Fault identification method of shield tunneling machine, electronic equipment, storage medium and program product

Abstract

The embodiment of the invention provides a fault identification method of a shield tunneling machine, electronic equipment, a storage medium and a program product. The method comprises the steps that a vibration signal of a bearing of the shield tunneling machine is obtained, iterative calculation is conducted on the vibration signal through a preset mantis shrimp algorithm, a signal decomposition parameter corresponding to the vibration signal is obtained, and the preset mantis shrimp algorithm at least comprises the parameter range of the preset signal decomposition parameter; decomposing the vibration signal according to a modal decomposition number and a filter length carried in the signal decomposition parameter to obtain a target modal component containing fault information; and determining an envelope spectrum corresponding to the target component, and determining the fault type of the shield tunneling machine according to the fault frequency in the envelope spectrum. The method is used for achieving the effect of determining the fault type of the shield tunneling machine.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of shield machines, and particularly relates to a fault identification method for a shield machine, an electronic device, a storage medium and a program product. BACKGROUND

[0002] In nuclear power engineering, a shield machine is used as an efficient and accurate underground engineering equipment, and is mainly used for water tunnel excavation and underground pipeline corridor construction. Bearings, as core rotating components, directly affect the safety and reliability of the shield machine. Once a fault occurs, it may cause equipment operation interruption and increase maintenance costs, and even cause safety hazards. Rolling bearings are widely used in various rotating machines in the fields of aerospace, nuclear power, ships, petrochemical industry, etc. Early fault diagnosis is of great significance to improve the safety and reliability of mechanical equipment.

[0003] The bearing vibration signal has nonlinear and non-stationary characteristics, and is affected by noise such as gear meshing and motor interference, so the fault characteristics are severely covered, and the extraction is extremely difficult. Existing fault diagnosis methods mainly include fixed frequency band division demodulation method and adaptive filtering demodulation method. Among them, the feature mode decomposition (FMD) is a new type of non-recursive decomposition theory. The signal is decomposed into different modal components by constructing an adaptive FIR filter through iterative updating of filter coefficients, which has the advantages of strong adaptability, high decomposition precision, etc.

[0004] However, the decomposition effect of FMD depends on the quality of the input parameters, and the selection of parameters has problems such as insufficient modal decomposition number leading to incomplete signal decomposition, excessive decomposition number introducing redundant components, and filter length being too short or too long.

[0005] In the prior art, the selection of modal decomposition number and filter length parameters in the FMD algorithm mainly depends on experience and artificial setting or traditional optimization algorithm optimization. These methods have strong subjectivity, slow convergence speed, and are easy to fall into local optimum, which cannot guarantee the global optimality of the parameters, resulting in low FMD decomposition precision and affecting the reliability of fault feature extraction.

[0006] There is an urgent need for a technical solution that can determine the fault type of a shield machine. SUMMARY

[0007] The fault identification method for a shield machine, the electronic device, the storage medium and the program product provided by the embodiments of the present application can determine the fault type of the shield machine.

[0008] In a first aspect, the embodiments of the present application provide a fault identification method of a shield tunneling machine, comprising: obtaining a vibration signal of a bearing of the shield tunneling machine, performing iterative calculation on the vibration signal by a horseshoe crab algorithm carrying preset parameters to obtain a target signal decomposition parameter corresponding to the vibration signal, wherein the preset parameters at least include a preset signal decomposition parameter interval; decomposing the vibration signal according to a modal decomposition number and a filter length carried in the target signal decomposition parameter to obtain a target modal component containing fault information; determining a corresponding envelope spectrum of the target component, and determining a fault type of the shield tunneling machine according to a fault frequency in the envelope spectrum.

[0009] In a possible implementation, an initial population set is generated according to the signal decomposition parameter interval, wherein the initial population set at least includes a plurality of initial signal decomposition parameters; the vibration signal is iteratively calculated by a dynamic optimization strategy corresponding to the initial population set to obtain the target signal decomposition parameter corresponding to the vibration signal.

[0010] In a possible implementation, a fitness value corresponding to decomposition of the vibration signal by each initial signal decomposition parameter in the initial population set is calculated to obtain a first fitness value set; a first fitness value is determined from the fitness set according to a preset rule, the dynamic optimization strategy is updated according to the first fitness value to obtain an upgraded optimization strategy; in a case where an upgrade number of the upgraded optimization strategy is greater than or equal to a preset optimization value, a target population set calculated by the final optimization strategy is determined; a second fitness value corresponding to decomposition of the vibration signal by each first signal decomposition parameter in the target population set is calculated to obtain a second fitness value set; a target second fitness value meeting the preset rule is determined from the second fitness value set, and a first signal decomposition parameter corresponding to the target second fitness value is determined as the target signal decomposition parameter.

[0011] In a possible implementation, the vibration signal is decomposed according to the modal decomposition number and the filter length to obtain a plurality of modal components; an envelope spectrum corresponding to each modal component in the plurality of modal components is determined, and a target modal component is determined from the plurality of modal components according to a peak value feature carried in the envelope spectrum.

[0012] In a possible implementation, a frequency value corresponding to a spectral line peak in the envelope spectrum is determined, the frequency value is matched with a preset fault frequency table to obtain a matching result, and a fault type of the shield tunneling machine is determined according to a fault frequency indicated by the matching result.

[0013] In one possible implementation, the vibration signal is iteratively searched according to the search strategy of the mantis shrimp algorithm to obtain the target signal decomposition parameters, wherein the initial search strategy is determined based on a preset polarization type index corresponding to the vibration signal; the search strategy includes at least one of the following: foraging strategy, attack strategy, and defense strategy; the operating parameters of the foraging strategy include:

[0014]

[0015] in The new values ​​of the modal decomposition number and filter length of the vibration signal in the (t+1)th iteration; This represents the target value found by the modality decomposition number and filter length; From the current value Compared with the target value The definition of the difference; express With the current vector The difference, where and ; The random diffusion coefficient is in the range [-1, 1].

[0016] The operating parameters of the attack strategy include: ,in, It is randomly generated. ;

[0017] The operating parameters of the defense strategy include: ; Where K is a scale factor randomly generated between 0 and 0.3;

[0018] Based on the search strategy corresponding to different polarization type indices, update the individual position vector composed of M and L in the mantis shrimp algorithm population, including:

[0019] Calculate the left polarization angle and the right polarization angle:

[0020] In each iteration, the left eye uses the current population. With the renewal of population To calculate the left polarization angle LPA:

[0021] ;

[0022] The right eye directly generates the right polarization angle RPA randomly according to the formula:

[0023] ;

[0024] The difference between the left polarization angle and the right polarization angle is compared. If the left polarization angle is smaller than the right polarization angle, the polarization type index value corresponding to the left polarization angle is determined as a new polarization index. If the right polarization angle is smaller than the left polarization angle, the polarization type index value corresponding to the right polarization angle is determined as a new polarization index. The polarization index is determined from a preset index library, which includes: multiple left polarization angles, multiple polarization indexes, and a mapping relationship between the multiple left polarization angles and the multiple polarization indexes; multiple right polarization angles, multiple polarization indexes, and a mapping relationship between the multiple right polarization angles and the multiple polarization indexes.

[0025] Secondly, embodiments of this application provide a fault identification device for a tunnel boring machine (TBM), comprising: an acquisition module, configured to acquire vibration signals of the bearings of the TBM, and iteratively calculate the vibration signals using a preset algorithm to obtain signal decomposition parameters corresponding to the vibration signals, wherein the preset algorithm includes at least a preset parameter range for the signal decomposition parameters; a decomposition module, configured to decompose the vibration signals according to the modal decomposition number and filter length carried in the signal decomposition parameters to obtain target modal components containing fault information; and a determination module, configured to determine the envelope spectrum corresponding to the target components, and determine the fault type of the TBM based on the fault frequency in the envelope spectrum.

[0026] Thirdly, embodiments of this application provide a fault identification device for a tunnel boring machine, including: a memory and a processor;

[0027] The memory stores computer-executed instructions;

[0028] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0029] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0030] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0031] The tunnel boring machine (TBM) fault identification method, electronic equipment, storage medium, and program product provided in this application first acquire the vibration signal of the TBM bearing. Since the impact characteristics of bearing faults are often masked by noise, optimal parameters are needed to accurately decompose the signal. The system uses a mantis egg case algorithm to iteratively search within a preset parameter range. This algorithm uses indicators such as "envelope entropy," which measure the clarity of impact characteristics, as optimization targets to automatically find the most suitable target signal decomposition parameters for the vibration signal. Then, the system uses the optimal parameter combination obtained above to accurately decompose the original vibration signal. This decomposition effectively separates noise from useful components, thereby accurately extracting the target modal components of the fault impact information. Finally, envelope spectrum analysis is performed on the target component. By comparing the prominent peak frequencies appearing in the envelope spectrum with the theoretical fault characteristic frequencies of various bearing components, the specific fault type of the TBM can be finally determined, achieving the effect of determining the fault type of the TBM. Attached Figure Description

[0032] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0033] Figure 1 The flowchart for fault identification of tunnel boring machines provided in this application Figure 1 ;

[0034] Figure 2 This is a schematic diagram of the mantis shrimp algorithm provided in this application;

[0035] Figure 3 The flowchart for fault identification of tunnel boring machines provided in this application Figure 2 ;

[0036] Figure 4 A schematic diagram of the fault identification device for the tunnel boring machine provided in this application;

[0037] Figure 5 A schematic diagram of the fault identification device for the tunnel boring machine provided in this application.

[0038] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0039] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0040] Figure 1 The flowchart for fault identification of tunnel boring machines provided in this application Figure 1 ,like Figure 1 As shown, the method includes:

[0041] S201. Obtain the vibration signal of the bearing of the tunnel boring machine, and perform iterative calculation on the vibration signal using the Shrimp Algorithm with preset parameters to obtain the target signal decomposition parameters corresponding to the vibration signal.

[0042] Among them, the preset parameters include at least a preset range of signal decomposition parameters.

[0043] For example, given the number of mode decompositions and the filter length, and a general search range (preset parameter interval), the mantis shrimp algorithm generates and tests different parameter combinations generation after generation within this interval, much like a mantis shrimp hunting. For each set of parameters, it is used to decompose the vibration signal and calculate a fitness value (such as envelope entropy). The smaller the envelope entropy, the clearer the impact characteristics of the decomposed components, and the better the parameter combination. The algorithm will eventually automatically find the "target signal decomposition parameters" that result in the optimal fitness value (such as the minimum envelope entropy).

[0044] This addresses the core pain point in signal decomposition algorithms: "parameters are dependent on experience and difficult to set."

[0045] S202. Decompose the vibration signal according to the modal decomposition number and filter length carried in the target signal decomposition parameters to obtain the target modal components containing fault information.

[0046] The optimal "modal decomposition number" and "filter length" obtained from S201 are used to decompose the input signal using an algorithm (such as VMD). After the algorithm runs, it outputs a series of modal components. By analyzing these components (such as calculating the envelope entropy or correlation coefficient of each component), the target modal component containing fault information is selected. This step ensures that subsequent analysis is based on the purest fault characteristics.

[0047] S203. Determine the envelope spectrum corresponding to the target component, and determine the fault type of the tunnel boring machine based on the fault frequency in the envelope spectrum.

[0048] Hilbert envelope demodulation is performed on the target modal components obtained from S202, and their envelope spectra are calculated. Obvious spectral peaks are identified on the envelope spectrum. These peak frequencies are then precisely compared with theoretical fault characteristic frequencies (such as inner ring, outer ring, and rolling element fault frequencies) calculated based on the bearing type and rotational speed. Once a match is found, the specific fault type of the tunnel boring machine bearing can be determined.

[0049] The tunnel boring machine (TBM) fault identification method provided in this application first acquires the vibration signal of the TBM bearing. Since the impact characteristics of bearing faults are often masked by noise, optimal parameters are needed to accurately decompose the signal. The system uses a mantis egg case algorithm to iteratively search within a preset parameter range. This algorithm uses indicators such as "envelope entropy," which measure the clarity of impact characteristics, as optimization targets to automatically find the most suitable target signal decomposition parameters for the vibration signal. Then, the system uses the optimal parameter combination obtained above to accurately decompose the original vibration signal. This decomposition effectively separates noise from useful components, thereby accurately extracting the target modal components of the fault impact information. Finally, envelope spectrum analysis is performed on the target component. By comparing the prominent peak frequencies appearing in the envelope spectrum with the theoretical fault characteristic frequencies of various bearing components, the specific fault type of the TBM can be finally determined, achieving the effect of determining the fault type of the TBM.

[0050] In an exemplary embodiment, an initial population set is generated based on the signal decomposition parameter range, wherein the initial population set includes at least a plurality of initial signal decomposition parameters; the vibration signal is iteratively calculated using a dynamic optimization strategy corresponding to the initial population set to obtain the target signal decomposition parameters corresponding to the vibration signal.

[0051] Optionally, the signal decomposition parameter range is a manually defined search range for parameters. For mode decomposition algorithms (such as FMD), key parameters include the number of mode decompositions (M) and the filter length (L). Ensure the search is performed within a reasonable range to avoid meaningless calculations.

[0052] In optimization algorithms, a "population" refers to a set of potential solutions. Here, each "individual" is a set of signal decomposition parameters (such as an (M, L) pair). The initial population set is a set of parameters randomly generated or generated according to certain rules based on the aforementioned parameter range at the beginning of the first generation of the algorithm.

[0053] Dynamic optimization strategies adaptively adjust the step size and direction of evolution based on the current distribution and quality of the population. For example, in the early stages of the search, large steps are taken to explore the entire population, while in the later stages, smaller steps are taken to refine the local search.

[0054] In an exemplary embodiment, the fitness value corresponding to the decomposition of the vibration signal by each initial signal decomposition parameter in the initial population set is calculated to obtain a first fitness value set; the first fitness value is determined from the fitness set according to a preset rule, and the dynamic optimization strategy is updated based on the first fitness value to obtain an upgrade optimization strategy; if the upgrade number of the upgrade optimization strategy is greater than or equal to a preset optimization value, the target population set calculated by the final optimization strategy is determined; the second fitness value corresponding to the decomposition of the vibration signal by each first signal decomposition parameter in the target population set is calculated to obtain a second fitness value set; the target second fitness value that conforms to the preset rule is determined from the second fitness value set, and the first signal decomposition parameter corresponding to the target second fitness value is determined as the target signal decomposition parameter.

[0055] For each set of parameters in the initial population, signal decomposition is performed, and its fitness value (such as envelope entropy) is calculated. All these values ​​constitute the "first fitness value set," which reflects the overall performance distribution of the initial parameter set. The "first fitness value" refers to a key benchmark value selected from the first set according to preset rules (such as maximum, minimum, or average value).

[0056] Based on this baseline value, the dynamic optimization strategy is updated for the first time, forming an "upgraded optimization strategy". For example, the algorithm adjusts the subsequent search step size based on the performance of the initial population, making the search behavior more targeted.

[0057] The aforementioned "strategy upgrade" process will be repeated multiple times (upgrade count ≥ preset optimization value). Each upgrade makes the optimization strategy "smarter" and better adapted to the characteristics of the current vibration signal. For each parameter in the "target population set" (called the first signal decomposition parameter), signal decomposition and fitness calculation are performed again to obtain the "second fitness value set". This calculation is the final performance confirmation.

[0058] Finally, from the set of second fitness values, the optimal "target second fitness value" is selected again according to preset rules (such as finding the minimum value of the minimum envelope entropy). The parameter with this optimal fitness value is finally determined as the "target signal decomposition parameter".

[0059] In an optional embodiment:

[0060] Randomly generate an initial population of size n and dimension m:

[0061] (1)

[0062] in, It is the population individual position vector composed of the parameters M and L to be optimized, i.e. , =1,2,3,...,n, representing the position coordinates of the i-th individual in the j-th dimension. ; Let i be the number of decomposition modes represented by the i-th mantis shrimp individual. , The maximum number of modal components preset for step 2; Let be the filter length represented by the i-th mantis shrimp individual; and Let be the upper and lower bounds of the j-th dimension; A random number that follows a uniform distribution in the range [0,1].

[0063] An initial polarization type index (PTI) value is defined for each individual in the initial population according to the following formula:

[0064] (2)

[0065] in, The function is used for rounding. Different polarization types are represented by the PTI value: PTI=1 represents vertical linear polarization; PTI=2 represents horizontal linear polarization; and PTI=3 represents circular polarization. Different polarization types will subsequently be assigned different search strategies.

[0066] The search strategy is determined based on the current polarization type index PTIi, and the individual positions in the population are updated.

[0067] 1) PTI=1 → Foraging strategy:

[0068] The random motion of mantis shrimp searching for food can be described as Brownian motion, and its position update can be described as follows:

[0069] (3)

[0070] in This represents the individual's new position in the (t+1)th iteration; This represents the optimal location for individual discovery; From current location With the best position The definition of the difference; express With the current vector The difference, where and ; It is the random diffusion coefficient in the range [-1, 1].

[0071] This strategy reflects the algorithm's initial exploration of solutions.

[0072] 2) PTI=2 → Attack Strategy:

[0073] Mantis shrimp's aggressive behavior is generally manifested in the swinging motion of its forepaws, which can be viewed as circular motion in a two-dimensional plane, and thus represented as:

[0074] (4)

[0075] in, It is randomly generated. By controlling the attack angle with θ, the individual attacks along the direction of the optimal solution in the circumferential direction. This strategy symbolizes the algorithm's outward search to escape local optima and avoid getting trapped in local optima.

[0076] 3) PTI=3 → Defense / Evasion Strategy:

[0077] When circularly polarized signals dominate, individuals choose to "approach" or "retreat" based on the level of threat:

[0078] When choosing defense:

[0079] (5)

[0080] When choosing to evade.

[0081] (6)

[0082] in, The scaling factor is randomly generated between 0 and 0.3. This strategy improves the algorithm's accuracy by making small scaling adjustments near the current optimal position.

[0083] 1) Calculate the left and right polarization angles.

[0084] The calculation of the polarization angle is inspired by the visual system of mantis shrimps, where the left and right eyes perform independent perception. In each iteration, the left eye uses the current population... With the renewal of population To calculate the left polarization angle LPA:

[0085] (7)

[0086] The right eye directly generates the right polarization angle RPA randomly according to formula (4):

[0087] (8)

[0088] 2) Determine the polarization type of the left and right eyes based on the polarization angles of the left and right eyes obtained in 1).

[0089] Left-polarization type LPT and right-polarization type RPT are determined according to equation (5):

[0090] (9)

[0091] Calculate the left-eye polarization angle difference (LAD) between the left polarization angle (LPA) and the polarization reference angle, including the reference angle. Similarly, the right-eye polarization angle difference (RAD) is calculated using the right polarization angle (RPA).

[0092] (10)

[0093] 4) Compare the magnitude of the polarization angle difference between the left and right eyes, and update the polarization type index value. .

[0094] (11)

[0095] That is, if Smaller, then retain As a new Otherwise, keep .

[0096] Step 1: Preprocess the vibration signals collected by the vibration sensors installed on the shield machine bearings to obtain the preprocessed signals.

[0097] Step 101: Use a bandpass filter to filter the noise-reduced signal and extract the specific frequency band signal components of 20-500Hz.

[0098] Step 102: Normalize the filtered signal to map the signal amplitude between 0 and 1, thus obtaining a normalized signal.

[0099] Step 2: Initialize the population, polarization type index PTI vector, and maximum number of iterations for the mantis shrimp optimization algorithm, and set the range of the mode decomposition number M and filter length L for the FMD algorithm.

[0100] Step 201: Initialize the population size N=50 and the identity matrix of dimension 50 for the mantis shrimp optimization algorithm;

[0101] Step 202: Convert the polarization type index vector Initialize to ;

[0102] Step 203: Set the maximum number of iterations to 150;

[0103] Step 204: Initialize other parameters required for the FMD algorithm, including the minimum value of the mode decomposition number M. =3 and maximum value =10, the minimum value of filter length L =1500 and the maximum value =2500.

[0104] Step 3: Based on the search strategy corresponding to different PTI values, update the individual position vector composed of M and L in the population, and obtain the modal components after FMD algorithm decomposition under the current parameters.

[0105] Step 301: Based on the current PTI vector Determine the corresponding search strategy;

[0106] Step 302: If =1, update using a foraging search strategy. ;

[0107] Step 303: If =2, update using an attack search strategy. ;

[0108] Step 304: If =3, update using a defensive / evasive search strategy. ;

[0109] Step 4: Update the PTI vector based on the current population and the updated population, such as... Figure 2 As shown, Figure 2 This is a schematic diagram of the mantis shrimp algorithm provided in this application.

[0110] Step 401: Calculate the left polarization angle of each individual in the current population. and right polarization angle ;

[0111] Step 402: Determine the left polarization type based on the left and right polarization angles. and right polarization type ;

[0112] Step 403: Calculate the left polarization angle difference and right polarization angle difference ;

[0113] Step 404: Compare the magnitude of the difference between the left and right polarization angles and update the polarization type index value. .

[0114] Step 5: Calculate the fitness and retain the corresponding optimal solution. Determine if the maximum number of iterations has been reached: if it is, use the current solution as the optimal parameter combination; if it is not, return to step 3 to continue the iterative search.

[0115] Step 6: Use the FMD algorithm to obtain the modal components containing fault information, calculate their envelope spectrum, extract fault feature frequencies, and realize bearing fault identification.

[0116] Step 601: Perform envelope spectrum analysis on the modal components and calculate the absolute value of the envelope spectrum of each modal component.

[0117] Step 602: Extract the peak position frequency of the envelope spectrum , as a fault characteristic frequency.

[0118] In an exemplary embodiment, the frequency value corresponding to the peak value of the spectral line in the envelope spectrum is determined, and the frequency value is matched with a preset fault frequency table to obtain a matching result; the fault type of the tunnel boring machine is determined according to the fault frequency indicated by the matching result.

[0119] The system will automatically scan the calculated envelope spectrum, identify all spectral peaks that are significantly higher than the background noise, and accurately record their corresponding frequency values.

[0120] When bearings or gears suffer localized damage (such as pitting or spalling), periodic impacts are generated during operation. The repetition frequency of these impacts is the fault characteristic frequency of the component. The function of the envelope spectrum is to demodulate these periodic impacts and display them as peak values ​​at the corresponding frequency points.

[0121] The one or more peak frequencies extracted in the first step are compared with the theoretical frequencies in the "preset fault frequency table". Due to actual speed fluctuations and calculation errors, they do not need to be exactly equal. As long as the measured peak frequency is very close to a certain theoretical fault frequency (for example, the error is within 2%), it can be considered a successful match.

[0122] If a significant peak frequency successfully matches the "bearing outer ring failure frequency" in the failure frequency table.

[0123] Diagnosis: The outer ring of the main drive bearing of the tunnel boring machine is damaged.

[0124] If the peak frequency matches the "bearing inner ring failure frequency".

[0125] Diagnosis: Damage exists in the inner ring of the bearing.

[0126] If no peak value is found in the envelope spectrum that significantly matches the fault frequency table.

[0127] Diagnostic results: The monitored bearings showed no obvious local faults in their current condition.

[0128] In one exemplary embodiment, the failure frequency table includes at least one of the following: outer ring failure frequency, inner ring failure frequency, and rolling element failure frequency.

[0129] Figure 3 The flowchart for fault identification of the tunnel boring machine provided in this application is as follows. Figure 1 Flowchart Figure 2,like Figure 3 As shown:

[0130] Start: Initiate the optimization process.

[0131] Initialization parameters: Set the parameters of the mantis shrimp algorithm itself, and define the search range of the parameters that the FMD algorithm needs to optimize (such as the number of modes and the penalty factor).

[0132] Adaptive search: Based on an indicator called "PTI" (which may represent population diversity, evolutionary stage, or individual performance), a search strategy is dynamically selected and the location is updated for each individual.

[0133] Update PTI vector: Based on the information of the current population and the updated population, refresh the PTI vector to guide the strategy selection in the next round.

[0134] Evaluation and selection: Calculate the fitness of all new positions and retain the historical best solution.

[0135] Termination Check: Check if the maximum number of iterations has been reached. If not, return to step 3 to continue optimization; if the maximum number of iterations has been reached, proceed to the next step.

[0136] Output the optimal solution: The process ends, and the optimal combination of FMD parameters found is output.

[0137] The core idea of ​​this flowchart is to achieve adaptive adjustment of the search strategy through the PTI mechanism, so as to find the optimal parameter configuration of the FMD algorithm more efficiently.

[0138] This application embodiment first acquires the vibration signal of the tunnel boring machine (TBM) bearing. Since the impact characteristics of bearing failures are often masked by noise, optimal parameters are needed to accurately decompose the signal. The system uses a mantis egg case algorithm to iteratively search within a preset parameter range. This algorithm uses indicators such as "envelope entropy," which measure the clarity of impact characteristics, as optimization targets, and automatically finds the most suitable target signal decomposition parameters for the vibration signal. Then, the system uses the obtained optimal parameter combination to accurately decompose the original vibration signal. This decomposition effectively separates noise from useful components, thereby accurately extracting the target modal components of the fault impact information. Finally, envelope spectrum analysis is performed on the target component. By comparing the prominent peak frequencies appearing in the envelope spectrum with the theoretical fault characteristic frequencies of various bearing components, the specific fault type of the TBM can be finally determined, achieving the effect of determining the fault type of the TBM.

[0139] Figure 4 A schematic diagram of the fault identification device for the tunnel boring machine provided in this application is shown below. Figure 4 As shown, the fault identification device 40 for the tunnel boring machine provided in this embodiment includes:

[0140] The acquisition module 401 is used to acquire the vibration signal of the bearing of the tunnel boring machine, and to perform iterative calculation on the vibration signal through a preset algorithm to obtain the signal decomposition parameters corresponding to the vibration signal. The preset algorithm includes at least the parameter range of the preset signal decomposition parameters.

[0141] The decomposition module 402 is used to decompose the vibration signal according to the mode decomposition number and filter length carried in the signal decomposition parameters to obtain the target mode component containing fault information.

[0142] The determination module 403 is used to determine the corresponding envelope spectrum of the target component and determine the fault type of the tunnel boring machine based on the fault frequency in the envelope spectrum.

[0143] In one possible implementation, the decomposition module 402 is used to generate an initial population set based on the signal decomposition parameter range, wherein the initial population set includes at least a plurality of initial signal decomposition parameters; and to iteratively calculate the vibration signal using a dynamic optimization strategy corresponding to the initial population set to obtain the target signal decomposition parameters corresponding to the vibration signal.

[0144] In one possible implementation, the decomposition module 402 is used to calculate the fitness value corresponding to the decomposition of the vibration signal by each initial signal decomposition parameter in the initial population set, to obtain a first fitness value set; determine the first fitness value from the fitness set according to a preset rule, update the dynamic optimization strategy based on the first fitness value, to obtain an upgrade optimization strategy; determine the target population set calculated by the final optimization strategy if the upgrade number of the upgrade optimization strategy is greater than or equal to a preset optimization value; calculate the second fitness value corresponding to the decomposition of the vibration signal by each first signal decomposition parameter in the target population set, to obtain a second fitness value set; determine the target second fitness value that conforms to the preset rule from the second fitness value set, and determine the first signal decomposition parameter corresponding to the target second fitness value as the target signal decomposition parameter.

[0145] In one possible implementation, the decomposition module 402 is used to decompose the vibration signal according to the modal decomposition number and the filter length to obtain multiple modal components; determine the envelope spectrum corresponding to each modal component among the multiple modal components; and determine the target modal component from the multiple modal components according to the peak characteristics carried by the envelope spectrum.

[0146] In one possible implementation, the determining module 403 is used to determine the frequency value corresponding to the peak value of the spectral line in the envelope spectrum, match the frequency value with a preset fault frequency table to obtain a matching result, and determine the fault type of the tunnel boring machine based on the fault frequency indicated by the matching result.

[0147] In one possible implementation, the determining module 403 is used to iteratively search the vibration signal according to the search strategy of the mantis shrimp algorithm to obtain the target signal decomposition parameters, wherein the initial search strategy of the search strategy is determined based on the preset polarization type index corresponding to the vibration signal.

[0148] The search strategy includes at least one of the following: foraging strategy, attack strategy, and defense strategy;

[0149] The operating parameters of the foraging strategy include:

[0150]

[0151] in The new values ​​of the modal decomposition number and filter length of the vibration signal in the (t+1)th iteration; This represents the target value found by the modality decomposition number and filter length; From the current value Compared with the target value The definition of the difference; express With the current vector The difference, where and ; The random diffusion coefficient is in the range [-1, 1].

[0152] The operating parameters of the attack strategy include: ,in, It is randomly generated. ;

[0153] The operating parameters of the defense strategy include: ; Where K is a scale factor randomly generated between 0 and 0.3;

[0154] Based on the search strategy corresponding to different polarization type indices, update the individual position vector composed of M and L in the mantis shrimp algorithm population, including:

[0155] Calculate the left polarization angle and the right polarization angle:

[0156] In each iteration, the left eye uses the current population. With the renewal of population To calculate the left polarization angle LPA:

[0157] ;

[0158] The right eye directly generates the right polarization angle RPA randomly according to the formula:

[0159] ;

[0160] Compare the difference between the left polarization angle and the right polarization angle. If the left polarization angle is smaller than the right polarization angle, then determine the polarization type index value corresponding to the left polarization angle as the new polarization index.

[0161] If the right polarization angle is smaller than the left polarization angle, then the polarization type index value corresponding to the right polarization angle is determined as a new polarization index. The polarization index is determined from a preset index library, which includes: multiple left polarization angles, multiple polarization indices, and a mapping relationship between the multiple left polarization angles and the multiple polarization indices; multiple right polarization angles, multiple polarization indices, and a mapping relationship between the multiple right polarization angles and the multiple polarization indices.

[0162] The fault identification device for tunnel boring machines provided in this embodiment can execute the method provided in the above-described method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0163] Figure 5 A schematic diagram of the fault identification device for the tunnel boring machine provided in this application. Figure 5 As shown, the electronic device 50 provided in this embodiment includes at least one processor 501 and a memory 502. Optionally, the device 50 further includes a communication component 503. The processor 501, memory 502, and communication component 503 are connected via a bus 504.

[0164] In a specific implementation, at least one processor 501 executes computer execution instructions stored in memory 502, causing at least one processor 501 to perform the above-described method.

[0165] The specific implementation process of processor 501 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0166] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0167] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0168] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0169] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0170] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0171] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0172] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0173] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0174] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0175] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0176] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0177] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0178] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A fault identification method for a tunnel boring machine, characterized in that, include: The vibration signal of the bearing of the tunnel boring machine is obtained, and the vibration signal is iteratively calculated by the Shrimp Algorithm with preset parameters to obtain the target signal decomposition parameters corresponding to the vibration signal. The preset parameters include at least a preset signal decomposition parameter range. The vibration signal is decomposed based on the modal decomposition number and filter length carried in the target signal decomposition parameters to obtain target modal components containing fault information. The envelope spectrum corresponding to the target component is determined, and the fault type of the tunnel boring machine is determined based on the fault frequency in the envelope spectrum.

2. The method according to claim 1, characterized in that, The vibration signal is iteratively calculated using a preset mantis shrimp algorithm to obtain the target signal decomposition parameters corresponding to the vibration signal, including: An initial population set is generated based on the signal decomposition parameter range, wherein the initial population set includes at least a plurality of initial signal decomposition parameters; The vibration signal is iteratively calculated using the dynamic optimization strategy corresponding to the initial population set to obtain the target signal decomposition parameters corresponding to the vibration signal.

3. The method according to claim 2, characterized in that, The vibration signal is iteratively calculated using a dynamic optimization strategy corresponding to the initial population set to obtain the target signal decomposition parameters corresponding to the vibration signal, including: Calculate the fitness value corresponding to the decomposition of the vibration signal by each initial signal decomposition parameter in the initial population set to obtain the first fitness value set; A first fitness value is determined from the fitness set according to a preset rule, and the dynamic optimization strategy is updated based on the first fitness value to obtain an upgrade optimization strategy; If the number of upgrades of the upgrade and optimization strategy is greater than or equal to the preset optimization value, the target population set calculated by the final optimization strategy is determined. Calculate the second fitness value corresponding to each first signal decomposition parameter in the target population set after decomposing the vibration signal, and obtain the second fitness value set; A target second fitness value that conforms to the preset rule is determined from the second fitness value set, and the first signal decomposition parameter corresponding to the target second fitness value is determined as the target signal decomposition parameter.

4. The method according to claim 1, characterized in that, The vibration signal is decomposed based on the modal decomposition number and filter length carried in the target signal decomposition parameters to obtain target modal components containing fault information, including: The vibration signal is decomposed according to the modal decomposition number and the filter length to obtain multiple modal components; The envelope spectrum corresponding to each of the plurality of modal components is determined, and the target modal component is determined from the plurality of modal components based on the first peak feature carried by the envelope spectrum.

5. The method according to claim 1, characterized in that, Determine the envelope spectrum corresponding to the target component, and determine the fault type of the tunnel boring machine based on the fault frequency in the envelope spectrum, including: Determine the frequency value corresponding to the peak value of the spectral line in the envelope spectrum, and match the frequency value with a preset fault frequency table to obtain the matching result; The fault type of the tunnel boring machine is determined based on the fault frequency indicated by the matching results.

6. The method according to claim 5, characterized in that, The fault frequency table includes at least one of the following: outer ring fault frequency, inner ring fault frequency, and rolling element fault frequency.

7. The method according to claim 1, characterized in that, The vibration signal is iteratively calculated using a mantis shrimp algorithm with preset parameters, including: The vibration signal is iteratively searched according to the search strategy of the mantis shrimp algorithm to obtain the target signal decomposition parameters. The initial search strategy is determined based on the preset polarization type index corresponding to the vibration signal. The search strategy includes at least one of the following: foraging strategy, attack strategy, and defense strategy; The operating parameters of the foraging strategy include: in The new values ​​of the modal decomposition number and filter length of the vibration signal in the (t+1)th iteration; This represents the target value found by the modality decomposition number and filter length; From the current value Compared with the target value The definition of the difference; express With the current vector The difference, where and ; The random diffusion coefficient is in the range [-1, 1]. The operating parameters of the attack strategy include: ,in, It is randomly generated. ; The operating parameters of the defense strategy include: ; Where K is a scale factor randomly generated between 0 and 0.3; Based on the search strategy corresponding to different polarization type indices, update the individual position vector composed of M and L in the mantis shrimp algorithm population, including: Calculate the left polarization angle and the right polarization angle: In each iteration, the left eye uses the current population. With the renewal of population To calculate the left polarization angle LPA: ； The right eye directly generates the right polarization angle RPA randomly according to the formula: ； Compare the difference between the left polarization angle and the right polarization angle. If the left polarization angle is smaller than the right polarization angle, then determine the polarization type index value corresponding to the left polarization angle as the new polarization index. If the right polarization angle is smaller than the left polarization angle, then the polarization type index value corresponding to the right polarization angle is determined as a new polarization index. The polarization index is determined from a preset index library, which includes: multiple left polarization angles, multiple polarization indices, and a mapping relationship between the multiple left polarization angles and the multiple polarization indices; multiple right polarization angles, multiple polarization indices, and a mapping relationship between the multiple right polarization angles and the multiple polarization indices.

8. A fault identification device for a tunnel boring machine, characterized in that, include: The acquisition module is used to acquire the vibration signal of the bearing of the tunnel boring machine, and to perform iterative calculation on the vibration signal through a preset mantis shrimp algorithm to obtain the signal decomposition parameters corresponding to the vibration signal. The preset mantis shrimp algorithm includes at least the preset parameter range of the signal decomposition parameters. The decomposition module is used to decompose the vibration signal according to the mode decomposition number and filter length carried in the signal decomposition parameters to obtain the target mode component containing fault information. The determination module is used to determine the envelope spectrum corresponding to the target component and determine the fault type of the tunnel boring machine based on the fault frequency in the envelope spectrum.

9. A fault identification device for a tunnel boring machine, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

Images