencies of (t) are not within the frequency constraint range, the vibration signal It is judged as a normal sample and the device is normal, that is, there is no target domain signal at this time.

[0174] 409. Ensure that the equipment is operating normally.

[0175] In one scenario, when the vibration signal All corresponding mode components x i If the characteristic frequencies of (t) do not meet the rotation-related frequency matching, then the vibration signal It is judged as a normal sample, that is, the electromechanical transmission equipment operates normally.

[0176] It can be understood that if the characteristic frequencies of all mode components in the vibration signal are within the frequency range corresponding to normal operation, it means that the electromechanical transmission equipment is operating normally.

[0177] After determining that the electromechanical transmission equipment is operating normally, step 415 may be continued to be executed. Of course, step 415 may also be omitted, and the operating status of the electromechanical transmission equipment may continue to be monitored. The specific selection may be based on the actual application scenario, and this application does not limit this.

[0178] 410. Calculate multidimensional indicators.

[0179] After determining that the working state of the electromechanical transmission equipment may be abnormal, the indicators of the target domain signal in multiple dimensions can be further calculated.

[0180] Specifically, the multidimensional indicators include but are not limited to one or more of the following:

[0181] Impact characteristic index: You can choose one of the following: kurtosis, information entropy, Gini index, sparse measure, etc.

[0182] Energy index: The energy ratio of this mode component in all mode components σ ri =σ i / sum(σ i )*100%;

[0183] Harmonic noise index: correlation coefficient aut at the mth order multiple of the target fault characteristic frequency m

[0184]

[0185] 411. Determine whether the multi-dimensional judgment condition is met. If not, execute step 409; if yes, execute step 412.

[0186] Generally, for the characteristics of rotating machinery such as bearings and gears, common distributed or local faults are often accompanied by severe impact characteristics, or the impact is not obvious, but the frequency multiplication is significantly increased. Therefore, when the rotation-related frequency meets the frequency constraint range, if the multidimensional index exceeds the threshold, the equipment abnormality can be identified. At this time, the corresponding target domain signal can be called the fault domain signal. The impact index, harmonic noise index and energy index can assist in fault identification and assessment of fault severity.

[0187] For example, the kurtosis index that measures the impact of a signal can be expressed as Fig.10 As shown in Figure 2, the kurtosis value is 5.5835, the fault characteristic frequency is 25.0061, and the first, second, and third order harmonic noise indicators at the fault characteristic frequency are 0.4677, 0.4711, and 0.4364, respectively. The singular value σ i The proportion σ in the sum of all the mode components singular values r =8.51%.

[0188] When the multi-dimensional feature judgment condition is not met, the target domain signal does not contain the fault domain signal, and the vibration signal For a normal sample, the equipment is normal, and the sample normal code is stored in the sample digital file; when the multi-dimensional feature judgment logic is met, the vibration signal For fault data samples and equipment abnormalities, the multi-dimensional feature indicators and equipment abnormality codes are stored in the sample digital archive.

[0189] 412. Determine the actual characteristic frequency of the device.

[0190] After determining the equipment fault, the rotation-related frequency can be used as the actual fault characteristic frequency of the electromechanical transmission equipment, and further processing can be performed based on the actual fault characteristic frequency.

[0191] 413. Perform fault diagnosis.

[0192] After the actual fault characteristic frequency of the electromechanical transmission equipment is determined, a more detailed fault diagnosis can be performed based on the actual fault characteristic frequency.

[0193] For example, a mapping relationship between fault frequency and fault type may be preset, and after the actual fault characteristic frequency is determined, the fault type of the electromechanical transmission device may be determined based on the actual fault characteristic frequency and the mapping relationship.

[0194] For example, as mentioned above Fig.10 As shown in the figure, the actual fault characteristic frequency is 25.0061. Compared with the preset mapping relationship, this frequency is the output shaft speed in the electromechanical transmission equipment. The output shaft only has bearing components and there is obvious impact. It can be determined as an output shaft bearing fault. For another example, for comparison, the multi-dimensional indicators of the vibration signal collected when the electromechanical transmission equipment is working normally can be as follows Fig.11 As shown in the figure, although the rotation-related frequency appears, it does not meet the multi-dimensional characteristics and judgment logic. The signal does not have impact characteristics and there is no obvious frequency multiplication abnormality. It belongs to the harmonic component generated during the normal operation of the rotating machinery. Since the multi-dimensional harmonic index of the signal is ≥0.4 and the fault component energy accounts for ≥8%, it can be judged that Fig.10 The failure is relatively serious.

[0195] 414. Store sample digital files.

[0196] Generally, a sample digital archive can be established for electromechanical transmission equipment to store data collected or calculated during the operation status monitoring of the electromechanical transmission equipment. For example, the difference between the existing data and the historical data can be compared, and the difference can be used to indicate the deterioration trend of the equipment, so that it can be used to track the health status and service life.

[0197] The data generated in the aforementioned steps 401 to 413 can all be stored in the sample digital archive, so that when the state of the electromechanical transmission equipment is monitored, the characteristic frequency changes of the electromechanical transmission equipment during operation can be known. Therefore, it is equivalent to establishing a high-value information digital archiving scheme for data samples. Based on the high-precision index calculation of this application, it can simultaneously carry out abnormality identification, fault diagnosis and fault severity assessment, with comprehensive information storage and convenient key result backtracking.

[0198] In this application, the actual fault characteristic frequency in the equipment fault vibration signal can be identified without a speed sensor, and the multi-dimensional characteristics can be accurately calculated, and the abnormal identification and fault diagnosis of bearings and gearboxes can be carried out accordingly. The multi-dimensional indicators based on the calculation of the actual fault characteristic frequency, as well as the key result records such as the relevant harmonic noise spectrum, can accurately reflect the high-value information in the vibration signal, and the digital archives of the data samples can be established accordingly, which is of great value for sample tracing and later big data development.

[0199] It can be understood that the method provided by the present application provides a way to accurately extract the actual fault characteristic frequency based on the envelope-related harmonic noise spectrum, which can extract the actual fault characteristic frequency without relying on the speed sensor, with reliable accuracy and reduced testing costs. The parameter input is relatively simple, and it can carry out abnormality identification and fault diagnosis based on the approximate speed range, while ensuring accuracy, greatly reducing the application threshold of predictive maintenance. Therefore, the identification of the actual fault characteristic frequency in the equipment fault vibration signal and the accurate calculation of the multi-dimensional characteristics are realized in the absence of a speed sensor, and the abnormality identification and fault diagnosis of bearings and gearboxes are carried out accordingly. And the multi-dimensional indicators based on the calculation of the actual fault characteristic frequency, as well as the key result records such as the related harmonic noise spectrum, can accurately reflect the high-value information in the vibration signal, and establish a digital archive for the data sample based on this, which is of great value for sample backtracking and later big data development.

[0200] The above has provided a detailed introduction to the method flow provided by the present application. Based on the above method flow, the structure of the device provided by the present application for executing the above method steps is introduced below.

[0201] See also Fig.12 , a schematic diagram of the structure of a device abnormality diagnosis device provided by the present application, comprising:

[0202] A rotation related frequency calculation module 1201 is used to obtain the rotation related frequency of the electromechanical transmission device;

[0203] The constraint calculation module 1202 is used to obtain a frequency constraint range of the electromechanical transmission device, where the frequency constraint range is obtained based on the parameters of the electromechanical transmission device, and the frequency constraint range can be used to represent a possible range of rotation frequencies of the electromechanical transmission device when a fault occurs;

[0204] The diagnosis module 1203 is used to determine whether the electromechanical transmission device has a fault according to the matching result between the rotation-related frequency and the frequency constraint range.

[0205] In a possible implementation, the rotation-related frequency calculation module 1201 is specifically used to: obtain vibration information, where the vibration information includes information generated by the collected vibration of the electromechanical transmission device; and obtain the rotation-related frequency of the electromechanical transmission device according to the vibration information.

[0206] In a possible implementation, the rotation-related frequency calculation module 1201 is specifically used to: decompose the vibration information to obtain at least one component, wherein each component represents the vibration condition of one or more components in the electromechanical transmission device; and calculate the rotation-related frequency based on the at least one component.

[0207] In a possible implementation, the rotation-related frequency calculation module 1201 is specifically used to include: obtaining at least one of the envelope signal, envelope spectrum signal or envelope harmonic noise spectrum of each component; calculating a frequency set of each component based on at least one of the envelope signal, envelope spectrum signal or envelope harmonic noise spectrum of each component, the frequency set including at least one of the harmonic frequency, the correlation frequency or the modulation frequency, the correlation frequency being the frequency corresponding to the maximum value of the correlation coefficient in each component; obtaining the rotation-related frequency based on at least one of the frequency sets of each component.

[0208] In a possible implementation, the rotation-related frequency calculation module 1201 is specifically configured to: if the frequencies in the frequency set satisfy a preset relationship, use at least one of the harmonic frequency, the related frequency or the modulation frequency as the rotation-related frequency.

[0209] In a possible implementation manner, the rotation-related frequency calculation module 1201 is specifically used to: perform standardization processing on the vibration information to obtain a standardized signal; and decompose the standardized signal to obtain at least one component.

[0210] In one possible implementation, the diagnostic module 1203 is specifically used to: if the rotation-related frequency is within the frequency constraint range, determine whether the electromechanical transmission device has a fault based on an indicator of at least one dimension corresponding to the rotation-related frequency; if the rotation-related frequency is not within the frequency constraint range, determine that the electromechanical transmission device has not a fault based on the vibration information.

[0211] In a possible implementation, the diagnostic module 1203 is specifically used to: if the rotation-related frequency is within the frequency constraint range, obtain an indicator of at least one dimension corresponding to each component; if the indicator of at least one dimension meets the preset conditions, determine that the electromechanical transmission equipment has a fault.

[0212] In one possible implementation, the index of at least one dimension includes one or more of the following: an impact index, an energy index or a harmonic noise index, the impact index includes at least one of kurtosis, information entropy, sparsity measure or Gini index, the energy index represents the proportion of the component corresponding to the rotation-related frequency, and the harmonic noise index is determined based on the rotation-related frequency and its multiples.

[0213] In a possible implementation manner, the diagnosis module 1203 is further configured to determine the fault type of the fault according to the rotation-related frequency and a preset mapping relationship.

[0214] In a possible embodiment, the above-mentioned device also includes an archiving module 1204, which is used to: obtain first information obtained from data calculated in the rotation-related frequency step of the electromechanical vibration equipment, and second information obtained from data calculated in the step of determining whether the electromechanical transmission equipment has a fault based on the matching result of the rotation-related frequency and the frequency constraint range; save the first information and the second information in the archived data, and the data saved in the archived data is used to monitor the status of the electromechanical transmission equipment.

[0215] In a possible implementation, the parameters of the electromechanical transmission equipment include one or more of the following: bearing model, gearbox model, number of teeth of each gear stage in the gearbox, fault characteristic order of the bearing inner ring, outer ring, cage or rolling element.

[0216] In a possible implementation manner, a vibration sensor is provided in the electromechanical transmission device, and the acquisition module 1201 is specifically used to obtain vibration information through information collected by the vibration sensor.

[0217] See also Fig.13 , a structural schematic diagram of another device abnormality diagnosis device provided by the present application is described as follows.

[0218] The device for diagnosing abnormality of equipment may include a processor 1301 and a memory 1302. The processor 1301 and the memory 1302 are interconnected via a line. The memory 1302 stores program instructions and data.

[0219] The memory 1302 stores the aforementioned Figure 2-Figure 11 The program instructions and data corresponding to the steps in .

[0220] Processor 1301 is used to execute the above Figure 2-Figure 11 The method steps are performed by the device abnormality diagnosis device shown in any embodiment.

[0221] Optionally, the device abnormality diagnosis apparatus may further include a transceiver 1303 for receiving or sending data.

[0222] In a possible implementation, the device for diagnosing equipment abnormality may also include a vibration sensor ( Fig.13 The vibration sensor may include a DAS or an inductive vibration sensor.

[0223] The present application also provides a computer-readable storage medium in which a program is stored. When the program is run on a computer, the computer executes the above-mentioned Figure 2-Figure 11 The illustrated embodiments describe steps in a method.

[0224] Optionally, the aforementioned Fig.13 The device abnormality diagnosis device shown in the figure is a chip

[0225] The embodiment of the present application also provides a device for diagnosing abnormalities of an equipment. The device for diagnosing abnormalities of an equipment can also be called a digital processing chip or chip. The chip includes a processing unit and a communication interface. The processing unit obtains program instructions through the communication interface. The program instructions are executed by the processing unit. The processing unit is used to execute the aforementioned Figure 2-Figure 11 method steps.

[0226] The embodiment of the present application also provides a digital processing chip. The digital processing chip integrates a circuit and one or more interfaces for implementing the above-mentioned processor 1301 or the function of the processor 1301. When the digital processing chip integrates a memory, the digital processing chip can complete the method steps of any one or more embodiments in the above-mentioned embodiments. When the digital processing chip does not integrate a memory, it can be connected to an external memory through a communication interface. The digital processing chip implements the device abnormality diagnosis device or the actions performed by the device abnormality diagnosis device in the above-mentioned embodiment according to the program code stored in the external memory.

[0227] The present application also provides a computer program product which, when executed on a computer, enables the computer to execute the above Figure 2-Figure 11 The illustrated embodiment describes the steps of the method.

[0228] The device abnormality diagnosis device or device abnormality diagnosis device provided in the embodiment of the present application may be a chip, and the chip includes: a processing unit and a communication unit, the processing unit may be, for example, a processor, and the communication unit may be, for example, an input / output interface, a pin or a circuit. The processing unit may execute the computer execution instructions stored in the storage unit to enable the chip in the server to execute the above Figure 2-Figure 11Optionally, the storage unit is a storage unit in the chip, such as a register, a cache, etc. The storage unit may also be a storage unit located outside the chip in the wireless access device, such as a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM), etc.

[0229] Specifically, the aforementioned processing unit or processor may be a central processing unit (CPU), a neural-network processing unit (NPU), a graphics processing unit (GPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0230] The processor mentioned in any of the above can be a general-purpose central processing unit, a microprocessor, an ASIC, or one or more processors for controlling the above Figure 2-Figure 11 The method is performed by a program of an integrated circuit.

[0231] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0232] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0233] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0234] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)), etc.

[0235] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

Claims

1. A method for diagnosing equipment abnormality, characterized in that: include: Acquire a rotation-related frequency of the electromechanical transmission device, wherein the rotation-related frequency includes a frequency generated by the electromechanical transmission device when the electromechanical transmission device rotates; Acquire a frequency constraint range of the electromechanical transmission device, wherein the frequency constraint range is obtained according to parameters of the electromechanical transmission device, and the frequency range includes a range of rotation frequencies when a fault occurs in the electromechanical transmission device; Whether the electromechanical transmission device fails is determined according to a matching result between the rotation-related frequency and the frequency constraint range.

2. The method according to claim 1, characterized in that The step of obtaining the rotation-related frequency of the electromechanical transmission device comprises: Acquiring vibration information, wherein the vibration information includes information generated by the collected vibration of the electromechanical transmission equipment; The rotation-related frequency of the electromechanical transmission device is acquired according to the vibration information, where the rotation-related frequency includes the frequency generated when the electromechanical transmission device rotates.

3. The method according to claim 2, characterized in that The step of obtaining the rotation-related frequency of the electromechanical vibration device according to the vibration information comprises: Decomposing the vibration information to obtain at least one component, wherein each component represents the vibration condition of one or more components in the electromechanical transmission device; The rotation-dependent frequency is calculated based on the at least one component.

4. The method according to claim 3, characterized in that The calculating the rotation-related frequency according to the at least one component comprises: Obtain at least one of an envelope signal, an envelope spectrum signal or an envelope harmonic noise spectrum of each component; According to at least one of the envelope signal, the envelope spectrum signal or the envelope harmonic noise spectrum of each component, a frequency set of each component is calculated, wherein the frequency set includes at least one of a harmonic frequency, a correlation frequency or a modulation frequency, and the correlation frequency is a frequency corresponding to a maximum value of a correlation coefficient in each component; The rotation-related frequency is obtained according to at least one of the frequency sets of each component.

5. The method according to claim 4, characterized in that The obtaining the rotation frequency according to at least one of the frequency sets of each component comprises: If the frequencies in the frequency set satisfy a preset relationship, at least one of the harmonic frequency, the related frequency or the modulation frequency is used as the rotation-related frequency.

6. The method according to any one of claims 3 to 5, characterized in that: Decomposing the vibration signal to obtain at least one component includes: Performing standardization processing on the vibration information to obtain a standardized signal; The standardized signal is decomposed to obtain the at least one component.

7. The method according to any one of claims 1 to 6, characterized in that The determining whether the electromechanical transmission device has a fault according to the matching result of the rotation-related frequency and the frequency constraint range includes: If the rotation-related frequency is within the frequency constraint range, determining whether the electromechanical transmission device has a fault according to an indicator of at least one dimension corresponding to the rotation-related frequency; If the rotation-related frequency is not within the frequency constraint range, it is determined that no fault occurs in the electromechanical transmission device based on the vibration information.

8. The method according to claim 7, characterized in that If the rotation-related frequency is within the frequency constraint range, determining that the electromechanical transmission device has a fault comprises: If the rotation-related frequency is within the frequency constraint range, obtaining an index of at least one dimension corresponding to the rotation-related frequency; If the index of the at least one dimension meets the preset condition, it is determined that the electromechanical transmission equipment fails.

9. The method according to claim 8, characterized in that The indicators of at least one dimension include one or more of the following: an impact indicator, an energy indicator or a harmonic noise indicator. The impact indicator includes at least one of kurtosis, information entropy, sparsity measurement or Gini index. The energy indicator represents the proportion of the component corresponding to the rotation-related frequency. The harmonic noise indicator is determined based on the multiple of the rotation-related frequency and the target frequency.

10. The method according to any one of claims 1 to 9, characterized in that If it is determined that the electromechanical transmission device fails, the method further includes: The fault type of the fault is determined according to the rotation-related frequency and the preset mapping relationship.

11. The method according to any one of claims 1 to 10, characterized in that The method further comprises: Acquire first information and second information, wherein the first information is obtained based on the data calculated in the step of acquiring the rotation-related frequency of the electromechanical vibration device, and the second information is obtained based on the data calculated in the step of determining whether the electromechanical transmission device has a fault based on a matching result of the rotation-related frequency and the frequency constraint range; The first information and the second information are saved in archived data, and the data saved in the archived data is used to monitor the status of the electromechanical transmission equipment.

12. The method according to any one of claims 1 to 11, characterized in that If the electromechanical transmission device includes a shaft system component, the rotation-related frequency includes the rotation frequency of the shaft system component; if the electromechanical transmission device includes a bearing, the rotation-related frequency includes the fault characteristic frequency of the bearing.

13. The method according to any one of claims 1 to 12, characterized in that The parameters of the electromechanical transmission equipment include one or more of the following: Bearing model, gearbox model, number of teeth on each gear in the gearbox, fault characteristic order of the bearing inner ring, outer ring, cage or rolling element.

14. The method according to any one of claims 1 to 13, characterized in that The electromechanical transmission device is provided with a vibration sensor, and the obtaining of vibration information includes: The vibration information is obtained through information collected by the vibration sensor.

15. A device for diagnosing abnormality of equipment, characterized in that: include: A rotation-related frequency calculation module, used to obtain the rotation-related frequency of the electromechanical transmission device, wherein the rotation-related frequency includes the frequency generated by the electromechanical transmission device when rotating; A constraint calculation module, used for obtaining a frequency constraint range of the electromechanical transmission device, wherein the frequency constraint range is obtained according to parameters of the electromechanical transmission device, and the frequency range includes a range of rotational frequencies when a fault occurs in the electromechanical transmission device; The diagnosis module is used to determine whether the electromechanical transmission device has a fault according to a matching result between the rotation-related frequency and the frequency constraint range.

16. A device for diagnosing equipment abnormality, characterized in that: The communication processing device includes: a processor, the processor is coupled to a memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory so that the equipment abnormality diagnosis device performs the method according to any one of claims 1 to 14.

17. The device according to claim 16, characterized in that The equipment abnormality diagnosis device also includes a vibration sensor, which is used to collect information generated by vibration of mechanical equipment.

18. A computer program product comprising instructions, characterized in that When the method is executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 14.

19. A computer-readable storage medium, characterized in that: The method comprises instructions which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 14.

Images