Shield Machine Abnormality Diagnosis Method and System Based on Maximum Information Non-parameters

Abstract

The present application provides a shield machine abnormal diagnosis method and system based on maximum information non-parameters. Among them, the method includes: during the tunneling process of the shield machine, acquiring the vibration signal of the cutter head bearing in the shield machine, and determining the spatial energy attenuation mode of the vibration waveform formed when the vibration signal propagates in the underground complex stratum; performing multi-frequency domain decomposition on the vibration signal to obtain a multi-band energy distribution set; using the maximum mutual information criterion to select the target abnormal band energy distribution from the multi-band energy distribution set, and the maximum mutual information criterion is determined by the maximum information non-parameters; generating a collaborative diagnosis feature vector according to the spatial energy attenuation mode and the target abnormal band energy distribution; inputting the collaborative diagnosis feature vector into a preset random forest classifier, and outputting a diagnosis result matching the abnormal tunneling state of the shield machine. The present application improves the accuracy and real-time performance of the abnormal state diagnosis of the shield machine in complex strata.

Description

Technical Field

[0001] This application relates to the technical field of maximum information non - parameters, and particularly to a shield machine anomaly diagnosis method and system based on maximum information non - parameters. Background Art

[0002] During the tunneling process of a shield machine in complex strata, abnormal states of the cutter head bearing (such as tool wear, formation mutation, etc.) will cause changes in the characteristics of vibration signals. Due to the complex underground environment and the interference of multiple factors on vibration signals, it is difficult for traditional methods to accurately identify anomalies.

[0003] Currently, there is a proposed solution that uses a diagnostic method based on vibration signal spectrum analysis and support vector machines. This method first extracts the spectral characteristics of vibration signals through fast Fourier transform, then uses principal component analysis for dimensionality reduction, and finally inputs them into a support vector machine model for classification. This method can identify abnormal states to a certain extent by using the frequency - domain energy distribution and machine learning algorithms.

[0004] This solution only relies on frequency - domain features and ignores the energy attenuation characteristics of vibration signals during spatial propagation, resulting in insufficient sensitivity to abnormal states such as formation mutation. In addition, principal component analysis is based on linear transformation and is difficult to effectively mine the non - linear correlation between vibration features and fault labels, affecting the accuracy of feature screening and thus reducing the reliability of diagnostic results. Summary of the Invention

[0005] This application provides a shield machine anomaly diagnosis method and system based on maximum information non - parameters to solve the problems of low accuracy and poor real - time performance in diagnosing abnormal states of shield machines in complex strata in the prior art.

[0006] In the first aspect, this application provides a shield machine anomaly diagnosis method based on maximum information non - parameters, including:

[0007] During the tunneling process of the shield machine, obtain the vibration signal of the cutter head bearing in the shield machine, and determine the spatial energy attenuation pattern of the vibration waveform formed when the vibration signal propagates in the complex underground strata;

[0008] Perform multi - frequency domain decomposition on the vibration signal to obtain a multi - band energy distribution set;

[0009] Use the maximum mutual information criterion to select the target abnormal - band energy distribution from the multi - band energy distribution set, and the maximum mutual information criterion is determined by maximum information non - parameters;

[0010] Generate a collaborative diagnosis feature vector according to the spatial energy attenuation pattern and the target abnormal - band energy distribution;

[0011] Input the collaborative diagnosis feature vector into a preset random forest classifier to output a diagnosis result matching the abnormal tunneling state of the shield machine.

[0012] Optionally, the selecting the target abnormal frequency band energy distribution from the multi-band energy distribution set by using the maximum mutual information criterion includes:

[0013] Based on the multi-band energy distribution set, with the energy distribution of the current frequency band as the first variable and a preset set of abnormal state labels as the second variable, construct a joint probability distribution table between the first variable and the second variable, where the set of abnormal state labels is pre-labeled based on the energy distribution characteristics corresponding to different abnormal types in historical data;

[0014] Based on the maximum information non-parameter in the maximum mutual information criterion, calculate the occurrence probability of each energy value in the first variable and the occurrence probability of each abnormal label in the second variable respectively;

[0015] Based on the joint probability distribution table and each of the occurrence probabilities, calculate the non-linear correlation index value between the energy distribution of the current frequency band and the abnormal state;

[0016] Mark the energy distribution corresponding to the frequency band with the non-linear correlation index value greater than the preset index threshold as the target abnormal frequency band energy distribution.

[0017] Optionally, the calculating the non-linear correlation index value between the energy distribution of the current frequency band and the abnormal state based on the joint probability distribution table and each of the occurrence probabilities includes:

[0018] Extract one by one from the joint probability distribution table the joint probability value of each energy value in the energy distribution of the current frequency band and the corresponding abnormal label;

[0019] For each combination of an energy value and an abnormal label, multiply the occurrence probability of the energy value in the first variable by the occurrence probability of the abnormal label in the second variable to obtain the corresponding independent probability product result;

[0020] Perform a ratio operation on the joint probability value and the independent probability product result of each combination and take the natural logarithm of the ratio operation result;

[0021] Multiply the joint probability value of each combination by the corresponding natural logarithm result to obtain the contribution value of each combination;

[0022] Accumulate the contribution values of all combinations to obtain the non-linear correlation index value of the current frequency band.

[0023] Optionally, generating a collaborative diagnosis feature vector according to the spatial energy attenuation pattern and the target abnormal frequency band energy distribution includes:

[0024] Extracting an energy attenuation rate parameter in the propagation direction, an energy difference parameter between adjacent sensor nodes, and an energy distribution symmetry parameter from the spatial energy attenuation pattern;

[0025] Extracting a frequency band span parameter, an energy concentration parameter, and an energy fluctuation amplitude parameter from the target abnormal frequency band energy distribution;

[0026] Taking the result of multiplying the energy attenuation rate parameter by the frequency band span parameter as a first weight coefficient, taking the result of multiplying the energy difference parameter by the energy concentration parameter as a second weight coefficient, and taking the result of multiplying the energy distribution symmetry parameter by the energy fluctuation amplitude parameter as a third weight coefficient;

[0027] Concatenating the first weight coefficient, the second weight coefficient, and the third weight coefficient in a preset order into a first multi-dimensional feature sequence, and adding a ratio parameter of the overlapping area between the spatial energy attenuation pattern and the abnormal frequency band energy distribution in the first multi-dimensional feature sequence to generate a collaborative diagnosis feature vector.

[0028] Optionally, adding the ratio parameter of the overlapping area between the spatial energy attenuation pattern and the abnormal frequency band energy distribution in the first multi-dimensional feature sequence to generate a collaborative diagnosis feature vector includes:

[0029] Counting the number of overlapping intervals between a first energy distribution interval, which is the energy distribution interval in the spatial energy attenuation pattern, and a second energy distribution interval, which is the energy distribution area in the abnormal frequency band energy distribution;

[0030] Dividing the number of overlapping intervals by the total number of distribution intervals corresponding to the spatial energy attenuation pattern to obtain the ratio parameter of the overlapping area;

[0031] Adding the ratio parameter of the overlapping area at the end of the last weight coefficient in the first multi-dimensional feature sequence to generate a second multi-dimensional feature sequence;

[0032] Generating a collaborative diagnosis feature vector based on the second multi-dimensional feature sequence.

[0033] Optionally, marking the energy distribution corresponding to the frequency band with the non-linear correlation index greater than the preset index threshold as the target abnormal frequency band energy distribution includes:

[0034] Comparing the non-linear correlation index value of the current frequency band with the preset index threshold;

[0035] When comparing the energy distributions corresponding to the frequency bands where the non-linear correlation index is greater than the preset index threshold, mark the energy distribution corresponding to the current frequency band as the preliminary abnormal frequency band energy distribution;

[0036] Perform verification processing on the preliminary abnormal frequency band energy distribution to generate the target abnormal frequency band energy distribution.

[0037] Optionally, the step of inputting the collaborative diagnosis feature vector into a preset random forest classifier and outputting a diagnosis result matching the abnormal tunneling state of the shield machine includes:

[0038] Input the collaborative diagnosis feature vector into a preset random forest classifier, and the preset random forest classifier disassembles the collaborative diagnosis feature vector into multiple independent feature items;

[0039] Through multiple decision paths in the preset random forest classifier, perform hierarchical judgment on each of the independent feature items;

[0040] According to the hierarchical judgment result, determine the abnormal type label corresponding to the independent feature item;

[0041] Match the abnormal type labels of all independent feature items with the abnormal states in the preset abnormal state mapping table;

[0042] If more than half of the independent feature items match the same abnormal state, then output the abnormal state and use the abnormal state as the diagnosis result.

[0043] In a second aspect, the present application provides a shield machine abnormal diagnosis system based on maximum information non-parameters, including:

[0044] An acquisition module, configured to acquire the vibration signal of the cutter head bearing in the shield machine during the tunneling process of the shield machine, and determine the spatial energy attenuation mode of the vibration waveform formed when the vibration signal propagates in the underground complex formation;

[0045] A decomposition module, configured to perform multi-frequency domain decomposition on the vibration signal to obtain a multi-frequency band energy distribution set;

[0046] A selection module, configured to select the target abnormal frequency band energy distribution from the multi-frequency band energy distribution set by using the maximum mutual information criterion, and the maximum mutual information criterion is determined by the maximum information non-parameters;

[0047] A generation module, configured to generate a collaborative diagnosis feature vector according to the spatial energy attenuation mode and the target abnormal frequency band energy distribution;

[0048] An input module for inputting the collaborative diagnosis feature vector into a preset random forest classifier and outputting a diagnosis result matching the abnormal tunneling state of the shield machine.

[0049] In a third aspect, the present application provides a computing device, including a processor and a memory. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the abnormal diagnosis method for a shield machine based on maximum information non-parameters according to any one of the first aspect.

[0050] In a fourth aspect, the present application provides a computer storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the abnormal diagnosis method for a shield machine based on maximum information non-parameters according to any one of the first aspect is implemented.

[0051] In the present application, an abnormal diagnosis method for a shield machine based on maximum information non-parameters is provided. The method includes: during the tunneling process of the shield machine, acquiring the vibration signal of the cutter head bearing in the shield machine and determining the spatial energy attenuation mode of the vibration waveform formed when the vibration signal propagates in the underground complex stratum; performing multi-frequency domain decomposition on the vibration signal to obtain a multi-band energy distribution set; using the maximum mutual information criterion to select the target abnormal band energy distribution from the multi-band energy distribution set, where the maximum mutual information criterion is determined by maximum information non-parameters; generating a collaborative diagnosis feature vector according to the spatial energy attenuation mode and the target abnormal band energy distribution; and inputting the collaborative diagnosis feature vector into a preset random forest classifier to output a diagnosis result matching the abnormal tunneling state of the shield machine.

[0052] The technical solution provided by the present application has the following beneficial effects:

[0053] By collecting the vibration signal of the cutter head bearing in real time and analyzing its energy attenuation characteristics during the propagation in the underground complex stratum, the present application effectively captures the influence of the change in the stratum structure on the vibration wave propagation, providing a spatial dimension feature basis for abnormal diagnosis. The signal decomposition technology is used to convert the vibration signal into a frequency domain energy distribution, completely retaining the fault feature information of different frequency bands and avoiding feature omission caused by single-frequency band analysis. The mutual information criterion constructed based on maximum information non-parameters accurately identifies the feature frequency band with the strongest correlation with the abnormal state from multiple frequency bands, improving the screening efficiency of abnormal features. The two-dimensional features of the spatial energy attenuation mode and the abnormal band energy distribution are fused to construct a composite feature vector with strong characterization ability, enhancing the completeness of the input features of the classifier. Through the hierarchical decision-making of multi-dimensional features by the ensemble learning algorithm, a high-precision matching of the abnormal type is achieved, ensuring the reliability of the diagnosis result.

[0054] Furthermore, the present application also constructs a joint probability distribution table of the energy distribution in the vibration signal frequency band and the preset anomaly label, calculates the independent probabilities of the energy value and the anomaly label respectively based on the maximum information non-parameter, and then quantifies the non-linear correlation index between the frequency band energy and the anomaly state. Finally, the frequency bands with index values exceeding the threshold are selected as the target anomaly features. This method breaks through the limitations of traditional linear analysis and realizes the accurate mining of the deep association between vibration signals and fault types.

[0055] Moreover, this solution effectively extracts the frequency band features in the vibration signal that have a strong non-linear association with the anomaly state through a combination of probability statistics and information theory, solves the problem of inaccurate extraction of fault features in complex stratum environments, and provides high-value feature inputs for subsequent diagnosis.

[0056] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0058] Figure 1 It is a flowchart of a shield machine anomaly diagnosis method based on the maximum information non-parameter provided by an embodiment of the present application;

[0059] Figure 2 It is a schematic structural diagram of a shield machine anomaly diagnosis system based on the maximum information non-parameter provided by an embodiment of the present application;

[0060] Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] In order to enable those skilled in the art to better understand the solution of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application.

[0062] In some of the processes described in the specification, claims, and above-mentioned drawings of this application, a number of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any order of execution. Additionally, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first", "second", etc. in this text are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0063] The researchers found that when the shield machine is tunneling in complex strata, the traditional vibration signal analysis method is difficult to accurately capture the correlation characteristics between the abnormal state of the cutter head bearing and the formation change, resulting in insufficient fault diagnosis accuracy. Based on this, the embodiment of this application provides a shield machine abnormal diagnosis method based on maximum information non-parameters. This method analyzes the spatial energy attenuation characteristics and multi-frequency band energy distribution of the vibration signal, uses the maximum mutual information criterion to screen key abnormal frequency bands, and fuses to generate a collaborative diagnosis feature vector, and finally realizes the accurate matching of the abnormal state through a random forest classifier. The technical solution of this application can be applied to the shield construction safety monitoring scenarios under complex geological conditions such as urban subway tunnels and underpass airport runways.

[0064] Next, the technical solutions in the embodiments of this application will be clearly and completely described in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of this application.

[0065] Figure 1 It is a flowchart of a shield machine abnormal diagnosis method based on maximum information non-parameters provided by the embodiment of this application, as Figure 1 shown, this method includes:

[0066] Step 101: During the tunneling process of the shield machine, obtain the vibration signal of the cutter head bearing in the shield machine, and determine the spatial energy attenuation mode of the vibration waveform formed when the vibration signal propagates in the underground complex strata.

[0067] In this step, the vibration signal of the cutter head bearing represents the mechanical vibration waveform data generated by the cutter head bearing during the tunneling of the shield machine, including time-domain amplitude and frequency characteristics. The underground complex stratum refers to the heterogeneous and variable geotechnical medium environment encountered during the tunneling of the shield machine. The spatial energy attenuation mode represents the quantitative characteristics of the vibration wave energy attenuation with the increase of the propagation distance during the propagation of the vibration wave in the stratum, including parameters such as attenuation rate and directionality.

[0068] In the embodiment of the present application, the vibration signal is collected in real time by a triaxial acceleration sensor installed on the cutter head bearing, and anti-interference filtering technology is used to eliminate equipment noise; a sensor array arranged on the outer shell of the shield machine is used to receive the stratum propagation signal, and a three-dimensional propagation model is established based on the wave equation theory; the signal energy envelope is extracted by wavelet time-frequency analysis, and the anisotropic attenuation coefficient is calculated in combination with the spatial position of the sensor, and finally a spatial energy attenuation mode including parameters such as the main attenuation direction and energy gradient is constructed.

[0069] For example, in an under-construction project of an airport runway, 8 vibration sensors arranged in a ring are used for real-time monitoring. It is found that at the interface between the concrete layer of the runway pavement and the underlying sand layer, the attenuation rate of the vibration signal in the vertical direction is higher than that in the horizontal direction. By comparing the signal intensity differences of sensors at different positions, the area with the change in the thickness of the runway structural layer is accurately identified.

[0070] Step 102: Perform multi-frequency domain decomposition on the vibration signal to obtain a multi-band energy distribution set.

[0071] In this step, multi-frequency domain decomposition represents the process of dividing the vibration signal into multiple sub-bands according to frequency components. The multi-band energy distribution set represents the set of energy values of each sub-band, characterizing the proportion of vibration energy of different frequency components, and is used to reflect the interference of the underground complex stratum structure. The dynamic change of the multi-band energy distribution set is directly related to the vibration energy attenuation caused by the reflection and diffraction of the underground medium interface.

[0072] In the embodiment of the present application, wavelet packet transform decomposition is performed on the preprocessed vibration signal, and a wavelet basis function is selected to achieve 6-layer decomposition; the energy values of the reconstructed signals of each node are calculated, and the frequency band is converted into 24 sub-bands that conform to the human ear perception characteristics through scale division; energy normalization processing is used to eliminate the amplitude difference, and finally a multi-band energy distribution set including the relative energy proportion of each frequency band is generated.

[0073] For example, in the above project, it is found through decomposition that the energy proportion of the vibration signal of the runway concrete layer in the high-frequency band is significantly higher than that of the sand layer. In particular, the energy ratio difference in the 315 - 400 Hz frequency band reaches 2.3 times (obtained by calculating the average energy ratio of this frequency band under the same working conditions of the concrete layer and the sand layer), and this feature is determined as the key index for stratum identification.

[0074] Step 103: Selecting a target abnormal frequency band energy distribution from the multi-frequency band energy distribution set by using a maximum mutual information criterion, wherein the maximum mutual information criterion is determined by a maximum information non-parameter.

[0075] In this step, the maximum mutual information criterion represents an indicator for measuring the statistical dependence between two variables. The target abnormal frequency band energy distribution specifically refers to the vibration energy distribution characteristics within a specific frequency range that is strongly correlated with the abnormal state of the shield machine, which is screened out by the maximum mutual information criterion. The maximum information nonparametric representation is a probability distribution calculation method based on kernel density estimation.

[0076] In an embodiment of the present application, a frequency band-fault label database containing historical abnormal cases is constructed; the window kernel density estimation method is used to calculate the joint probability distribution; the maximum information coefficient algorithm improved based on the divergence theory is used to calculate the correlation between each frequency band and the abnormal type; the dynamic threshold screening method is used to determine the frequency band, and finally the target abnormal frequency band energy distribution that is strongly associated with the typical risk of runway underpass is output.

[0077] For example, in response to the unique pavement settlement risk of runway underpass, the system automatically screened out the 80-160Hz frequency band as the key monitoring object. This selection was based on the calculation of the correlation coefficient between energy mutation and settlement deformation in this frequency band in historical cases (coefficient > 0.85), and successfully warned of three potential settlement areas during actual excavation.

[0078] Step 104: Generate a collaborative diagnosis feature vector according to the spatial energy attenuation pattern and the target abnormal frequency band energy distribution.

[0079] In this step, the collaborative diagnosis feature vector represents a multi-dimensional feature array that fuses spatial propagation and frequency domain features.

[0080] In an embodiment of the present application, a canonical correlation analysis is performed on the three principal components of the spatial attenuation pattern and the six characteristic parameters of the abnormal frequency band; the linear inseparable features are converted to a high-dimensional space by kernel function mapping; the top seven most discriminative features are retained by ranking the feature importance; and a 13-dimensional collaborative diagnosis feature vector is constructed by adding a spatiotemporal correlation factor.

[0081] For example, when crossing the critical area of ​​the runway, an abnormal peak appeared in the cross-feature of "vertical attenuation rate × 160Hz frequency band energy" in the fusion feature vector generated by the system. After inversion analysis, it was found that there was a pavement joint defect at that location.

[0082] Step 105: Input the collaborative diagnosis feature vector into a preset random forest classifier, and output a diagnosis result that matches the abnormal excavation state of the shield machine.

[0083] In this step, the preset random forest classifier includes an ensemble learning model with 200 decision trees. The abnormal tunneling state refers to the classification system of typical fault modes shown when the shield machine operates abnormally in complex strata. The diagnostic result is a structured decision report output by the system, including the determination of abnormal types, severity grading, and positioning information in three dimensions. For example: Class B2 fault (confidence level 89%), secondary warning, positioning: at the 368th ring + 1.2m, bearing abnormality 0.8m to the right of the Y-axis.

[0084] In the embodiment of the present application, a random forest classifier optimized by an ensemble strategy is adopted, with 100 decision trees set for parallel computing; the feature weights are dynamically adjusted based on error estimation; a transfer learning mechanism is introduced to adapt to different geological conditions; and a triple diagnostic result including abnormal types, position coordinates, and risk levels is output.

[0085] For example, during continuous tunneling, the system pre-warns of possible differential settlement in the central area of the runway according to the change trend of the feature vector, and the diagnostic result shows "settlement risk level II, located at the 11 o'clock direction of the cutter head".

[0086] This solution realizes multi-dimensional accurate perception of shield tunneling abnormalities by establishing a collaborative analysis system for the spatial propagation and frequency domain characteristics of vibration signals. Engineering applications show that the system can stably identify typical abnormalities such as tool wear and formation mutations, and the warning timeliness is improved compared with conventional methods, providing a full-process diagnostic ability for complex stratum construction that runs through abnormality discovery, positioning, and evaluation, effectively reducing construction risks and optimizing maintenance decisions.

[0087] To solve the problem of insufficient recognition accuracy of abnormal frequency bands during the tunneling process of the shield machine in complex strata, in some embodiments, step 103: The selecting the target abnormal frequency band energy distribution from the multi-band energy distribution set by using the maximum mutual information criterion includes:

[0088] Step 201: Based on the multi-band energy distribution set, with the energy distribution of the current frequency band as the first variable and the preset abnormal state label set as the second variable, construct a joint probability distribution table between the first variable and the second variable, where the abnormal state label set is pre-labeled based on the energy distribution characteristics corresponding to different abnormal types in historical data.

[0089] In step 201, the joint probability distribution table refers to a two-dimensional table recording the co-occurrence frequencies of each energy value and abnormal labels, where the first variable is the interval division after discretization of the frequency band energy value, and the second variable is the pre-defined abnormal type label set.

[0090] In the embodiments of the present application, first, the vibration signals in the historical data are processed by energy value binning, divided into several equal-width intervals; at the same time, a label system including typical anomalies such as tool wear and formation mutation is established; then, the co-occurrence frequencies of each energy interval and each anomaly label are counted, and the Laplace smoothing process is used to avoid the zero-probability problem, and finally a complete joint probability distribution table is constructed.

[0091] Step 202: Based on the maximum information non-parameter in the maximum mutual information criterion, calculate the occurrence probability of each energy value in the first variable and the occurrence probability of each anomaly label in the second variable respectively.

[0092] In step 202, the occurrence probability includes the marginal probability of the energy value and the marginal probability of the anomaly label, which respectively represent the independent occurrence frequencies of each energy interval and anomaly label in the historical data.

[0093] In the embodiments of the present application, the kernel density estimation method is used to calculate the probability distribution of the energy value, and the discrete data is smoothed by the Gaussian kernel function; for the anomaly label probability, the maximum likelihood estimation is directly based on the label occurrence frequency; both probability calculations are normalized to ensure that the sum of probabilities is 1.

[0094] Step 203: Based on the joint probability distribution table and each of the occurrence probabilities, calculate the non-linear correlation index value between the current frequency band energy distribution and the abnormal state.

[0095] In step 203, the non-linear correlation index value is a metric value that quantifies the association strength between the frequency band energy and the abnormal state.

[0096] In the embodiments of the present application, first calculate the ratio of the product of the joint probability and the marginal probability, then multiply by the joint probability after taking the logarithm, and finally perform a weighted sum for all possible energy-label combinations; use the Monte Carlo integration method to handle the probability calculation problem of continuous variables to ensure the calculation accuracy of the index value.

[0097] Step 204: Mark the energy distribution corresponding to the frequency band with the non-linear correlation index value greater than the preset index threshold as the target abnormal frequency band energy distribution.

[0098] In step 204, the preset index threshold is a discrimination boundary determined by analyzing the distribution difference of the index values between the normal working conditions and abnormal working conditions in the historical data.

[0099] In the embodiments of the present application, the curve analysis method is used to determine the optimal threshold to maximize the difference between the true positive rate and the false positive rate; the selected abnormal frequency bands are verified twice to ensure that their energy distribution patterns have a stable association characteristic with the target anomaly.

[0100] The following is a specific example:

[0101] In an under - runway crossing project at an airport, when the system screens for abnormal frequency bands in the vibration signals collected from the cutter head bearing, it first establishes a label library containing 5 types of typical abnormalities such as pavement settlement and tool wear based on 6 - month historical construction data (the basis for label setting is the record of 32 actual abnormal events). For the 24 frequency bands obtained by decomposing the vibration signals in the current tunneling section (the frequency band division uses the engineering vibration analysis frequency band), the 80 - 160Hz frequency band with energy fluctuations is selected as the analysis object. The energy monitoring values of this frequency band for 10 consecutive minutes (sampling frequency 1000Hz, processed by moving average) are divided into 20 equal - width intervals as the first variable, and a joint probability distribution table (a total of 100 valid samples) is constructed with the abnormal label library. Through kernel density estimation, the occurrence probability of the energy value of this frequency band during settlement abnormality is calculated to be 0.38 (based on the statistical analysis of 28 settlement events in historical data), and at the same time, the marginal probability of this abnormal label is calculated to be 0.25 (the proportion of settlements in 32 abnormal events). Finally, the calculated non - linear correlation index value of this frequency band is 0.21 (calculated through the mutual information formula), which exceeds the preset threshold of 0.18 (determined by curve analysis of 100 groups of verification data). Therefore, this frequency band is marked as the target abnormal frequency band for pavement settlement monitoring in the under - runway crossing project.

[0102] In the embodiment of the present application, this solution realizes the intelligent screening of abnormal frequency bands through the establishment of an accurate probability statistical model and an innovative non - linear correlation measurement method, not only improving the accuracy of diagnosis but also reducing the false alarm rate, providing a reliable technical guarantee for under - crossing projects in sensitive areas such as airport runways.

[0103] To solve the problem of insufficient quantification accuracy of the correlation between frequency - band energy and abnormal states in shield tunneling construction, in some embodiments, step 203: calculating the non - linear correlation index value between the current frequency - band energy distribution and the abnormal state based on the joint probability distribution table and each of the occurrence probabilities includes:

[0104] Step 301: Extract, one by one from the joint probability distribution table, the joint probability values of each energy value in the current frequency - band energy distribution and the corresponding abnormal label.

[0105] In step 301, the joint probability value refers to the probability of the simultaneous occurrence of a specific energy interval and a specific abnormal label, reflecting the statistical law of their co - occurrence.

[0106] In the embodiment of the present application, the sliding window method is used to extract data from the distribution table. The window width is set to 10 consecutive sampling points, and each time it slides 1 point to ensure data continuity; after counting the frequencies of the energy - label combinations within each window, the Laplace smoothing technique is used to process the zero - frequency problem, and finally, a stable joint probability estimate is obtained.

[0107] Step 302: For each combination of the energy value and the anomaly label, multiply the occurrence probability of the energy value in the first variable by the occurrence probability of the anomaly label in the second variable to obtain a corresponding independent probability product result.

[0108] In step 302, the independent probability product result represents the theoretical co-occurrence probability that should exist when the energy value and the anomaly label are independent of each other.

[0109] In the embodiments of the present application, the theoretical probability is calculated by the marginal probability multiplication rule, where the occurrence probability of the energy value is obtained by Gaussian kernel density estimation, and the anomaly label probability is obtained by historical frequency estimation; the product result is normalized to ensure that the sum of the theoretical probabilities of all combinations is 1.

[0110] Step 303: Perform a ratio operation on the joint probability value and the independent probability product result of each combination, and take the natural logarithm of the ratio operation result.

[0111] In step 303, the ratio operation reflects the deviation degree of the actual observed probability from the theoretical independent probability, and the natural logarithm processing is used to amplify the difference.

[0112] In the embodiments of the present application, fixed-point number operations are used to ensure the calculation accuracy, and an exception handling mechanism is set to handle division-by-zero errors; the logarithmic operation is approximated by Taylor expansion to improve the operation efficiency while ensuring the accuracy.

[0113] Step 304: Multiply the joint probability value of each combination by the corresponding natural logarithm result to obtain the contribution value of each combination.

[0114] In step 304, the contribution value quantifies the contribution degree of a single energy-label combination to the overall correlation.

[0115] In the embodiments of the present application, a weight coefficient is introduced to adjust the contribution weight of each combination, and appropriate gain is given to rare but important anomaly patterns; parallel computing technology is used to accelerate the calculation of the contribution values of large-scale combinations.

[0116] Step 305: Accumulate the contribution values of all combinations to obtain the non-linear correlation index value of the current frequency band.

[0117] In the embodiments of the present application, a hierarchical accumulation algorithm is used to avoid numerical overflow, and the final result is normalized so that its value range is limited between 0 and 1, which is convenient for subsequent threshold comparison.

[0118] The following is a specific example:

[0119] In the specific implementation of an airport runway under-crossing project, when analyzing the vibration energy data in the 80 - 160 Hz frequency band, the system first extracts data from the constructed joint probability distribution table (including 100 groups of samples with 20 energy intervals and 5 types of abnormal labels). Taking the combination of the energy interval [120, 140] and the "pavement settlement" label as an example: By statistically analyzing historical data, it is found that this combination actually appears 28 times (out of a total of 100 samples). The calculated joint probability value is 0.28; the occurrence probability of this energy interval is 0.35 (calculated based on the fact that this interval appears 35 times in all samples), and the occurrence probability of the "pavement settlement" label is 0.25 (the proportion of settlements in historical abnormal events). The product of the two gives the independent probability product result of 0.0875. Dividing the joint probability value of 0.28 by the independent probability of 0.0875 gives a ratio of 3.2. Taking the natural logarithm, it is approximately 1.16, and then multiplying by the joint probability value of 0.28 gives a contribution value of this combination of approximately 0.325. The system repeats the above calculation process for all 300 valid combinations of 20 energy intervals and 5 types of abnormal labels (some combinations are filtered due to insufficient data), and finally accumulates to obtain a non-linear correlation index value of 0.21 for this frequency band (this value is confirmed to be valid through statistical analysis of 100 groups of verification samples).

[0120] In the embodiment of the present application, this solution realizes the accurate characterization of the non-linear association between vibration characteristics and construction anomalies by establishing a refined probability statistical model and an innovative correlation quantification method, providing a reliable basis for abnormal early warning in shield construction in sensitive areas such as airport runways.

[0121] To solve the problem of insufficient multi-dimensional feature fusion in shield construction, in some embodiments, step 104: generating a collaborative diagnosis feature vector according to the spatial energy attenuation pattern and the energy distribution of the target abnormal frequency band includes:

[0122] Step 401: Extract the energy attenuation rate parameter in the propagation direction, the energy difference parameter between adjacent sensor nodes, and the energy distribution symmetry parameter from the spatial energy attenuation pattern.

[0123] In step 401, the propagation direction specifically refers to the direction of the energy transfer path of the vibration waveform between the cutter head bearing of the shield machine and the formation sensor. The energy attenuation rate parameter refers to the amount of energy loss per unit distance of the vibration wave in a specific propagation direction. Adjacent sensor nodes refer to the vibration signal acquisition devices deployed in the formation outside the cutter head bearing of the shield machine. The energy difference parameter is the standard deviation of the energy values received by adjacent sensor nodes. The energy distribution symmetry parameter specifically refers to the degree of difference in the distribution of the vibration wave energy at the centrally symmetric positions of the cutter head bearing. It is used to quantify the asymmetric characteristics of the spatial distribution of the vibration energy and reflect the distortion effect of the formation inhomogeneity on the propagation of the vibration wave. Specifically, it is calculated through three elements: the energy reception ratio of the symmetric sensor node pair, the position offset of the energy extreme point, and the tilt angle of the main energy axis.

[0124] In the embodiment of the present application, the least squares method is used to fit the energy attenuation curve of each propagation direction to obtain the rate parameter; the energy difference between nodes is calculated based on the annularly arranged sensor array; the symmetry parameter is calculated by the mirror comparison method, and finally a three-dimensional spatial feature group is formed.

[0125] Step 402: Extract the frequency band span parameter, energy concentration parameter, and energy fluctuation amplitude parameter from the energy distribution of the target abnormal frequency band.

[0126] In step 402, the frequency band span parameter is the width of the frequency range of the abnormal frequency band. The energy concentration parameter is the proportion of the energy of the first three main peaks in the frequency band. The energy fluctuation amplitude parameter is the range of the change of the frequency band energy over time.

[0127] In the embodiment of the present application, the frequency band boundary is determined by spectral envelope analysis to calculate the span; the peak search algorithm is used to extract the main energy components; the time-domain statistical analysis is used to obtain the fluctuation characteristics, and finally a three-dimensional frequency domain feature group is constructed.

[0128] Step 403: Take the result of multiplying the energy attenuation rate parameter by the frequency band span parameter as the first weight coefficient, take the result of multiplying the energy difference parameter by the energy concentration parameter as the second weight coefficient, and take the result of multiplying the energy distribution symmetry parameter by the energy fluctuation amplitude parameter as the third weight coefficient.

[0129] In step 403, the weight coefficient is an enhanced feature obtained by cross-multiplying the features and is used to highlight the key diagnostic information.

[0130] In the embodiment of the present application, the Cartesian product combination of the spatial features and the frequency domain features is performed, and the dynamic weighting algorithm is used to adjust the product coefficient to ensure the unity of the feature dimensions in each dimension, and finally three groups of enhanced feature values are generated.

[0131] Step 404: Concatenate the first weight coefficient, the second weight coefficient, and the third weight coefficient in a preset order to form a first multi-dimensional feature sequence, and append the ratio parameter of the overlapping region between the spatial energy attenuation pattern and the abnormal frequency band energy distribution to the first multi-dimensional feature sequence to generate a collaborative diagnosis feature vector.

[0132] In step 404, the ratio parameter of the overlapping region is the ratio of the intersection of the spatial energy distribution and the frequency band energy distribution in the time-frequency domain.

[0133] In the embodiment of the present application, the time-frequency matrix dot product operation is used to calculate the overlapping region; the ratio parameter is obtained by the area integration method; the features are concatenated after being sorted according to the diagnostic importance, and finally a collaborative diagnosis feature vector is generated.

[0134] The following is a specific example:

[0135] During the implementation of an airport runway under-crossing project, when the system performs feature fusion for the identified target abnormal frequency band of 80 - 160 Hz, three key parameters are first extracted from the spatial energy attenuation pattern: the vertical direction energy attenuation rate parameter is 0.15 per second (obtained by fitting the data of 5 vibration sensors arranged from 1 meter to 5 meters behind the cutter head), the energy difference parameter between adjacent sensor nodes is 0.3 (calculated as the standard deviation of the energy received by 8 circumferentially arranged sensors), and the energy distribution symmetry parameter is 0.85 (obtained by comparing the energy reception ratio of sensors at symmetric positions at 3 o'clock and 9 o'clock of the cutter head). At the same time, the frequency band span parameter of 80 Hz (the difference between the upper and lower limits of the frequency band determined by spectral analysis), the energy concentration parameter of 0.6 (the proportion of the energy of the first three main peaks in the total energy within this frequency band), and the energy fluctuation amplitude parameter of 1.2 V (the maximum and minimum amplitude difference within a 10-minute monitoring period) are extracted from the target abnormal frequency band. Multiply the attenuation rate by the frequency band span to obtain the first weight coefficient of 12, multiply the energy difference by the concentration to obtain the second weight coefficient of 0.18, and multiply the symmetry by the fluctuation amplitude to obtain the third weight coefficient of 1.02. The ratio parameter of the overlapping region of 0.45 (the ratio of the region where the spatial energy distribution and the frequency band energy distribution appear simultaneously) is calculated through the time-frequency analysis matrix, and finally a collaborative diagnosis vector containing these 4 features is generated.

[0136] In the embodiment of the present application, this solution realizes the deep collaboration between the spatial propagation characteristics and the frequency domain characteristics through an innovative multi-source feature fusion method, improves the reliability of abnormal diagnosis and the early warning ability under complex working conditions, and provides an effective technical guarantee for shield construction in sensitive areas such as airport runways.

[0137] In order to solve the problem of insufficient spatio-temporal correlation representation in the feature fusion process, in some embodiments, step 404: attaching the ratio parameter of the overlapping region between the spatial energy decay pattern and the abnormal frequency band energy distribution in the multi-dimensional feature sequence to generate a collaborative diagnosis feature vector, includes:

[0138] Step 501: Count the number of overlapping intervals between the first energy distribution interval and the second energy distribution interval. The first energy distribution interval is the energy distribution interval in the spatial energy decay pattern, and the second energy distribution interval is the energy distribution area in the abnormal frequency band energy distribution.

[0139] In step 501, the number of overlapping intervals refers to the number of regions where the spatial energy distribution and the frequency band energy distribution simultaneously appear in the time-frequency domain, reflecting the spatio-temporal correlation strength of the two features.

[0140] In the embodiments of the present application, by establishing a time-frequency distribution matrix, the heat map of the spatial energy distribution and the frequency spectrum map of the frequency band energy distribution are compared point by point, and the sliding window matching algorithm is used to count the number of overlapping regions that meet the double energy threshold. The window size is adaptively adjusted according to the sampling frequency.

[0141] Step 502: Divide the number of overlapping intervals by the total number of distribution intervals corresponding to the spatial energy decay pattern to obtain the ratio parameter of the overlapping region.

[0142] In the embodiments of the present application, the area integration method is used to calculate the total distribution interval, the dimension influence is eliminated through normalization processing, and the boundary effect is processed by introducing a smoothing coefficient, and finally the standardized ratio parameter is obtained.

[0143] Step 503: Add the ratio parameter of the overlapping region to the end of the last weight coefficient in the first multi-dimensional feature sequence to generate a second multi-dimensional feature sequence.

[0144] In step 503, the second multi-dimensional feature sequence is a new sequence that enhances the spatio-temporal correlation features on the basis of the original features.

[0145] In the embodiments of the present application, a feature buffer is designed to ensure data alignment, the dynamic memory allocation technology is used to achieve feature expansion, and the data integrity is ensured through a verification mechanism.

[0146] Step 504: Generate a collaborative diagnosis feature vector based on the second multi-dimensional feature sequence.

[0147] In the embodiments of the present application, a feature compression algorithm is used to reduce the dimension, and the transmission efficiency is improved through quantization coding, and finally a standardized feature vector suitable for the input of the classifier is generated.

[0148] The following is a specific example:

[0149] During the implementation of an airport runway under-crossing project, when the system generates collaborative diagnostic feature vectors, it first divides 20 detection intervals for the spatial energy attenuation pattern (from 1 meter to 5 meters behind the cutter head, with each interval being 0.2 meters), and at the same time divides 15 frequency bands for the abnormal frequency band of 80 - 160 Hz (each frequency band being 5.33 Hz). Through time-frequency analysis, it is found that there are 9 overlapping intervals in areas such as the depth interval of 3.2 - 4.0 meters and the frequency band of 105 - 125 Hz (obtained by comparing the spatial sensor data with the spectrum analysis results). Divide the number of overlapping intervals, 9, by the total number of spatial intervals, 20, to obtain the overlapping area ratio parameter of 0.45. Then, splice the previously calculated three groups of weight coefficients [12, 0.18, 1.02] and the ratio parameter 0.45 in sequence to form a new four-dimensional feature sequence [12, 0.18, 1.02, 0.45].

[0150] In the embodiment of the present application, this solution effectively improves the characterization ability of the feature vector for complex working conditions through an innovative spatio-temporal correlation feature enhancement method, enabling the system to more accurately identify hidden engineering risks and providing reliable technical support for underground crossing projects in sensitive areas such as airport runways.

[0151] To solve the problem of a relatively high false alarm rate during the screening process of abnormal frequency bands, in some embodiments, step 204: marking the energy distribution corresponding to the frequency band with a non-linear correlation index greater than the preset index threshold as the target abnormal frequency band energy distribution includes:

[0152] Step 601: Compare the non-linear correlation index value of the current frequency band with the preset index threshold.

[0153] In step 601, the preset index threshold is a discrimination criterion determined by analyzing the distribution differences of the index values in historical data for normal and abnormal working conditions.

[0154] In the embodiment of the present application, a kernel density estimation method is used to construct an index value probability distribution model. By determining the optimal threshold point, the ratio of the true positive rate to the false positive rate is maximized, and an appropriate safety margin is set according to the engineering safety level.

[0155] Step 602: When the comparison result is that the energy distribution corresponding to the frequency band with a non-linear correlation index greater than the preset index threshold, mark the energy distribution corresponding to the current frequency band as the preliminary abnormal frequency band energy distribution.

[0156] In step 602, the preliminary abnormal frequency band energy distribution refers to the candidate abnormal frequency band data obtained through preliminary threshold screening.

[0157] In the embodiments of the present application, a dynamic threshold comparison mechanism is established to perform sliding window verification on the index values of each frequency band, and a double-buffer technology is used to store candidate frequency band data to ensure the real-time and continuous nature of the screening process.

[0158] Step 603: Verify and process the preliminary abnormal frequency band energy distribution to generate a target abnormal frequency band energy distribution.

[0159] In step 603, the verification process refers to verifying whether there are at least two consecutive frequency bands in adjacent frequency bands whose index values exceed the index threshold. If there is a phenomenon of synchronous threshold exceeding in the consecutive frequency bands, the mark of the preliminary abnormal frequency band energy distribution is retained; if not, the preliminary abnormal frequency band energy distribution is removed from the mark list.

[0160] In the embodiments of the present application, first, the continuous abnormality of the frequency band energy is tested through time series analysis, then the matching degree between the abnormal frequency band and the formation characteristics is verified in combination with geological exploration data, and finally, Bayesian network is used for comprehensive evaluation.

[0161] The following is a specific example:

[0162] During the implementation of an airport runway undercrossing project, when the system determines the abnormality of the 80 - 160 Hz frequency band, first, the non-linear correlation index value of 0.21 calculated for this frequency band (obtained by mutual information calculation based on 100 groups of samples) is compared with the preset threshold of 0.18 (determined by analyzing 50 groups of normal working conditions and 50 groups of abnormal working condition data). Since the index value exceeds the threshold, the system marks this frequency band as a preliminary abnormal frequency band. Subsequently, verification processing is carried out: by analyzing the monitoring data of 3 consecutive tunneling rings (each ring advances 1.2 meters), it is found that the energy value of this frequency band continuously remains above 1.5 times the baseline value (the baseline value is the average energy value during the normal construction of the first 30 rings); at the same time, by comparing the geological radar data, it is confirmed that the coincidence degree between the energy abnormal area and the runway pavement joint position reaches 85% (calculated by the spatial position matching algorithm). Finally, the system officially determines this frequency band as the target abnormal frequency band.

[0163] In the embodiments of the present application, this solution effectively reduces the false alarm rate of abnormal frequency band identification by establishing a strict two-stage verification mechanism, and at the same time ensures the detection rate of important engineering risks, providing a reliable frequency band screening method for underground crossing projects in sensitive areas such as airport runways.

[0164] To solve the accuracy problem of multi-feature collaborative decision-making in shield tunneling abnormal diagnosis, in some embodiments, step 105: inputting the collaborative diagnosis feature vector into a preset random forest classifier and outputting a diagnosis result matching the tunneling abnormal state of the shield machine includes:

[0165] Step 701: Input the collaborative diagnosis feature vector into a preset random forest classifier, and the preset random forest classifier disassembles the collaborative diagnosis feature vector into multiple independent feature items.

[0166] In step 701, an independent feature item refers to a single feature component in the collaborative diagnosis feature vector that has clear engineering significance. It includes the feature parameters of the spatial energy attenuation pattern and the feature parameters of the energy distribution in the target abnormal frequency band.

[0167] In the embodiment of the present application, a feature importance ranking algorithm is used to analyze the collaborative diagnosis feature vector, the independent contribution degree of each feature is determined based on the Gini coefficient, and features with strong correlation are combined through feature grouping technology, and finally disassembled into multiple independent feature items with clear physical meanings.

[0168] Step 702: Perform hierarchical judgment on each of the independent feature items through multiple decision paths in the preset random forest classifier.

[0169] In step 702, hierarchical judgment refers to the process of gradually discriminating features from the root node to the leaf node of the decision tree.

[0170] In the embodiment of the present application, each decision tree is set with 10 hierarchical judgment nodes, the information gain ratio is used as the splitting criterion, the tree depth is controlled through pre-pruning technology, and multi-path parallel evaluation is performed on each feature item.

[0171] Step 703: Determine the abnormal type label corresponding to the independent feature item according to the hierarchical judgment result.

[0172] In step 703, the abnormal type label is the abnormal category code corresponding to the leaf node of the decision tree.

[0173] In the embodiment of the present application, a marking system including 5-bit codes is established. The first two bits represent the major abnormal category, and the last three bits represent the specific subcategory. The final label is determined through the leaf node voting mechanism.

[0174] Step 704: Match the abnormal type labels of all the independent feature items with the abnormal states in the preset abnormal state mapping table.

[0175] In step 704, the preset abnormal state mapping table is a correspondence table between feature labels and engineering abnormal states.

[0176] In the embodiment of the present application, the mapping relationship is constructed based on 300 groups of historical abnormal cases, the fuzzy matching algorithm is used to process atypical features, and the confidence threshold is set to filter out low-reliability matches.

[0177] Step 705: If more than half of the independent feature items match the same abnormal state, output the abnormal state and use the abnormal state as the diagnostic result.

[0178] In the embodiment of the present application, a weighted voting mechanism is adopted to assign higher weights to important features, and a double confirmation mechanism is set to ensure the reliability of the result. Finally, a structured diagnostic report including the abnormal type, location, and risk level is output.

[0179] The following is a specific example:

[0180] In an under-construction project of an airport runway, the system inputs a collaborative diagnostic vector containing 4 features such as the spatial attenuation parameter (12) and the frequency band correlation parameter (0.21) into a preset random forest classifier. The classifier first disassembles the vector into 4 independent feature items (the parameters are from the previous feature fusion calculation). Through parallel judgment by 100 decision trees, the spatial attenuation parameter is judged to be abnormal in 82 trees (the threshold is set to 1.5 times the historical normal value), and the frequency band correlation parameter shows abnormality in 79 trees (based on the mutual information threshold of 0.18). According to the preset marking rules ("starting with '10' indicates pavement abnormality, and '20' indicates tool abnormality"), these two feature items are respectively marked with "10012" (pavement joint abnormality) and "10015" (pavement thickness abnormality). In the abnormal state mapping table, both of these marks correspond to the major category of "runway pavement structure abnormality". Since more than 60 feature items (accounting for 60% of the total) match this type of abnormality, the system finally outputs a diagnostic result of "runway pavement structure abnormality, risk level II".

[0181] In the embodiment of the present application, this solution realizes high-precision diagnosis of shield tunneling anomalies through an innovative multi-level feature decision-making mechanism, improves the recognition accuracy of engineering risks under complex working conditions, and provides reliable technical support for the underground crossing construction of major projects such as airport runways.

[0182] Figure 2 The structural schematic diagram of a shield machine anomaly diagnosis system provided for the embodiment of the present application is as Figure 2 shown, and the system includes:

[0183] An acquisition module 21, configured to acquire the vibration signal of the cutter head bearing in the shield machine during the tunneling process of the shield machine, and determine the spatial energy attenuation mode of the vibration waveform formed when the vibration signal propagates in the underground complex formation.

[0184] A decomposition module 22, configured to perform multi-frequency domain decomposition on the vibration signal to obtain a multi-band energy distribution set.

[0185] A selection module 23 is configured to select a target abnormal frequency band energy distribution from the set of multi-band energy distributions by using a maximum mutual information criterion, which is determined by a maximum information non-parameter.

[0186] A generation module 24 is configured to generate a collaborative diagnosis feature vector according to the spatial energy attenuation pattern and the target abnormal frequency band energy distribution.

[0187] An input module 25 is configured to input the collaborative diagnosis feature vector into a preset random forest classifier and output a diagnosis result matching the abnormal tunneling state of the shield machine.

[0188] Figure 2 The abnormal diagnosis system of the shield machine based on the maximum information non-parameter can execute Figure 1 The abnormal diagnosis method of the shield machine based on the maximum information non-parameter described in the illustrated embodiment, and its implementation principle and technical effects will not be elaborated. For the abnormal diagnosis system of the shield machine based on the maximum information non-parameter in the above embodiment, the specific manners in which each module and unit perform operations have been described in detail in the embodiment related to the method, and will not be elaborated here.

[0189] In a possible design, Figure 2 The abnormal diagnosis system of the shield machine based on the maximum information non-parameter in the illustrated embodiment can be implemented as a computing device, such as Figure 3 shown, and the computing device may include a storage component 31 and a processing component 32;

[0190] The storage component 31 stores one or more computer instructions, and the one or more computer instructions are called and executed by the processing component 32.

[0191] The processing component 32 is the above Figure 1 The abnormal diagnosis method of the shield machine based on the maximum information non-parameter in the illustrated embodiment.

[0192] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above methods. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for performing the above methods.

[0193] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component 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 memory, flash memory, magnetic disks or optical discs.

[0194] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0195] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0196] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0197] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above processing component, storage component, etc. may be basic server resources leased or purchased from a cloud computing platform.

[0198] The embodiments of the present application also provide a computer storage medium storing a computer program, which when executed by a computer can implement the above-mentioned Figure 1 abnormal diagnosis method for a shield machine based on maximum information non-parameters shown in the embodiments.

[0199] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0200] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

[0201] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0202] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A shield machine abnormal diagnosis method based on maximum information non-parameters, characterized in that Including: During the tunneling process of a shield machine, obtain the vibration signal of the cutter head bearing in the shield machine, and determine the spatial energy attenuation mode of the vibration waveform formed when the vibration signal propagates in the underground complex stratum; Perform multi-frequency domain decomposition on the vibration signal to obtain a multi-band energy distribution set; Use the maximum mutual information criterion to select the target abnormal band energy distribution from the multi-band energy distribution set, and the maximum mutual information criterion is determined by the maximum information non-parameter; Generate a collaborative diagnosis feature vector according to the spatial energy attenuation mode and the target abnormal band energy distribution; Input the collaborative diagnosis feature vector into a preset random forest classifier, and output a diagnosis result matching the abnormal tunneling state of the shield machine; The generating a collaborative diagnosis feature vector according to the spatial energy attenuation mode and the target abnormal band energy distribution includes: Extract the energy attenuation rate parameter in the propagation direction, the energy difference parameter between adjacent sensor nodes, and the energy distribution symmetry parameter from the spatial energy attenuation mode; Extract the band span parameter, the energy concentration parameter, and the energy fluctuation amplitude parameter from the target abnormal band energy distribution; Use the result of multiplying the energy attenuation rate parameter by the band span parameter as the first weight coefficient, the result of multiplying the energy difference parameter by the energy concentration parameter as the second weight coefficient, and the result of multiplying the energy distribution symmetry parameter by the energy fluctuation amplitude parameter as the third weight coefficient; Splice the first weight coefficient, the second weight coefficient, and the third weight coefficient in a preset order into a first multi-dimensional feature sequence, and append the overlapping area ratio parameter of the spatial energy attenuation mode and the abnormal band energy distribution in the first multi-dimensional feature sequence to generate a collaborative diagnosis feature vector.

2. The method according to claim 1, wherein The using the maximum mutual information criterion to select the target abnormal band energy distribution from the multi-band energy distribution set includes: Based on the multi-band energy distribution set, use the energy distribution of the current band as the first variable and the preset abnormal state label set as the second variable to construct a joint probability distribution table between the first variable and the second variable, where the abnormal state label set is pre-labeled based on the energy distribution characteristics corresponding to different abnormal types in historical data; Based on the maximum information non-parameter in the maximum mutual information criterion, calculate the occurrence probability of each energy value in the first variable and the occurrence probability of each abnormal label in the second variable respectively; Based on the joint probability distribution table and each of the occurrence probabilities, calculate the non-linear correlation index value between the current band energy distribution and the abnormal state; Mark the energy distribution corresponding to the band with the non-linear correlation index value greater than the preset index threshold as the target abnormal band energy distribution.

3. The method according to claim 2, wherein The calculating the non-linear correlation index value between the current band energy distribution and the abnormal state based on the joint probability distribution table and each of the occurrence probabilities includes: Extract the joint probability value of each energy value in the current band energy distribution and the corresponding abnormal label one by one from the joint probability distribution table; For each combination of the energy value and the anomaly label, multiply the occurrence probability of the energy value in the first variable by the occurrence probability of the anomaly label in the second variable to obtain a corresponding independent probability product result; Perform a ratio operation on the joint probability value and the independent probability product result of each combination, and take the natural logarithm of the ratio operation result; Multiply the joint probability value of each combination by the corresponding natural logarithm result to obtain the contribution value of each combination; Accumulate the contribution values of all combinations to obtain the non-linear correlation index value of the current frequency band.

4. The method according to claim 1, wherein Attaching the ratio parameter of the overlapping area between the spatial energy attenuation pattern and the abnormal frequency band energy distribution to the first multi-dimensional feature sequence to generate a collaborative diagnosis feature vector includes: Count the number of overlapping intervals between the first energy distribution interval and the second energy distribution interval, where the first energy distribution interval is the energy distribution interval in the spatial energy attenuation pattern, and the second energy distribution interval is the energy distribution area in the abnormal frequency band energy distribution; Divide the number of overlapping intervals by the total number of distribution intervals corresponding to the spatial energy attenuation pattern to obtain the ratio parameter of the overlapping area; Add the ratio parameter of the overlapping area to the end of the last weight coefficient in the first multi-dimensional feature sequence to generate a second multi-dimensional feature sequence; Generate a collaborative diagnosis feature vector based on the second multi-dimensional feature sequence.

5. The method according to claim 2, wherein Marking the energy distribution corresponding to the frequency band with the non-linear correlation index greater than the preset index threshold as the target abnormal frequency band energy distribution includes: Compare the non-linear correlation index value of the current frequency band with the preset index threshold; When the comparison result is that the non-linear correlation index is greater than the preset index threshold for the energy distribution of the frequency band, mark the energy distribution corresponding to the current frequency band as the preliminary abnormal frequency band energy distribution; Perform verification processing on the preliminary abnormal frequency band energy distribution to generate the target abnormal frequency band energy distribution.

6. The method according to claim 1, wherein Inputting the collaborative diagnosis feature vector into a preset random forest classifier and outputting a diagnosis result matching the abnormal tunneling state of the shield machine includes: Input the collaborative diagnosis feature vector into a preset random forest classifier, and the preset random forest classifier disassembles the collaborative diagnosis feature vector into multiple independent feature items; Perform hierarchical judgment on each independent feature item through multiple decision paths in the preset random forest classifier; Determine the abnormal type label corresponding to the independent feature item according to the hierarchical judgment result; Match the abnormal type labels of all independent feature items with the abnormal states in the preset abnormal state mapping table; If more than half of the independent feature items match the same abnormal state, output the abnormal state and use the abnormal state as the diagnosis result.

7. A shield machine abnormal diagnosis system based on maximum information non-parameters, characterized in that, Including: An acquisition module for acquiring the vibration signal of the cutter head bearing in the shield machine during the tunneling process of the shield machine, and determining the spatial energy attenuation pattern of the vibration waveform formed when the vibration signal propagates in the underground complex formation; A decomposition module for performing multi-frequency domain decomposition on the vibration signal to obtain a multi-band energy distribution set; A selection module, configured to select a target abnormal frequency band energy distribution from the set of multi-band energy distributions by using a maximum mutual information criterion, where the maximum mutual information criterion is determined by a maximum information non-parameter; A generation module, configured to generate a collaborative diagnosis feature vector according to the spatial energy attenuation pattern and the target abnormal frequency band energy distribution; An input module, configured to input the collaborative diagnosis feature vector into a preset random forest classifier and output a diagnosis result matching the abnormal tunneling state of the shield machine; The generating the collaborative diagnosis feature vector according to the spatial energy attenuation pattern and the target abnormal frequency band energy distribution includes: Extracting an energy attenuation rate parameter in the propagation direction, an energy difference parameter between adjacent sensor nodes, and an energy distribution symmetry parameter from the spatial energy attenuation pattern; Extracting a frequency band span parameter, an energy concentration parameter, and an energy fluctuation amplitude parameter from the target abnormal frequency band energy distribution; Taking the result of multiplying the energy attenuation rate parameter by the frequency band span parameter as a first weight coefficient, taking the result of multiplying the energy difference parameter by the energy concentration parameter as a second weight coefficient, and taking the result of multiplying the energy distribution symmetry parameter by the energy fluctuation amplitude parameter as a third weight coefficient; Concatenating the first weight coefficient, the second weight coefficient, and the third weight coefficient in a preset order into a first multi-dimensional feature sequence, and adding a ratio parameter of the overlapping area between the spatial energy attenuation pattern and the abnormal frequency band energy distribution in the first multi-dimensional feature sequence to generate a collaborative diagnosis feature vector.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an abnormal diagnosis method for a shield machine based on a maximum information non-parameter as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a computer, it implements an abnormal diagnosis method for a shield machine based on a maximum information non-parameter as described in any one of claims 1 to 6.

Images