CN-122020379-A - Intelligent fault diagnosis method and system for main driving device of shield tunneling machine

CN122020379ACN 122020379 ACN122020379 ACN 122020379ACN-122020379-A

Abstract

The invention discloses an intelligent fault diagnosis method and system for a main driving device of a shield machine, which comprise the steps of obtaining multi-source monitoring information of the main driving device of the shield machine, extracting characteristic indexes based on the multi-source monitoring information, obtaining corresponding multi-source characteristics through normalization processing, carrying out characteristic fusion based on the multi-source characteristics to obtain fusion characteristics, inputting the fusion characteristics into a trained deep confidence network to obtain fault types.

Inventors

SHI YAN
YANG ZHIGUO
SONG HUAN
LI LEI
You Shaoqiang
SHEN XIANGKAI
WANG SHUHAO
LIU YONGCHAO
CHANG QIANG

Assignees

北京航空航天大学

Dates

Publication Date: 20260512
Application Date: 20260128

Claims (10)

1. The intelligent fault diagnosis method for the main driving device of the shield tunneling machine is characterized by comprising the following steps of: acquiring multi-source monitoring information of a main driving device of the shield machine; Extracting characteristic indexes based on the multi-source monitoring information, and obtaining corresponding multi-source characteristics through normalization processing; performing feature fusion through self-adaptive weights based on the multi-source features to obtain fusion features; and inputting the fusion characteristics into a trained deep confidence network to obtain a diagnosis result.
2. The intelligent diagnosis method for faults of a main driving system of a shield machine according to claim 1, further comprising the step of performing secondary assessment on the confidence level of a diagnosis result through fuzzy Bayesian inference.
3. The intelligent diagnosis method for faults of a main driving system of a shield machine according to claim 2, wherein the step of fuzzy Bayesian reasoning comprises the following steps: calculating fuzzy membership according to a preset membership function and the fault probability distribution in the diagnosis result; and constructing a Bayesian confidence matrix according to the prior occurrence probability of various faults, and obtaining a corrected confidence value through the joint calculation of the fuzzy membership and the prior probability.
4. The intelligent diagnosis method for faults of a main driving device of a shield machine according to claim 1, wherein a multi-source feature fusion model is established for the feature fusion, and the multi-source feature fusion model is as follows: Wherein, the 、 And For different adaptive weight coefficients, As a feature of the hydraulic pressure signal, As a characteristic of the current signal, Is a vibration signal characteristic.
5. The intelligent diagnosis method for faults of the main driving device of the shield machine according to claim 1 or 4, wherein the feature index extraction comprises the following steps: wherein X is a characteristic index, The mean value is represented as such, Is the standard deviation; Characterizing the overall energy level of the signal as a root mean square value; The kurtosis is used for representing the steepness degree of the distribution form of the signal waveform; representing wavelet packet energy, reflecting the energy distribution of the signal in a specific frequency band; the peak frequency is used to represent the frequency component with the highest energy in the spectrum.
6. A fault intelligent diagnosis system for a shield tunneling machine main driving device is characterized by comprising: the data acquisition module is used for acquiring multi-source monitoring information of the main driving device of the shield machine; the feature extraction module is used for extracting feature indexes based on the multi-source monitoring information and obtaining corresponding multi-source features through normalization processing; The feature fusion module is used for carrying out feature fusion based on the multisource features to obtain fusion features, and adopting an improved weighted principal component analysis and multiservice feature fusion model to uniformly map feature vectors from hydraulic, electric and vibration signals to a low-dimensional feature space; And the fault identification module is used for confirming the fault category according to the fusion characteristic through the trained deep confidence network.
7. The intelligent fault diagnosis system for a main driving device of a shield machine according to claim 6, wherein the data acquisition module comprises a pressure sensor, a flow sensor, a vibration acceleration sensor and a main motor current.
8. The intelligent fault diagnosis system for the shield tunneling machine main driving device according to claim 6 is characterized by further comprising a fuzzy reasoning module, wherein the fuzzy reasoning module is used for carrying out secondary assessment on the confidence of the diagnosis result through fuzzy Bayesian reasoning.
9. The intelligent fault diagnosis system for a main driving device of a shield tunneling machine according to claim 6, wherein said fuzzy inference module comprises: The reasoning configuration sub-module is used for setting the fuzzy grades and not configuring corresponding membership functions for each fuzzy grade, acquiring prior occurrence probability of various faults and constructing a Bayesian confidence matrix; the confidence correction sub-module is used for receiving the diagnosis result and confirming fuzzy membership according to the membership function, and is used for calculating a corrected confidence value according to the fuzzy membership and the Bayesian confidence matrix.
10. The intelligent fault diagnosis system for the main driving device of the shield machine according to claim 6, further comprising an early warning module, wherein the early warning module is used for judging the confidence level of the fault class, and early warning is performed when the confidence level exceeds a preset threshold value.

Description

Intelligent fault diagnosis method and system for main driving device of shield tunneling machine Technical Field The invention relates to the technical field of fault detection, in particular to an intelligent fault diagnosis method and system for a main driving device of a shield machine. Background In the field of state monitoring and fault diagnosis of a main driving system of a shield machine, the prior art mainly faces the dilemma of single monitoring means and insufficient diagnosis reliability. The traditional diagnosis scheme usually only relies on single type signals such as vibration, current or hydraulic pressure to analyze, and because the shield machine has complex and severe working conditions, the single signals are easy to interfere and difficult to comprehensively reflect the health state of the system, and false alarm or missing alarm is extremely easy to cause. Although some technologies attempt to collect multi-source signals, most stay on the level of simple data superposition or threshold comparison, and lack the capability of effectively unifying and deeply fusing heterogeneous signals such as hydraulic signals, electric signals and vibration signals, so that complex mapping relations between faults and multi-source information cannot be deeply mined. In addition, the existing diagnosis method based on the traditional machine learning model has limited characterization capability on complex nonlinear faults, the performance of the method highly depends on complete historical fault data, and sufficient samples are difficult to obtain in practical application, so that the generalization capability of the model and the reasoning capability on unknown faults are insufficient, and the accuracy of diagnosis and early warning efficiency are restricted. Therefore, how to improve the accuracy of fault detection and implement early warning is a problem that needs to be solved by those skilled in the art. Disclosure of Invention In view of the above problems, the present invention is provided to provide a method and a system for intelligent diagnosis of faults of a main driving device of a shield machine, which overcome or at least partially solve the above problems. In order to achieve the above purpose, the present invention adopts the following technical scheme: a fault intelligent diagnosis method for a main driving device of a shield tunneling machine comprises the following steps: acquiring multi-source monitoring information of a main driving device of the shield machine; Extracting characteristic indexes based on the multi-source monitoring information, and obtaining corresponding multi-source characteristics through normalization processing; performing feature fusion through self-adaptive weights based on the multi-source features to obtain fusion features; and inputting the fusion characteristics into a trained deep confidence network to obtain a diagnosis result. Preferably, the method further comprises the step of performing secondary evaluation on the confidence level of the diagnosis result through fuzzy Bayesian inference. Preferably, the step of fuzzy bayesian reasoning includes: calculating fuzzy membership according to a preset membership function and the fault probability distribution in the diagnosis result; and constructing a Bayesian confidence matrix according to the prior occurrence probability of various faults, and obtaining a corrected confidence value through the joint calculation of the fuzzy membership and the prior probability. Preferably, a multi-source feature fusion model is established to perform the feature fusion, and the multi-source feature fusion model is as follows: Wherein, the 、AndFor different adaptive weight coefficients,As a feature of the hydraulic pressure signal,As a characteristic of the current signal,Is a vibration signal characteristic. Preferably, the extracting the characteristic index includes: wherein X is a characteristic index, The mean value is represented as such,Is the standard deviation; Characterizing the overall energy level of the signal as a root mean square value; The kurtosis is used for representing the steepness degree of the distribution form of the signal waveform; representing wavelet packet energy, reflecting the energy distribution of the signal in a specific frequency band; the peak frequency is used to represent the frequency component with the highest energy in the spectrum. A fault intelligent diagnosis system for a main driving device of a shield tunneling machine comprises: the data acquisition module is used for acquiring multi-source monitoring information of the main driving device of the shield machine; the feature extraction module is used for extracting feature indexes based on the multi-source monitoring information and obtaining corresponding multi-source features through normalization processing; The feature fusion module is used for carrying out feature fusion based on the multisource features to obt