Shield machine abnormal data detection method and system based on weighted adaptive coefficient of variation

Abstract

The present invention discloses a method and system for detecting abnormal data of a shield machine based on a weighted adaptive coefficient of variation. The method comprises the following steps: acquiring data from multiple different sensors on the shield machine; preprocessing the sensor data to obtain preprocessed data; obtaining a corresponding volatility index and a coefficient of variation within a window based on the preprocessed data; obtaining an adaptive weight corresponding to the preprocessed data based on the volatility index; obtaining a weighted adaptive coefficient of variation for all preprocessed data based on the coefficient of variation within the window and the adaptive weight; obtaining an abnormality detection threshold based on historical normal data of the shield machine and prior experience; obtaining abnormal data based on a comparison of the weighted adaptive coefficient of variation with the abnormality detection threshold; and cleaning the abnormal data and extracting fault features. The method achieves abnormality detection and efficient cleaning of multi-parameter data of shield machine sensors, thereby improving the accuracy and reliability of shield machine data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of shield machine abnormal data detection, more particularly to a shield machine abnormal data detection method and system based on a weighted adaptive coefficient of variation. BACKGROUND

[0002] As a key equipment in tunnel construction, the running state of a shield machine directly affects the safety and efficiency of the project. However, due to the complex operating environment, the sensor data of the shield machine is often disturbed by noise, outliers and fault signals. The existing technology mainly processes the data by the following methods:

[0003] 1. Traditional data cleaning method: usually uses simple statistical methods such as mean filtering and median filtering for data cleaning; these methods are effective for smooth and low-noise data, but when dealing with nonlinear, non-stationary and multi-dimensional data, it is difficult to maintain the integrity and effectiveness of the data, and it is easy to cause the loss of fault features.

[0004] 2. Threshold method: a fixed threshold is set according to experience to detect abnormal data that exceeds the threshold range; however, the data of the shield machine fluctuates greatly under different construction environments, and the fixed threshold is difficult to adapt to the dynamic changes of the complex environment, which may lead to misjudgment and missed judgment.

[0005] 3. Machine learning-based anomaly detection: some existing technologies introduce machine learning models for fault detection, which can improve the accuracy of detection, but the complexity of the model and the dependence on the training process are strong, making it difficult to apply to online monitoring and early warning in real time.

[0006] The main shortcomings of the above-mentioned existing technologies include: difficulty in adapting to dynamic changes in the construction environment; limitations in processing multi-dimensional data, which cannot fully integrate parameter information; difficulty in balancing real-time performance and accuracy.

[0007] Therefore, how to effectively integrate multi-parameter data of the shield machine and then realize efficient detection and cleaning of abnormal data of the shield machine is a problem that needs to be solved by those skilled in the art. SUMMARY

[0008] Therefore, the present application provides a shield machine abnormal data detection method and system based on a weighted adaptive coefficient of variation, which realizes abnormal detection and efficient cleaning of multi-parameter data of the shield machine sensors, improves the accuracy and reliability of the shield machine data, and provides strong support for real-time fault warning and safe operation of the shield machine.

[0009] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0010] The shield machine abnormal data detection method based on the weighted adaptive coefficient of variation comprises:

[0011] acquiring a plurality of different sensor data on a shield tunneling machine;

[0012] performing preprocessing based on the sensor data to obtain preprocessed data;

[0013] obtaining a corresponding volatility index and a window-in coefficient of variation based on the preprocessed data;

[0014] obtaining an adaptive weight corresponding to the preprocessed data based on the volatility index;

[0015] obtaining a weighted adaptive coefficient of variation of all the preprocessed data based on the window-in coefficient of variation and the adaptive weight;

[0016] obtaining an anomaly detection threshold based on historical normal data of the shield tunneling machine and prior experience;

[0017] obtaining abnormal data based on a comparison between the weighted adaptive coefficient of variation and the anomaly detection threshold;

[0018] performing cleaning based on the abnormal data and extracting fault features.

[0019] Preferably, the sensor data includes shield tunneling machine pressure data, shield tunneling machine temperature data, shield tunneling machine vibration data, and shield tunneling machine current data.

[0020] Preferably, the volatility index acquisition method is:

[0021] obtaining a window-in mean value μ i (t) of the i-th preprocessed data based on the preprocessed data;

[0022] obtaining a volatility index δ i (t) of the i-th preprocessed data at time t based on the window-in mean value μ i (t):

[0023]

[0024] wherein N represents a total window data length, k represents a current window step length, X i' (t-k) represents the i-th preprocessed data at time t-k.

[0025] Preferably, the window-in coefficient of variation acquisition method is:

[0026] obtaining a window-in coefficient of variation CV i (t) corresponding to the i-th preprocessed data based on the volatility index δ i (t) and the window-in mean value μ i (t):

[0027]

[0028] wherein, ε represents a small positive number to prevent the denominator from being zero.

[0029] Preferably, the adaptive weight is specifically:

[0030] w i (t)=α·w i (t-1)+(1-α)·f(δ i (t));

[0031] wherein, w i (t) represents the adaptive weight of the i-th preprocessed data at time t, α represents a smoothing coefficient, w i (t-1) represents the adaptive weight of the i-th preprocessed data at time t-1, and f represents a mapping function.

[0032] Preferably, the abnormality detection threshold obtaining method is:

[0033] a normal weighted adaptive variation coefficient is calculated based on historical normal data of the shield tunneling machine and prior experience;

[0034] a normal weighted adaptive variation coefficient mean μ WACV' is obtained based on the normal weighted adaptive variation coefficient;

[0035] a weighted adaptive variation coefficient standard deviation σ WACV' is obtained based on the normal weighted adaptive variation coefficient mean μ WACV' ;

[0036] an abnormality detection upper threshold θ upper and an abnormality detection lower threshold θ lower are obtained based on μ WACV' and σ WACV' :

[0037] θ upper =μ WACV' +q·σ WACV'

[0038] θ lower =μ WACV' -q·σ WACV'

[0039] wherein, q represents a threshold adjustment coefficient;

[0040] the abnormality detection upper threshold θ upper and the abnormality detection lower threshold θ lower jointly constitute the abnormality detection threshold.

[0041] Preferably, the specific method for obtaining abnormal data is:

[0042] Based on the comparison between the weighted adaptive coefficient of variation and the abnormal detection threshold:

[0043] If the weighted adaptive coefficient of variation is greater than the upper abnormal detection threshold or less than the lower abnormal detection threshold, it is determined as abnormal data; otherwise, it is determined as normal data.

[0044] Preferably, it further includes: based on the proportion of the weighted adaptive coefficient of variation exceeding the abnormal detection threshold, dividing early warning levels:

[0045] First-level early warning: the weighted adaptive coefficient of variation exceeds the upper abnormal detection threshold or the lower abnormal detection threshold by 10-20%;

[0046] Second-level early warning: the weighted adaptive coefficient of variation exceeds the upper abnormal detection threshold or the lower abnormal detection threshold by 20-30%;

[0047] Third-level early warning: the weighted adaptive coefficient of variation exceeds the upper abnormal detection threshold or the lower abnormal detection threshold by more than 30%.

[0048] Preferably, based on the abnormal data, cleaning and extracting fault features are performed, specifically including:

[0049] Based on the abnormal data, interpolation processing is performed to obtain cleaned data;

[0050] Based on the cleaned data, the fault features are extracted;

[0051] The fault features include: time features, parameter features, and statistical features.

[0052] The shield machine abnormal data detection system based on the weighted adaptive coefficient of variation includes: a data acquisition and processing module, a coefficient of variation acquisition module, a threshold acquisition module, a result output module, and a result processing module.

[0053] The data acquisition and processing module is configured to acquire a plurality of different sensor data on the shield machine; based on the sensor data, pre-processing is performed to obtain pre-processed data.

[0054] The coefficient of variation acquisition module is configured to obtain corresponding volatility indicators and window-in coefficients of variation based on the pre-processed data; obtain adaptive weights corresponding to the pre-processed data based on the volatility indicators; and obtain a weighted adaptive coefficient of variation of all the pre-processed data based on the window-in coefficients of variation and the adaptive weights.

[0055] The threshold obtaining module is configured to obtain an anomaly detection threshold based on the weighted adaptive coefficient of variation;

[0056] The result output module is configured to obtain abnormal data based on a comparison between the weighted adaptive coefficient of variation and the anomaly detection threshold.

[0057] The result processing module is configured to clean and extract fault features based on the abnormal data.

[0058] Compared with the prior art, the shield machine abnormal data detection method and system based on a weighted adaptive coefficient of variation has the following beneficial effects:

[0059] 1. Improved anomaly detection accuracy: The weighted adaptive coefficient of variation method is used to comprehensively analyze the multi-parameter data of the shield machine, and the fault sensitivity of each parameter is fully utilized. Through the adaptive weight adjustment mechanism, the algorithm can respond to changes in the running state and environment in real time, improving the accuracy of anomaly detection and reducing the probability of misjudgment and missed judgment.

[0060] 2. Enhanced real-time and robustness: The sliding window is used to dynamically monitor and analyze the data, and the threshold is adjusted in real time to quickly detect abnormal points in the data. The WACV has low computational complexity and is suitable for online real-time monitoring, has strong robustness, and can effectively cope with complex and variable environmental disturbances during the operation of the shield machine.

[0061] 3. Strong multi-parameter data fusion capability: The weighted coefficient of variation method is used to fuse the multi-parameter data of the shield machine, which can effectively process multi-dimensional, nonlinear and non-stationary data, improve the overall effect of data cleaning and fault detection, and realize reasonable evaluation of the volatility of each parameter data through adaptive weight adjustment, improving the ability of multi-parameter data fusion.

[0062] 4. Flexible and effective early warning mechanism: Through the real-time fault early warning mechanism, the system can divide different warning levels according to the change degree of the WACV value, timely trigger the warning, and notify the operator to take appropriate measures to avoid the expansion of the fault. The hierarchical warning provides flexible risk prompts, which helps to ensure the safe operation of the shield machine.

[0063] 5. Data cleaning and fault feature extraction: The present application can not only detect and clean abnormal data, but also extract key fault features, providing a high-quality data basis for subsequent fault diagnosis and analysis, and improving the accuracy and efficiency of fault diagnosis.

[0064] 6. Easy to engineering application: the method has low computational complexity, easy to implement, suitable for deployment and promotion in practical engineering; its real-time online monitoring characteristics make it can be directly applied to the monitoring system of the shield machine, providing a practical solution for the state monitoring and fault warning of engineering equipment.

[0065] 7. The application has significant advantages in improving the accuracy, real-time and reliability of shield machine operation data, and can effectively solve the shortcomings of existing technology in data cleaning and anomaly detection, providing strong support for the safe operation and maintenance management of shield machines. BRIEF DESCRIPTION OF DRAWINGS

[0066] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of the provided drawings.

[0067] Figure 1 The flow chart of the shield machine abnormal data detection method based on the weighted adaptive coefficient of variation provided by the present application.

[0068] Figure 2 The flow chart of the abnormal detection threshold acquisition method provided by the present application.

[0069] Figure 3 The structure schematic diagram of the shield machine abnormal data detection system based on the weighted adaptive coefficient of variation provided by the present application.

[0070] Figure 4 The structure schematic diagram of the shield machine abnormal data detection system based on the weighted adaptive coefficient of variation provided by the present application.

[0071] Figure 5 The structure block diagram of the computer equipment provided by the present application. DETAILED DESCRIPTION

[0072] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0073] Embodiment 1

[0074] As Figure 1As shown, the embodiment of the application discloses a shield machine abnormal data detection method based on weighted adaptive coefficient of variation, comprising:

[0075] Obtain a plurality of different sensor data on the shield machine;

[0076] Preprocess based on sensor data to obtain preprocessed data;

[0077] Based on the preprocessed data, the corresponding volatility index and the window coefficient of variation are obtained;

[0078] Based on the volatility index, the adaptive weight corresponding to the preprocessed data is obtained;

[0079] Based on the window coefficient of variation and the adaptive weight, the weighted adaptive coefficient of variation of all preprocessed data is obtained;

[0080] Based on the historical normal data of the shield machine and the prior experience, an abnormal detection threshold is obtained;

[0081] Based on the comparison between the weighted adaptive coefficient of variation and the abnormal detection threshold, the abnormal data is obtained;

[0082] Based on the abnormal data, cleaning and fault feature extraction are performed.

[0083] Embodiment 2

[0084] The embodiment of the application discloses a shield machine abnormal data detection method based on weighted adaptive coefficient of variation, comprising:

[0085] Obtain a plurality of different sensor data on the shield machine:

[0086] Preferably, m different sensors are arranged on the shield machine to be detected, and the data collected by each sensor at time t is X i (t), wherein i=1, 2,..., m represents different sensors, t represents time, and reflects the physical quantity of the shield machine during operation.

[0087] Preferably, in this embodiment, pressure sensors, temperature sensors, vibration sensors and current sensors are arranged to collect shield machine pressure data, shield machine temperature data, shield machine vibration data and shield machine current data, which are used for subsequent analysis and processing.

[0088] Preprocess based on sensor data to obtain preprocessed data:

[0089] Preferably, data normalization is used for preprocessing, so as to eliminate the influence of different sensor data magnitudes.

[0090] Preferably, the minimum-maximum normalization method is used in this embodiment:

[0091]

[0092] wherein X i' (t) represents normalized sensor data, i.e. preprocessed data, X min,i and X max,i represent the minimum and maximum values of X i (t), respectively.

[0093] Based on the preprocessed data, a corresponding volatility indicator and a window- wise coefficient of variation are obtained:

[0094] Preferably, the volatility indicator obtaining method is:

[0095] Based on the preprocessed data, a window-wise mean value μ i (t) of the ith preprocessed data is obtained:

[0096] Based on the window-wise mean value μ i (t), a volatility indicator δ i (t) of the ith preprocessed data at time t is obtained:

[0097]

[0098] wherein N represents the total window data length in the sliding window, k represents the current window step length, X i' (t-k) represents the ith preprocessed data at time t-k.

[0099] Preferably, the window-wise mean value μ i (t) of the ith preprocessed data is:

[0100]

[0101] Preferably, the window-wise coefficient of variation obtaining method is:

[0102] Based on the volatility indicator δ i (t) and the window-wise mean value μ i (t), a window-wise coefficient of variation CV i (t) corresponding to the ith preprocessed data is obtained:

[0103]

[0104] wherein represents a small positive number to prevent the denominator from being zero.

[0105] Preferably, the window-wise mean value μ i (t) and the volatility indicator δ i (t) correspond to the central tendency and dispersion degree of the data.

[0106] The adaptive weight corresponding to the pretreatment data is obtained based on the volatility index:

[0107] Preferably, the initial weight w i (0) of each pretreatment data is set

[0108]

[0109] The initial weight w i (0) is a basic value of the set weight, which reflects the initial sensitivity of the i-th pretreatment data X i to fault detection, and is initially assigned based on experience or historical data. Over time, the weight is dynamically adjusted through a smoothing adjustment formula.

[0110] Preferably, the adaptive weight is specifically:

[0111] w i (t)=α·w i (t-1)+(1-α)·f(δ i (t));

[0112] wherein w i (t) represents the adaptive weight of the i-th pretreatment data at time t, α represents a smoothing coefficient for controlling the speed of adaptive adjustment, 0<α<1, w i (t-1) represents the adaptive weight of the i-th pretreatment data at time t-1, and f represents a mapping function.

[0113] Preferably, the present application realizes adaptive adjustment of the weight through the smoothing coefficient α and the volatility index δ i (t), so that the algorithm can respond to changes in the running state and the environment in real time, improve the accuracy of anomaly detection, and reduce the probability of misjudgment and missed judgment; through adaptive adjustment of the weight, reasonable evaluation of the volatility of each parameter data is realized, and the ability of multi-parameter data fusion is improved.

[0114] The weighted adaptive coefficient of variation of all pretreatment data is obtained based on the coefficient of variation within the window and the adaptive weight:

[0115] Preferably, the weighted adaptive coefficient of variation WACV(t) of all pretreatment data is obtained by fusing the coefficient of variation CV i (t) within the window corresponding to all pretreatment data and the adaptive weight w i (t):

[0116]

[0117] Preferably, different coefficients of variation CV i (t) within the window corresponding to multiple sensor data are fused to obtain the weighted adaptive coefficient of variation WACV(t) of all pretreatment data.(t) The fusion of multi-parameter data of the shield machine is realized, multi-dimensional, nonlinear and non-stationary data can be effectively processed, and the overall effect of data cleaning and fault detection is improved.

[0118] Preferably, the weighted adaptive coefficient of variation WACV(t) comprehensively reflects the overall volatility of the multi-parameter data of the shield machine, and is used for anomaly detection.

[0119] The anomaly detection threshold is obtained based on the historical normal data of the shield machine:

[0120] Preferably, as shown in Figure 2 , the anomaly detection threshold acquisition method is:

[0121] The normal weighted adaptive coefficient of variation WACV'(t) is calculated based on the historical normal data of the shield machine and prior experience;

[0122] The normal weighted adaptive coefficient of variation mean μ WACV' is obtained based on the normal weighted adaptive coefficient of variation WACV'(t);

[0123] The normal weighted adaptive coefficient of variation standard deviation σ WACV' is obtained based on the normal weighted adaptive coefficient of variation mean μ WACV' ;

[0124] The anomaly detection upper threshold θ upper and the anomaly detection lower threshold θ lower are obtained based on μ WACV' and σ WACV' :

[0125] θ upper = μ WACV' + q· σ WACV'

[0126] θ lower = μ WACV' - q· σ WACV'

[0127] Wherein, q represents a threshold adjustment coefficient;

[0128] The anomaly detection upper threshold θ upper and the anomaly detection lower threshold θ lower together constitute the anomaly detection threshold.

[0129] Preferably, the normal weighted adaptive coefficient of variation mean μ WACV' is:

[0130]

[0131] Wherein, T represents the total window length, i.e. the total step length.

[0132] The normal weighted adaptive coefficient of variation standard deviation σ is preferably updated in real time WACV' is:

[0133]

[0134] The normal weighted adaptive coefficient of variation standard deviation σ is preferably updated in real time WACV' and σ WACV' to achieve dynamic adjustment of the anomaly detection threshold.

[0135] The application uses a sliding window to dynamically monitor and analyze data, and combines a real-time adjusted threshold to quickly detect abnormal points in the data.

[0136] Based on the comparison between the weighted adaptive coefficient of variation and the anomaly detection threshold, abnormal data is obtained:

[0137] The specific method for obtaining abnormal data is preferably:

[0138] Based on the comparison between the weighted adaptive coefficient of variation and the anomaly detection threshold:

[0139] If the weighted adaptive coefficient of variation WACV(t) is greater than the upper anomaly detection threshold θ upper or less than the lower anomaly detection threshold θ lower , it is determined to be abnormal data; otherwise, it is determined to be normal data.

[0140] The application also preferably includes: based on the proportion of the weighted adaptive coefficient of variation exceeding the anomaly detection threshold, dividing the warning levels:

[0141] First-level warning: the weighted adaptive coefficient of variation WACV(t) is greater than the upper anomaly detection threshold θ upper or less than the lower anomaly detection threshold θ lower , exceeding 10-20%;

[0142] Second-level warning: the weighted adaptive coefficient of variation WACV(t) is greater than the upper anomaly detection threshold θ upper or less than the lower anomaly detection threshold θ lower , exceeding 20-30%;

[0143] Third-level warning: the weighted adaptive coefficient of variation WACV(t) is greater than the upper anomaly detection threshold θ upper or less than the lower anomaly detection threshold θ lower , exceeding 30% or more.

[0144] The application preferably includes a warning processing: triggering a warning mechanism, recording abnormal information, and taking corresponding measures.

[0145] Preferably, through the real-time fault early warning mechanism, the system can divide different early warning levels according to the degree of change of the WACV value, timely trigger early warning, and notify the operator to take corresponding measures to avoid the expansion of the fault. The hierarchical early warning provides flexible risk prompts, which helps to ensure the safe operation of the shield machine.

[0146] Based on the abnormal data, the cleaning and extraction of fault features are carried out.

[0147] Preferably, based on the abnormal data, the interpolation processing is carried out to obtain the cleaned data

[0148]

[0149] X i” (t-1) represents the abnormal data at t-1 time, X i” (t+1) represents the abnormal data at t+1 time.

[0150] Preferably, based on the cleaned data, the fault features are extracted.

[0151] The fault features include time features, parameter features and statistical features.

[0152] Preferably, the time features include the frequency and duration of abnormal occurrence; the parameter features include the change rate of each parameter during the abnormality; and the statistical features include the WACV mean and WACV standard deviation.

[0153] For the detected abnormal data, the system uses interpolation and other methods for cleaning to ensure the continuity of the data. The cleaned data is used to extract fault features, including time features and parameter features, to provide high-quality data for subsequent fault diagnosis.

[0154] Embodiment 3

[0155] As shown in Figure 3 , the shield machine abnormal data detection system based on the weighted adaptive coefficient of variation includes a data acquisition and processing module, a coefficient of variation acquisition module, a threshold acquisition module, a result output module and a result processing module.

[0156] The data acquisition and processing module is used to acquire a plurality of different sensor data on the shield machine; based on the sensor data, the pre-processing data is obtained.

[0157] The coefficient of variation acquisition module is used to obtain the corresponding volatility index and window coefficient of variation based on the pre-processing data; based on the volatility index, the adaptive weight corresponding to the pre-processing data is obtained; based on the window coefficient of variation and the adaptive weight, the weighted adaptive coefficient of variation of all pre-processing data is obtained.

[0158] a threshold obtaining module configured to obtain an anomaly detection threshold based on the weighted adaptive coefficient of variation;

[0159] a result output module configured to obtain abnormal data based on a comparison between the weighted adaptive coefficient of variation and the anomaly detection threshold;

[0160] a result processing module configured to clean and extract fault features based on the abnormal data.

[0161] Preferably, each module in the system of the present application can correspondingly implement the above method.

[0162] Embodiment 4

[0163] As shown in the figure, the shield machine abnormal data detection system based on the weighted adaptive coefficient of variation comprises a data acquisition and processing module, a coefficient of variation obtaining module, a threshold obtaining module, a result output module and a result processing module. Figure 4

[0164] The data acquisition and processing module is configured to obtain a plurality of different sensor data on the shield machine, and to obtain preprocessed data based on the preprocessing of the sensor data.

[0165] The coefficient of variation obtaining module is configured to obtain corresponding volatility indicators and window-in coefficients of variation based on the preprocessed data, to obtain adaptive weights corresponding to the preprocessed data based on the volatility indicators, and to obtain weighted adaptive coefficients of variation of all the preprocessed data based on the window-in coefficients of variation and the adaptive weights.

[0166] The threshold obtaining module is configured to obtain an anomaly detection threshold based on the weighted adaptive coefficient of variation.

[0167] The result output module is configured to obtain abnormal data based on a comparison between the weighted adaptive coefficient of variation and the anomaly detection threshold.

[0168] The result processing module is configured to clean and extract fault features based on the abnormal data.

[0169] Preferably, each module in the system of the present application can correspondingly implement the above method.

[0170] Preferably, the system further comprises a hierarchical early warning module.

[0171] The hierarchical early warning module is configured to divide early warning levels based on a proportion of the weighted adaptive coefficient of variation exceeding the anomaly detection threshold.

[0172] First-level early warning: the weighted adaptive coefficient of variation WACV(t) is greater than the upper anomaly detection threshold θ upper or less than the lower anomaly detection threshold θ lower by 10-20%;

[0173] ​Second warning: the weighted adaptive coefficient of variation WACV(t) is greater than the upper threshold of anomaly detection upper or less than the lower threshold of anomaly detection lower more than 20-30%;

[0174] Third warning: the weighted adaptive coefficient of variation WACV(t) is greater than the upper threshold of anomaly detection upper or less than the lower threshold of anomaly detection lower more than 30%.

[0175] Warning processing: trigger the warning mechanism, record abnormal information, and take corresponding measures.

[0176] Embodiment 5

[0177] Based on the same inventive concept, the application also provides a computer device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other through the communication bus;

[0178] The memory is used to store a computer program.

[0179] The processor is used to execute the program stored on the memory, and can realize the shield machine abnormal data detection method based on the weighted adaptive coefficient of variation in Embodiments 1 or 2.

[0180] As shown in Figure 5 The electronic device can include a processor 41, a communication interface 42, a memory 43 and a communication bus 44, wherein the processor 41, the communication interface 42 and the memory 43 complete communication with each other through the communication bus 44. The processor 41 can call the logical instructions in the memory 43 to execute the shield machine abnormal data detection method based on the weighted adaptive coefficient of variation in Embodiments 1 or 2.

[0181] In addition, the logic instructions in the memory 43 described above can be realized in the form of a software function unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the parts that contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0182] Via the technical solutions described above, the present application discloses a shield machine abnormal data detection method and system based on a weighted adaptive coefficient of variation, which has the following beneficial effects:

[0183] 1. Improve the accuracy of abnormal detection: through the weighted adaptive coefficient of variation method, the multi-parameter data of the shield machine is comprehensively analyzed, and the fault sensitivity of each parameter is fully utilized; through the adaptive weight adjustment mechanism, the algorithm can respond to the changes of the running state and the environment in real time, improve the accuracy of abnormal detection, and reduce the probability of misjudgment and omission.

[0184] 2. Enhance real-time and robustness: the present application uses a sliding window to dynamically monitor and analyze data, and combines a real-time adjusted threshold to quickly detect abnormal points in the data; the WACV has low computational complexity and is suitable for online real-time monitoring, has strong robustness, and can effectively cope with complex and variable environmental interference in the running process of the shield machine.

[0185] 3. Strong multi-parameter data fusion capability: the present application uses the weighted coefficient of variation method to fuse the multi-parameter data of the shield machine, which can effectively process multi-dimensional, nonlinear and non-stationary data, improve the overall effect of data cleaning and fault detection; through adaptive adjustment of the weight, reasonable evaluation of the volatility of each parameter data is realized, and the ability of multi-parameter data fusion is improved.

[0186] 4. Flexible and effective early warning mechanism: through the real-time fault early warning mechanism, the system can divide different warning levels according to the change degree of the WACV value, trigger the warning in time, notify the operator to take corresponding measures, and avoid the expansion of the fault; the hierarchical warning provides flexible risk prompt, which helps to ensure the safe operation of the shield machine.

[0187] 5、Data cleaning and fault feature extraction: The application can not only detect and clean abnormal data, but also extract key fault features, providing high-quality data basis for subsequent fault diagnosis and analysis, and improving the accuracy and efficiency of fault diagnosis.

[0188] 6、Easy to engineering application: The method has low computational complexity and is easy to implement, suitable for deployment and promotion in actual engineering; Its real-time online monitoring characteristics make it can be directly applied to the monitoring system of the shield machine, providing a practical solution for the state monitoring and fault warning of engineering equipment.

[0189] 7、The application has significant advantages in improving the accuracy, real-time and reliability of shield machine operation data, and can effectively solve the shortcomings of existing technologies in data cleaning and anomaly detection, providing strong support for the safe operation and maintenance management of shield machines.

[0190] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. For the device disclosed by the embodiments, since it corresponds to the method disclosed by the embodiments, the description is relatively simple, and the related parts can be referred to the method part.

[0191] The above description of the disclosed embodiments enables those skilled in the art to implement or use the application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the application. Therefore, the application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A shield machine abnormal data detection method based on weighted adaptive variation coefficient is characterized by: include: Acquire data from multiple sensors on the shield machine; Performing preprocessing based on the sensor data to obtain preprocessed data; Obtaining corresponding volatility index and coefficient of variation within a window based on the preprocessed data; The volatility index is obtained as follows: Based on the preprocessed data, the mean μ in the window of the i-th preprocessed data is obtained. i (t); Based on the mean μ within the window i (t) Get the volatility index δ of the i-th preprocessed data at time t i (t): Among them, N represents the total window data length in the sliding window, k represents the current window step, and X i′ (tk) represents the i-th preprocessed data at time tk; The method for obtaining the coefficient of variation within the window is: Based on the volatility indicator δ i (t) and the mean μ in the window i (t) Obtain the coefficient of variation CV within the window corresponding to the i-th preprocessed data i (t): in, Represents a small positive number to prevent the denominator from being zero; Obtaining an adaptive weight corresponding to the preprocessed data based on the volatility index; The adaptive weight is specifically: w i (t)=α·w i (t-1)+(1-α)·f(δ i (t)); Among them, w i (t) represents the adaptive weight of the i-th preprocessed data at time t, α represents the smoothing coefficient, w i (t-1) represents the adaptive weight of the i-th preprocessed data at time t-1, and f represents the mapping function; Obtaining a weighted adaptive coefficient of variation of all the preprocessed data based on the coefficient of variation within the window and the adaptive weight; The anomaly detection threshold is obtained based on the shield machine's historical normal data and prior experience; The method for obtaining the anomaly detection threshold is: The normal weighted adaptive variation coefficient is calculated based on the historical normal data of the shield machine and prior experience; Based on the normal weighted adaptive variation coefficient, a normal weighted adaptive variation coefficient mean μ is obtained. WACV′ ; Based on the normal weighted adaptive coefficient of variation mean μ WACV′ Get the standard deviation of the weighted adaptive coefficient of variation σ WACV′ ; Based on μ WACV′ and σ WACV′ Get the upper threshold θ for anomaly detection upper and the anomaly detection threshold θ lower : i upper =μ WACV′ +q·s WACV′ i lower =μ WACV′ -q·s WACV′ Wherein, q represents the threshold adjustment coefficient; The abnormality detection upper threshold θ upper and the anomaly detection lower threshold θ lower together constitute the anomaly detection threshold; Obtaining abnormal data based on a comparison between the weighted adaptive coefficient of variation and the abnormality detection threshold; Cleaning is performed based on the abnormal data and fault features are extracted.

2. The shield machine abnormal data detection method based on weighted adaptive variation coefficient according to claim 1 is characterized in that: The sensor data includes: shield machine pressure data, shield machine temperature data, shield machine vibration data and shield machine current data.

3. The shield machine abnormal data detection method based on weighted adaptive variation coefficient according to claim 1 is characterized in that: The specific method to obtain abnormal data is: Based on the comparison between the weighted adaptive coefficient of variation and the anomaly detection threshold: If the weighted adaptive variation coefficient is greater than the anomaly detection upper threshold or less than the anomaly detection lower threshold, it is determined to be the abnormal data; otherwise, it is determined to be normal data.

4. The shield machine abnormal data detection method based on weighted adaptive variation coefficient according to claim 3 is characterized in that: Also includes: Based on the proportion of the weighted adaptive coefficient of variation exceeding the anomaly detection threshold, the warning level is divided: Level 1 warning: The weighted adaptive coefficient of variation is greater than the upper threshold of abnormality detection or less than the lower threshold of abnormality detection by more than 10-20%; Level 2 warning: The weighted adaptive coefficient of variation is greater than the upper threshold of abnormality detection or less than the lower threshold of abnormality detection by more than 20-30%; Level 3 warning: The weighted adaptive variation coefficient is greater than the upper threshold value of abnormality detection or less than the lower threshold value of abnormality detection by more than 30%.

5. The shield machine abnormal data detection method based on weighted adaptive variation coefficient according to claim 1 is characterized in that: Cleaning and extracting fault features based on the abnormal data specifically includes: Performing interpolation processing based on the abnormal data to obtain cleaned data; extracting the fault feature based on the cleaning data; The fault characteristics include: time characteristics, parameter characteristics and statistical characteristics.

6. A shield machine abnormal data detection system based on weighted adaptive coefficient of variation, applied to a shield machine abnormal data detection method based on weighted adaptive coefficient of variation as claimed in any one of claims 1 to 5, characterized in that: include: Data acquisition and processing module, coefficient of variation acquisition module, threshold acquisition module, result output module and result processing module; The data acquisition and processing module is used to obtain data from multiple different sensors on the shield machine; Performing preprocessing based on the sensor data to obtain preprocessed data; The coefficient of variation acquisition module is used to obtain the corresponding volatility index and the coefficient of variation within the window based on the preprocessed data; Obtaining an adaptive weight corresponding to the preprocessed data based on the volatility index; Obtaining a weighted adaptive coefficient of variation of all the preprocessed data based on the coefficient of variation within the window and the adaptive weight; The threshold acquisition module is used to obtain an anomaly detection threshold based on the weighted adaptive coefficient of variation; The result output module is configured to obtain abnormal data based on the comparison between the weighted adaptive coefficient of variation and the abnormality detection threshold; The result processing module is used to clean and extract fault features based on the abnormal data.

Images