CN117969092A - Fault detection method, equipment and medium for main bearing of shield tunneling machine - Google Patents
Fault detection method, equipment and medium for main bearing of shield tunneling machine Download PDFInfo
- Publication number
- CN117969092A CN117969092A CN202410370862.7A CN202410370862A CN117969092A CN 117969092 A CN117969092 A CN 117969092A CN 202410370862 A CN202410370862 A CN 202410370862A CN 117969092 A CN117969092 A CN 117969092A
- Authority
- CN
- China
- Prior art keywords
- main bearing
- analyzed
- vibration signal
- fault
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 34
- 230000005641 tunneling Effects 0.000 title claims description 29
- 239000013598 vector Substances 0.000 claims abstract description 62
- 238000000034 method Methods 0.000 claims abstract description 53
- 238000012706 support-vector machine Methods 0.000 claims abstract description 29
- 238000004458 analytical method Methods 0.000 claims abstract description 25
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000012423 maintenance Methods 0.000 claims description 25
- 238000006073 displacement reaction Methods 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 10
- 238000004088 simulation Methods 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 9
- 238000000354 decomposition reaction Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 239000007788 liquid Substances 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 7
- 239000006061 abrasive grain Substances 0.000 claims description 6
- 239000012535 impurity Substances 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 230000000739 chaotic effect Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 239000012530 fluid Substances 0.000 claims description 2
- 239000000306 component Substances 0.000 description 73
- 239000003921 oil Substances 0.000 description 25
- 230000008569 process Effects 0.000 description 17
- 230000002159 abnormal effect Effects 0.000 description 3
- 239000010687 lubricating oil Substances 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005094 computer simulation Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000005299 abrasion Methods 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003638 chemical reducing agent Substances 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 208000008918 voyeurism Diseases 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Acoustics & Sound (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The embodiment of the specification discloses a fault detection method, equipment and medium for a main bearing of a shield machine, which are applied to the technical field of shield machine detection and are used for solving the problems that the existing detection mode is difficult to detect in real time and has low reliability. The method comprises the following steps: collecting main bearing operation data of the shield machine to be detected based on a preset sensor; acquiring historical main bearing operation data corresponding to the main bearing operation data, and extracting a first feature vector of the main bearing according to the historical main bearing operation data and the main bearing operation data; integrating the vibration signals acquired by the preset sensors to obtain vibration signals to be analyzed, and preprocessing the vibration signals to be analyzed to obtain current vibration signals to be analyzed; performing time-frequency domain analysis on the current vibration signal to be analyzed to extract a second characteristic vector of the current vibration signal to be analyzed; inputting the first characteristic vector and the second characteristic vector into a preset support vector machine classifier to locate main bearing fault parts of the shield machine to be detected in real time.
Description
Technical Field
The specification relates to the field of shield tunneling machine detection, in particular to a fault detection method, equipment and medium for a main bearing of a shield tunneling machine.
Background
The shield tunneling machine is short for shield tunneling machine, and is a special engineering machine for tunneling. The shield tunneling machine integrates light, mechanical, electric, liquid, sensing and information technologies, has the functions of excavating and cutting soil body, conveying soil slag, assembling tunnel lining, measuring, guiding and correcting deviation and the like, relates to multi-discipline technologies such as geology, civil engineering, machinery, mechanics, hydraulic pressure, electric, control, measurement and the like, and is designed and manufactured correspondingly according to different geology, so that the reliability requirement is extremely high. The main bearing is used as a key core component of the shield machine, continuously bears huge axial force, radial force and overturning moment in the tunneling process of the shield machine, and can also bear impact load transmitted by the cutter head. Therefore, in order to ensure the reliability and the service life of the main bearing and further ensure the reliable operation of the shield machine, the method is very important for fault detection in the operation process of the main bearing of the shield machine.
Because the main bearing of the shield machine is arranged in the main drive of the shield machine, the structure is closed, and the existing judgment of the health condition of the main bearing of the shield machine is that a serviceman checks the condition in the bearing through a peeping window arranged on the bearing during the overhaul of the main bearing of the shield machine. Or in the tunneling process, the operator preliminarily judges whether the main bearing fails according to various operation parameters of the shield tunneling machine and personal experience, verifies the main bearing lubricating oil by performing physical and chemical index analysis, removes a group of speed reducers after confirming the failure, leaks out the main bearing, and observes and confirms the failure position and damage condition of the main bearing. The shutdown detection and detection mode based on manual experience is poor in effect and low in efficiency, and particularly the actual health condition of the bearing in operation cannot be reflected in real time, so that excessive maintenance or insufficient maintenance is easily caused.
Disclosure of Invention
In order to solve the technical problems, one or more embodiments of the present disclosure provide a method, an apparatus, and a medium for detecting a fault of a main bearing of a shield machine.
One or more embodiments of the present disclosure adopt the following technical solutions:
one or more embodiments of the present disclosure provide a fault detection method for a main bearing of a shield tunneling machine, where the method includes:
collecting main bearing operation data of the shield machine to be detected based on a plurality of preset sensors; wherein the main bearing operation data includes: oil data, vibration data and axial displacement data;
acquiring historical main bearing operation data corresponding to the main bearing operation data, and extracting a first feature vector of the main bearing according to the historical main bearing operation data, the oil liquid data and the axial displacement data; wherein the first feature vector corresponds to the main bearing ring gear feature;
Integrating the vibration signals acquired by the preset sensors to obtain vibration signals to be analyzed, and preprocessing the vibration signals to be analyzed to obtain current vibration signals to be analyzed;
Performing time-frequency domain analysis on the current vibration signal to be analyzed to extract a second feature vector of the current vibration signal to be analyzed; the second feature vector corresponds to the internal and external gear ring features of the main bearing;
Inputting the first characteristic vector and the second characteristic vector into a preset support vector machine classifier to position the main bearing fault part of the shield machine to be detected.
Optionally, in one or more embodiments of the present disclosure, before collecting main bearing operation data of the shield tunneling machine to be detected based on the plurality of preset sensors, the method further includes:
According to the equipment model of the shield machine to be detected, obtaining structural information corresponding to the shield machine to be detected, inputting the structural information into preset analysis software, and constructing a main bearing dynamics simulation model of the shield machine to be detected;
acquiring a reference shield machine corresponding to the equipment model and the working environment of the shield machine to be detected so as to acquire a fault sample of the reference shield machine;
Determining a main bearing dynamics parameter sample of the shield tunneling machine to be detected based on the main bearing dynamics simulation model and the fault sample, and inputting the main bearing dynamics parameter sample into a preset modeling language to obtain a digital twin model corresponding to the main bearing of the shield tunneling machine to be detected; wherein the main bearing kinetic parameter sample comprises: the bearing is externally connected, the bearing has an inner diameter and a bearing pitch diameter;
determining a fault data transmission path corresponding to each fault sample based on a digital twin model corresponding to each fault sample;
Determining a corresponding sensor type according to the fault type corresponding to the fault sample, and determining a corresponding sensor arrangement direction based on the fault data transmission path;
And determining the layout positions of the preset sensors according to the acquisition range corresponding to each sensor type and the sensor layout direction so as to realize the layout of a plurality of preset sensors.
Optionally, in one or more embodiments of the present disclosure, obtaining historical main bearing operation data corresponding to the main bearing operation data, so as to extract a first feature vector of the main bearing according to the historical main bearing operation data, the oil liquid data and the axial displacement data, and specifically includes:
Determining the acquisition time of the main bearing operation data to call a plurality of historical main bearing operation data of the main bearing of the shield machine to be detected, which correspond to the acquisition time; wherein the historical main bearing operation data comprises: historical oil data and historical vibration data;
Acquiring a first difference value between the oil data and adjacent historical oil data and a second difference value between adjacent historical oil data;
If the fluctuation value of the first difference value and the second difference value is larger than the preset fluctuation value, acquiring axial displacement data corresponding to the inner gear ring of the main bearing, and determining the current working condition posture of the main bearing of the shield machine to be detected according to the axial displacement data;
acquiring oil data corresponding to the variation value, and extracting part data corresponding to the oil data and the current working condition posture as a first feature vector of the main bearing; wherein, the fluid data includes: abrasive grain impurity element data, kinematic viscosity, abrasive grain impurity distribution data.
Optionally, in one or more embodiments of the present disclosure, the integrating the vibration signal collected by each preset sensor to obtain the vibration signal to be analyzed specifically includes:
determining parts corresponding to the preset sensors according to the fault data transmission paths, and grouping the vibration signals based on the corresponding relation between the preset sensors and the parts to obtain vibration signal groups corresponding to the parts;
Acquiring acquisition starting time and acquisition ending time corresponding to different vibration signals in each vibration signal group, and sequencing the acquisition starting time according to the time sequence of the acquisition starting time to obtain an acquisition starting time sequence corresponding to each vibration signal group;
Sorting the collection end time corresponding to different vibration signals in each vibration signal group based on time sequence to obtain collection end time sequence corresponding to each vibration signal group;
Determining a current acquisition time range of each vibration signal group according to the cut-off time of the acquisition starting time sequence and the start time of the acquisition ending time sequence, so as to time align vibration signals in each vibration signal group based on the current acquisition time range and obtain a plurality of vibration signal groups to be combined;
and acquiring vibration signal waveforms in each vibration signal group to be combined, and integrating a plurality of vibration signals to be analyzed based on the union of the vibration signal waveforms.
Optionally, in one or more embodiments of the present disclosure, preprocessing the vibration signal to be analyzed to obtain a current vibration signal to be analyzed specifically includes:
decomposing the vibration signal to be analyzed based on an empirical mode decomposition algorithm to obtain content mode components corresponding to the vibration signal to be analyzed in each layer of frequency range;
calculating a mode probability value corresponding to each connotation mode component, determining a frequency component value of each connotation mode component based on the mode probability value, and determining a first connotation mode component to be filtered and a second connotation mode component not to be filtered according to the probability component value of each connotation mode component;
analyzing the first connotation modal component through a bilinear time-frequency distribution algorithm so as to encode the first connotation modal component into an analysis signal component;
obtaining an instantaneous frequency estimation value of the analytic signal component based on the peak value estimation of the analytic signal component, and filtering based on the instantaneous frequency estimation value to obtain a filtered connotation modal component;
And reconstructing the filtered connotation modal component and the second connotation modal component to obtain a current vibration signal to be analyzed.
Optionally, in one or more embodiments of the present disclosure, performing time-frequency domain analysis on the current vibration signal to be analyzed to extract a second feature vector of the current vibration signal to be analyzed, specifically including:
performing envelope demodulation on the current vibration signal to be analyzed based on Hiber transformation to obtain a low-frequency characteristic signal corresponding to the current vibration signal to be analyzed, so as to extract time domain characteristics and frequency domain characteristics of the low-frequency characteristic signal;
Performing empirical mode decomposition on the current vibration signal to be analyzed to extract characteristic component energy moment and sample entropy characteristics of the current vibration signal to be analyzed, and generating an initial characteristic set corresponding to the current vibration signal to be analyzed according to the time domain characteristics, the frequency domain characteristics, the characteristic component energy moment and the sample entropy;
Determining the similarity of each initial feature in the initial feature set according to Euclidean distance between each initial feature in the initial feature set, constructing an adjacency graph corresponding to the initial feature set based on the similarity, and determining a corresponding adjacency matrix according to the similarity between each output feature in the adjacency graph;
And carrying out structural optimization on the adjacency matrix based on a random gradient algorithm to obtain a low-dimensional feature set corresponding to the initial feature set as a second feature vector of the vibration signal to be analyzed currently.
Optionally, in one or more embodiments of the present disclosure, before inputting the first feature vector and the second feature vector into a preset support vector machine classifier to locate a main bearing fault component of the shield tunneling machine to be detected, the method further includes:
acquiring corresponding sample characteristics according to the model of the main bearing, and dividing the sample characteristics to obtain training sample characteristics and test sample characteristics;
Initializing an initial support vector machine classifier to obtain initial parameters of the initial support vector machine classifier; wherein the initial parameters include: penalty coefficient, kernel parameter, search range;
initializing the population of the initial parameters based on chaotic mapping to obtain initialized population parameters, and determining the initial positions of all individuals in the population based on an inverse mapping mode;
Calculating the fitness value of each individual in the population, taking the initial parameter corresponding to the optimal fitness value as the optimal parameter, updating the initial support vector machine classifier according to the optimal parameter, and training the updated initial support vector machine classifier based on the training sample characteristics and the test sample characteristics to obtain a preset support vector machine meeting the requirements.
Optionally, in one or more embodiments of the present disclosure, after inputting the first feature vector and the second feature vector into a preset support vector machine classifier to locate a main bearing fault component of the shield tunneling machine to be detected, the method further includes:
determining maintenance links corresponding to the main bearing fault parts, and determining a plurality of corresponding maintenance personnel according to the maintenance links;
And determining waiting time corresponding to each maintenance personnel according to the working state of each maintenance personnel, so as to screen the plurality of maintenance personnel based on the waiting time and determine designated maintenance personnel corresponding to the main bearing fault parts.
One or more embodiments of the present disclosure provide a fault detection apparatus for a main bearing of a shield tunneling machine, the apparatus including:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to perform any one of the methods described above.
One or more embodiments of the present specification provide a non-volatile computer storage medium storing computer-executable instructions configured to perform any of the methods described above.
The above-mentioned at least one technical scheme that this description embodiment adopted can reach following beneficial effect:
And extracting a first characteristic vector of the main bearing related to faults in the oil liquid data and the axial displacement data by referring to the historical bearing operation data, so that the problem that in the prior art, when the detection of the inner gear ring is realized only based on displacement and rotation speed detection, the determination of the internal fault part of the inner gear ring of the main bearing is difficult to realize is solved. And the time-frequency domain analysis is carried out on the current vibration signal to be analyzed so as to extract the second characteristic vector of the current vibration signal to be analyzed, thereby realizing the acquisition of the characteristics of the internal gear ring and the external gear ring of the main bearing. By the joint analysis of the first feature vector and the second feature vector, the characteristics of the inner gear ring of the main bearing and the characteristics of the outer gear ring of the main bearing are considered, the comprehensive detection of the main bearing of the shield machine to be detected is realized, the detection accuracy is improved, and the problems of low efficiency and low reliability when fault detection is carried out based on the experience of operators are solved.
Drawings
In order to more clearly illustrate the embodiments of the present description or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some of the embodiments described in the present description, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
Fig. 1 is a schematic flow chart of a fault detection method of a main bearing of a shield machine according to an embodiment of the present disclosure;
fig. 2 is a schematic diagram of an internal structure of a fault detection device of a main bearing of a shield machine according to an embodiment of the present disclosure;
Fig. 3 is a schematic diagram of an internal structure of a nonvolatile storage medium according to an embodiment of the present disclosure.
Detailed Description
The embodiment of the specification provides a fault detection method, equipment and medium for a main bearing of a shield machine.
In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present disclosure.
As shown in fig. 1, the embodiment of the present disclosure provides a fault detection method for a main bearing of a shield tunneling machine. As can be seen from fig. 1, in one or more embodiments of the present disclosure, a method for detecting a fault of a main bearing of a shield machine includes the following steps:
S101: collecting main bearing operation data of the shield machine to be detected based on a plurality of preset sensors; wherein the main bearing operation data includes: oil data, vibration data and axial displacement data.
In order to obtain the running data of the main bearing of the shield machine to be detected, thereby realizing fault detection in the running process of the main bearing of the shield machine and reducing the fault rate of the main bearing, in the embodiment of the specification, the main bearing is detected by a plurality of preset sensors, for example: displacement sensor, acceleration sensor etc. for detect shield constructs quick-witted main bearing operation in-process for example: main bearing operation data such as oil data, vibration data, axial displacement data and the like.
The main bearing of the shield machine is provided with a specific sensor mounting interface, so that the accuracy of sensor detection is improved, and the problem of redundant data overmuch caused by repeated collection of the same type of sensors is avoided. In one or more embodiments of the present disclosure, before collecting main bearing operation data of a shield machine to be detected based on a plurality of preset sensors, the method further includes the following steps:
Firstly, in order to facilitate visual analysis of a shield machine to be detected, in the embodiment of the specification, a main bearing dynamics simulation model of a main bearing dynamics simulation model assumes that the shield machine to be analyzed is a model XYZ shield machine, and structural information of the shield machine needs to be acquired. Then the structural information of the shield machine of the model can be obtained by referring to the equipment manual and technical data, and the structural information comprises parameters such as the external size, the inner diameter and the like of the main bearing. And inputting the structural information into preset analysis software, such as ADAMS software, and constructing a main bearing dynamics simulation model of the shield tunneling machine to be detected. The preset ADAMS software is software for creating a completely parameterized mechanical system geometric model by using an interactive graphic environment, a part library, a constraint library and a force library, and a person skilled in the art clearly how to perform model simulation based on the ADAMS software. After the ADAMS software is used for constructing the dynamic simulation model of the main bearing, the ADAMS software can carry out statics and kinematics analysis according to the input structural information so as to simulate the mechanical behavior of the main bearing during working. Next, a reference shield machine corresponding to the equipment model and the working environment of the shield machine to be detected needs to be acquired to acquire a fault sample thereof. The reference shield machine should have a similar structure and working environment to the shield machine to be detected. And extracting fault samples such as abnormal vibration, abnormal temperature and the like of the main bearing from the operation data of the reference shield machine.
And determining a main bearing dynamics parameter sample of the shield machine to be detected by utilizing a main bearing dynamics simulation model of the shield machine to be detected constructed through ADAMS software and a fault sample of the reference shield machine. And then inputting the fault sample into a dynamic simulation model of the main bearing, and carrying out statics and kinematics correlation analysis so as to determine dynamic parameter samples, such as bearing external connection, bearing inner diameter, bearing pitch diameter and the like, of the main bearing of the shield tunneling machine to be detected when the fault occurs. And inputting the current dynamic parameter sample into a preset modeling language, and obtaining a digital twin model corresponding to the main bearing of the shield tunneling machine to be detected. The digital twin model is to correspond the dynamic behavior of the actual shield machine to the digital model thereof by an analog and simulation method so as to realize real-time monitoring and diagnosis. In order to determine the layout path of the sensor, determining a fault data transmission path according to the digital twin model corresponding to each fault sample. And determining the type of the sensor according to the type of the fault sample, and determining the layout direction of the sensor according to the fault data transmission path. And then determining the layout positions of the preset sensors according to the acquisition range corresponding to each sensor type and the layout direction of each sensor determined in the process, so as to realize the layout of a plurality of preset sensors. And by arranging a plurality of preset sensors, the dynamic parameter data of the main bearing of the shield machine to be detected can be acquired in real time.
S102: acquiring historical main bearing operation data corresponding to the main bearing operation data, and extracting a first feature vector of the main bearing according to the historical main bearing operation data, the oil liquid data and the axial displacement data; wherein the first feature vector corresponds to the main bearing ring gear feature.
In order to judge whether the deflection degree of the inner gear ring of the main bearing is abnormal locally or not and the impact load degree of the inner gear ring of the main bearing is detected in the using process, the problem that in the prior art, when the inner gear ring is detected only based on displacement and rotation speed detection, the determination of internal fault parts of the inner gear ring of the main bearing is difficult to realize and the fault positioning accuracy is low is solved. According to the embodiment of the specification, after the historical main bearing operation data corresponding to the main bearing operation data are obtained, oil liquid data and axial displacement data in the main bearing operation data of the shield machine to be detected are collected according to the obtained historical main bearing operation data and a plurality of preset sensors, so that first feature vectors of the main bearing, which are related to faults, in the oil liquid data and the axial displacement data are extracted by referring to the historical bearing operation data. Among these, it is understood that the oil data includes: abrasive grain impurity element data, kinematic viscosity, abrasive grain impurity distribution data reflecting wear particles in bearing lubricating oil and lubricating oil performance, so that the first feature vector corresponds to the ring gear feature of the main bearing.
Specifically, in one or more embodiments of the present disclosure, historical main bearing operation data corresponding to main bearing operation data is obtained, so that a first feature vector of a main bearing is extracted according to the historical main bearing operation data, oil data and axial displacement data, and the method specifically includes the following steps:
Firstly, determining the acquisition time of main bearing operation data, and calling a plurality of historical main bearing operation data of the main bearing of the shield machine to be detected, which correspond to the acquisition time. Wherein, it should be noted that the historical main bearing operation data includes: historical oil data and historical vibration data. By acquiring the first difference value between the oil data and the adjacent historical oil data and the second difference value between the adjacent historical oil data, it is convenient to determine whether the oil data such as abrasion elements have mutation or not, and then detection of the inner gear ring is achieved. If the variation values of the first difference value and the second difference value are larger than the preset variation value, acquiring axial displacement data corresponding to the inner gear ring of the main bearing, and determining the current working condition posture of the main bearing of the shield machine to be detected according to the axial displacement data. And then extracting the oil data and the part data corresponding to the current working condition posture by acquiring the oil data corresponding to the variation value, and taking the oil data and the part data as a first characteristic vector of the main bearing.
S103: and integrating the vibration signals acquired by the preset sensors to obtain vibration signals to be analyzed, and preprocessing the vibration signals to be analyzed to obtain current vibration signals to be analyzed.
In order to enhance the accuracy of vibration signal identification, vibration signals acquired by the preset sensors are integrated to obtain vibration signals to be analyzed containing multi-dimensional information, so that the current vibration signals to be analyzed are obtained by preprocessing the vibration signals to be analyzed. Specifically, in one or more embodiments of the present disclosure, the vibration signals collected by the preset sensors are integrated to obtain a vibration signal to be analyzed, which specifically includes the following steps:
firstly, determining parts corresponding to each preset sensor according to a fault data transmission path, and grouping vibration signals according to the corresponding relation between the preset sensor and the parts, so as to obtain vibration signal groups corresponding to the parts. And then, in order to unify the signal lengths in the vibration signal groups, acquiring acquisition starting time and acquisition ending time corresponding to different vibration signals in each vibration signal group, and further sequencing the acquisition starting time according to the time sequence of each acquisition starting time to obtain an acquisition starting time sequence corresponding to each vibration signal group. And sequencing the acquisition end time corresponding to different vibration signals in each vibration signal group according to the time sequence obtained in the process, so as to obtain the acquisition end time sequence corresponding to each vibration signal group. And determining the current acquisition time range of each vibration signal group according to the cut-off time of the acquisition starting time sequence and the starting time of the acquisition ending time sequence, so as to time align the vibration signals in each vibration signal group according to the current acquisition time range and obtain a plurality of vibration signal groups to be combined. And obtaining vibration signal waveforms in each vibration signal group to be combined, and further obtaining a plurality of vibration signals to be analyzed according to the union integration of the vibration signal waveforms.
Further, in one or more embodiments of the present disclosure, in order to extract a signal component including fault information in a vibration signal to be analyzed as a current vibration signal to be analyzed, so as to improve analysis efficiency and reduce analysis cost, the vibration signal to be analyzed is preprocessed, and a specific process of obtaining the current vibration signal to be analyzed is as follows:
Firstly, decomposing a vibration signal to be analyzed based on an empirical mode decomposition algorithm to obtain connotation mode components corresponding to the frequency range of each layer of the vibration signal to be analyzed. The method for analyzing the time-frequency of the nonlinear self-adaptation of the empirical mode decomposition algorithm can effectively extract local features and time-frequency information of signals, and further obtain connotation mode components corresponding to the frequency range of each layer of vibration signals to be analyzed. And then calculating a mode probability value corresponding to each connotation mode component, determining a frequency component value of each connotation mode component according to the mode probability value, and determining a first connotation mode component to be filtered and a second connotation mode component not to be filtered according to the probability component value of each connotation mode component. It will be appreciated that the higher the pattern probability value, the higher the noise content tends to be the invalid signal, so the connotation mode component with the pattern probability value greater than the preset pattern probability value is taken as the first connotation mode component, and the other components are taken as the second connotation mode components which do not need to be filtered. And then analyzing the first connotation modal component obtained in the process by a bilinear time-frequency distribution algorithm, so that the first connotation modal component is encoded into an analysis signal component. And obtaining an instantaneous frequency estimation value of the analytic signal component according to the peak value estimation of the basic analytic signal component, and filtering according to the instantaneous frequency estimation value to obtain a filtered connotation modal component. Through the bilinear time-frequency distribution algorithm, the problem that effective signal components are easy to miss when analysis is carried out only on the basis of a time domain or only on the basis of a frequency domain is avoided, and the detection accuracy and reliability are improved. And by reconstructing the filtered connotation modal component and the second connotation modal component, the vibration signal to be analyzed is effectively obtained for the inclusion in different dimensions.
Specifically, for the above-mentioned content, under a certain application scenario, it is assumed that we are analyzing a vibration signal of an industrial device, then an empirical mode decomposition algorithm (EMPIRICAL MODE DECOMPOSITION, abbreviated as EMD) is used to decompose the vibration signal to be analyzed, so as to obtain connotation mode components corresponding to different frequency ranges. In the embodiment of the specification, firstly, local maximum value points and local minimum value points of the vibration signal to be analyzed are found, and the envelope curve of the extreme point is obtained through interpolation. And then calculating the average value of the extreme point envelope curve to obtain a mean curve. And subtracting the mean value curve from the original signal to obtain an initial connotation modal component, judging whether the up-and-down fluctuation times of oscillation of the initial connotation modal component are equal or the difference is at most 1, and if the mean value is 0 in any local interval, reserving the initial connotation modal component as an connotation modal component, and obtaining a modal component after the vibration signal is decomposed through iterative processing of the process, wherein the connotation modal component represents characteristic components of the vibration signal to be analyzed on different frequency ranges.
Then, after obtaining the connotation modal components, calculating the ratio of the occurrence times of the connotation modal components to the sum of the total number of the connotation modal components by counting the occurrence times of the connotation modal components, and obtaining the mode probability value of each connotation modal component. The following are to be described: the pattern probability value may reflect the noise content in the signal, with higher values indicating greater noise and lower signal quality. We can set a preset pattern probability value, such as 0.5, as the threshold. According to the mode probability value, we set the content mode components with the mode probability value greater than 0.5 as the first content mode components needing to be filtered, and the other components as the second content mode components needing not to be filtered. And then analyzing the first connotation modal component obtained in the process by a bilinear time-frequency distribution algorithm, so that the first connotation modal component is encoded into an analysis signal component. In the time-frequency analysis process, the input signal is decomposed into a series of time-frequency small areas, and each small area has a moment and a corresponding frequency. For each time-frequency small region, the absolute value of the amplitude value at each moment and frequency can be calculated to obtain the energy value in each small region. These energy values represent the characteristic intensities of the input signal at different moments and frequencies. The calculated energy is used as the code of the analysis signal component. The encoded analytic signal component reflects the distribution of the first connotation modal component on the time-frequency domain. The resolved signal components may be regarded as a representation of the extracted time and frequency domain information within a specific frequency range of the input signal. Through the process, the first connotation modal component can be encoded into the analysis signal component, so that the time-frequency characteristics of the vibration signal to be analyzed in a specific frequency range and the change rule of the vibration signal can be known in more detail, and fault diagnosis and prediction maintenance are facilitated.
S104: performing time-frequency domain analysis on the current vibration signal to be analyzed to extract a second feature vector of the current vibration signal to be analyzed; the second feature vector corresponds to the main bearing inner and outer ring gear features.
After the current vibration signal to be analyzed is obtained based on the step S103, time-frequency domain analysis is performed on the current vibration signal to be analyzed through fourier transformation, so as to extract a second feature vector of the current vibration signal to be analyzed. It should be noted that the second feature vector corresponds to the outer ring gear feature of the main bearing. The method facilitates the subsequent comprehensive analysis of the inner gear ring characteristics and the outer gear ring characteristics, and is beneficial to realizing the comprehensive fault detection of the main bearing of the shield tunneling machine. Specifically, in one or more embodiments of the present disclosure, performing time-frequency domain analysis on a current vibration signal to be analyzed to extract a second eigenvector of the current vibration signal to be analyzed, specifically includes:
Firstly, carrying out envelope demodulation on the current vibration signal to be analyzed based on a Hilbert transform mode in a signal processing method so as to obtain a low-frequency characteristic signal corresponding to the current vibration signal to be analyzed, thereby realizing the extraction of the time domain characteristic and the frequency domain characteristic of the low-frequency characteristic signal. The Hilbert transform is a mathematical tool commonly used in signal processing, and is used for analyzing phase information and amplitude modulation of a signal, and the signal is obtained by post-processing fourier transform of a signal, so that the signal can be expanded from a real number domain to a complex number domain, and further, the time domain features and the frequency domain features of a low-frequency feature signal can be extracted. And then carrying out empirical mode decomposition on the current vibration signal to be analyzed, so as to extract characteristic component energy moment and sample entropy characteristics of the current vibration signal to be analyzed, and generating an initial characteristic set corresponding to the current vibration signal to be analyzed according to the time domain characteristics, the frequency domain characteristics, the characteristic component energy moment and the sample entropy. And then, according to the Euclidean distance between the initial features in the acquired initial feature set, the similarity between the initial features in the initial feature set can be determined so as to construct an adjacency graph corresponding to the initial feature set according to the similarity, and a corresponding adjacency matrix is determined according to the similarity between the output features in the adjacency graph.
And performing structural optimization on the obtained adjacent matrix based on a random gradient algorithm, so as to obtain a low-dimensional feature set corresponding to the initial feature set as a second feature vector of the vibration signal to be analyzed. It should be noted that the optimization effect of the random gradient descent algorithm depends on the selection of the learning rate to a great extent, the learning rate is too large to cause convergence, and the convergence speed is slow if the learning rate is too small, so that the learning rate of the random gradient descent algorithm for structure optimization is determined based on historical experience. It should be further described that, based on a random gradient algorithm, structural optimization is performed on the adjacency matrix, so as to obtain a low-dimensional feature set corresponding to the initial feature set as a second feature vector of the vibration signal to be analyzed, and the method can be performed in an application scenario based on the following modes: first, the adjacency matrix is converted into a graph data structure, e.g., a list of nodes and edges. Then, the feature vector is initialized randomly, and super parameters such as learning rate, iteration number and the like are used. The gradient of each node is calculated using the feature vector and the adjacency matrix of the initial feature. The gradient may be obtained by solving the residual between the adjacency matrix and the eigenvector. The feature vectors are then updated using a gradient descent method or other optimization algorithm based on the calculated gradients and learning rates. The adjacency matrix can be optimized in structure by iterating the process of calculating the node gradient and updating the feature vector in the process.
S105: inputting the first characteristic vector and the second characteristic vector into a preset support vector machine classifier to position the main bearing fault part of the shield machine to be detected.
In order to realize comprehensive real-time detection of the main bearing of the shield machine to be detected, the fault rate of the main bearing of the shield machine is reduced, and the first feature vector and the second feature vector are input into a preset support vector machine classifier in the embodiment of the specification, so that the input vector is analyzed through the preset support vector machine classifier, and the main bearing fault parts of the shield machine to be detected are positioned. By the joint analysis of the first feature vector and the second feature vector, the characteristics of the inner gear ring of the main bearing and the characteristics of the outer gear ring of the main bearing are considered, so that the comprehensive detection of the main bearing of the shield machine to be detected is realized, and the detection accuracy is improved.
Further, in one or more embodiments of the present disclosure, before inputting the first feature vector and the second feature vector into the preset support vector machine classifier to locate a main bearing fault component of the shield machine to be detected, the method further includes the following steps:
firstly, corresponding sample characteristics are acquired according to the model of the main bearing, and the sample characteristics are divided to obtain training sample characteristics and test sample characteristics. And initializing the initial support vector machine classifier to obtain initial parameters of the initial support vector machine classifier. Wherein, it should be noted that the initial parameters include: penalty coefficient, kernel parameters, search range. In order to avoid the problem that the population obtained by randomly sequencing the initial parameters is easy to have uneven distribution, in the embodiment of the specification, the initial parameters are subjected to population initialization through chaotic mapping, so that initialized population parameters are obtained, and then the initial positions of individuals in the population are determined through an inverse mapping mode. And then, calculating the fitness value of each individual in the population, taking the initial parameter corresponding to the calculated optimal fitness value as the optimal parameter, updating the initial support vector machine classifier according to the optimal parameter, and training the updated initial support vector machine classifier based on the training sample characteristics and the test sample characteristics to obtain the preset support vector machine meeting the requirements. The preset support vector machine which accords with the main bearing fault detection scene of the shield machine to be detected is obtained through optimizing the initial support vector machine, and a reliable basis is provided for subsequent fault positioning.
Further, in order to inform corresponding maintenance personnel to perform row inspection in time and avoid loss caused by further development of faults, in one or more embodiments of the present disclosure, after inputting the first feature vector and the second feature vector into a preset support vector machine classifier to locate a main bearing fault component of the shield machine to be detected, the method further includes the following processes: first, determining maintenance links corresponding to fault parts of each main bearing, and accordingly determining a plurality of corresponding maintenance personnel according to the maintenance links. And then, in order to realize optimal allocation, determining waiting time corresponding to each maintenance personnel according to the working state of each maintenance personnel, and further screening a plurality of maintenance personnel according to the waiting time to determine designated maintenance personnel corresponding to the main bearing fault parts.
As shown in fig. 2, the embodiment of the present specification provides a fault detection device for a main bearing of a shield tunneling machine. As can be seen from fig. 2, in one or more embodiments of the present disclosure, a fault detection apparatus for a main bearing of a shield machine includes:
at least one processor 201; and
A memory 202 communicatively coupled to the at least one processor 201; wherein,
The memory 202 stores instructions executable by the at least one processor 201, the instructions being executable by the at least one processor 201 to enable the at least one processor 201 to: performing any of the methods described above.
As shown in fig. 3, the present embodiment provides a nonvolatile storage medium. As can be seen from fig. 3, one or more of the non-volatile storage media of the present description store computer-executable instructions 301, the computer-executable instructions 301 being capable of performing any of the methods described above.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus, devices, non-volatile computer storage medium embodiments, the description is relatively simple, as it is substantially similar to method embodiments, with reference to the section of the method embodiments being relevant.
The foregoing describes specific embodiments of the present disclosure. In some cases, the acts or steps recited in the embodiments of the specification may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing is merely one or more embodiments of the present description and is not intended to limit the present description. Various modifications and alterations to one or more embodiments of this description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, or the like, which is within the spirit and principles of one or more embodiments of the present specification, is intended to be included within the scope of the present specification.
Claims (10)
1. The fault detection method of the main bearing of the shield tunneling machine is characterized by comprising the following steps:
collecting main bearing operation data of the shield machine to be detected based on a plurality of preset sensors; wherein the main bearing operation data includes: oil data, vibration data and axial displacement data;
acquiring historical main bearing operation data corresponding to the main bearing operation data, and extracting a first feature vector of the main bearing according to the historical main bearing operation data, the oil liquid data and the axial displacement data; wherein the first feature vector corresponds to the main bearing ring gear feature;
Integrating the vibration signals acquired by the preset sensors to obtain vibration signals to be analyzed, and preprocessing the vibration signals to be analyzed to obtain current vibration signals to be analyzed;
Performing time-frequency domain analysis on the current vibration signal to be analyzed to extract a second feature vector of the current vibration signal to be analyzed; the second feature vector corresponds to the internal and external gear ring features of the main bearing;
Inputting the first characteristic vector and the second characteristic vector into a preset support vector machine classifier to position the main bearing fault part of the shield machine to be detected.
2. The method for detecting faults of the main bearing of the shield machine according to claim 1, wherein before the main bearing operation data of the shield machine to be detected is collected based on a plurality of preset sensors, the method further comprises:
According to the equipment model of the shield machine to be detected, obtaining structural information corresponding to the shield machine to be detected, inputting the structural information into preset analysis software, and constructing a main bearing dynamics simulation model of the shield machine to be detected;
acquiring a reference shield machine corresponding to the equipment model and the working environment of the shield machine to be detected so as to acquire a fault sample of the reference shield machine;
Determining a main bearing dynamics parameter sample of the shield tunneling machine to be detected based on the main bearing dynamics simulation model and the fault sample, and inputting the main bearing dynamics parameter sample into a preset modeling language to obtain a digital twin model corresponding to the main bearing of the shield tunneling machine to be detected; wherein the main bearing kinetic parameter sample comprises: the bearing is externally connected, the bearing has an inner diameter and a bearing pitch diameter;
determining a fault data transmission path corresponding to each fault sample based on a digital twin model corresponding to each fault sample;
Determining a corresponding sensor type according to the fault type corresponding to the fault sample, and determining a corresponding sensor arrangement direction based on the fault data transmission path;
And determining the layout positions of the preset sensors according to the acquisition range corresponding to each sensor type and the sensor layout direction so as to realize the layout of a plurality of preset sensors.
3. The method for detecting a fault of a main bearing of a shield tunneling machine according to claim 1, wherein the obtaining historical main bearing operation data corresponding to the main bearing operation data, so as to extract a first feature vector of the main bearing according to the historical main bearing operation data, the oil data and the axial displacement data, specifically includes:
Determining the acquisition time of the main bearing operation data to call a plurality of historical main bearing operation data of the main bearing of the shield machine to be detected, which correspond to the acquisition time; wherein the historical main bearing operation data comprises: historical oil data and historical vibration data;
Acquiring a first difference value between the oil data and adjacent historical oil data and a second difference value between adjacent historical oil data;
If the fluctuation value of the first difference value and the second difference value is larger than the preset fluctuation value, acquiring axial displacement data corresponding to the inner gear ring of the main bearing, and determining the current working condition posture of the main bearing of the shield machine to be detected according to the axial displacement data;
acquiring oil data corresponding to the variation value, and extracting part data corresponding to the oil data and the current working condition posture as a first feature vector of the main bearing; wherein, the fluid data includes: abrasive grain impurity element data, kinematic viscosity, abrasive grain impurity distribution data.
4. The method for detecting the faults of the main bearing of the shield machine according to claim 2, wherein the method for integrating the vibration signals collected by the preset sensors to obtain the vibration signals to be analyzed specifically comprises the following steps:
determining parts corresponding to the preset sensors according to the fault data transmission paths, and grouping the vibration signals based on the corresponding relation between the preset sensors and the parts to obtain vibration signal groups corresponding to the parts;
Acquiring acquisition starting time and acquisition ending time corresponding to different vibration signals in each vibration signal group, and sequencing the acquisition starting time according to the time sequence of the acquisition starting time to obtain an acquisition starting time sequence corresponding to each vibration signal group;
Sorting the collection end time corresponding to different vibration signals in each vibration signal group based on time sequence to obtain collection end time sequence corresponding to each vibration signal group;
Determining a current acquisition time range of each vibration signal group according to the cut-off time of the acquisition starting time sequence and the start time of the acquisition ending time sequence, so as to time align vibration signals in each vibration signal group based on the current acquisition time range and obtain a plurality of vibration signal groups to be combined;
and acquiring vibration signal waveforms in each vibration signal group to be combined, and integrating a plurality of vibration signals to be analyzed based on the union of the vibration signal waveforms.
5. The method for detecting the fault of the main bearing of the shield machine according to claim 1, wherein the preprocessing is performed on the vibration signal to be analyzed to obtain a current vibration signal to be analyzed, specifically comprising:
decomposing the vibration signal to be analyzed based on an empirical mode decomposition algorithm to obtain content mode components corresponding to the vibration signal to be analyzed in each layer of frequency range;
calculating a mode probability value corresponding to each connotation mode component, determining a frequency component value of each connotation mode component based on the mode probability value, and determining a first connotation mode component to be filtered and a second connotation mode component not to be filtered according to the probability component value of each connotation mode component;
analyzing the first connotation modal component through a bilinear time-frequency distribution algorithm so as to encode the first connotation modal component into an analysis signal component;
obtaining an instantaneous frequency estimation value of the analytic signal component based on the peak value estimation of the analytic signal component, and filtering based on the instantaneous frequency estimation value to obtain a filtered connotation modal component;
And reconstructing the filtered connotation modal component and the second connotation modal component to obtain a current vibration signal to be analyzed.
6. The method for detecting the fault of the main bearing of the shield machine according to claim 1, wherein the performing time-frequency domain analysis on the vibration signal to be analyzed to extract the second eigenvector of the vibration signal to be analyzed specifically includes:
performing envelope demodulation on the current vibration signal to be analyzed based on Hiber transformation to obtain a low-frequency characteristic signal corresponding to the current vibration signal to be analyzed, so as to extract time domain characteristics and frequency domain characteristics of the low-frequency characteristic signal;
Performing empirical mode decomposition on the current vibration signal to be analyzed to extract characteristic component energy moment and sample entropy characteristics of the current vibration signal to be analyzed, and generating an initial characteristic set corresponding to the current vibration signal to be analyzed according to the time domain characteristics, the frequency domain characteristics, the characteristic component energy moment and the sample entropy;
Determining the similarity of each initial feature in the initial feature set according to Euclidean distance between each initial feature in the initial feature set, constructing an adjacency graph corresponding to the initial feature set based on the similarity, and determining a corresponding adjacency matrix according to the similarity between each output feature in the adjacency graph;
And carrying out structural optimization on the adjacency matrix based on a random gradient algorithm to obtain a low-dimensional feature set corresponding to the initial feature set as a second feature vector of the vibration signal to be analyzed currently.
7. The method for detecting a fault of a main bearing of a shield machine according to claim 1, wherein before inputting the first feature vector and the second feature vector into a preset support vector machine classifier to locate a main bearing fault component of the shield machine to be detected, the method further comprises:
acquiring corresponding sample characteristics according to the model of the main bearing, and dividing the sample characteristics to obtain training sample characteristics and test sample characteristics;
Initializing an initial support vector machine classifier to obtain initial parameters of the initial support vector machine classifier; wherein the initial parameters include: penalty coefficient, kernel parameter, search range;
initializing the population of the initial parameters based on chaotic mapping to obtain initialized population parameters, and determining the initial positions of all individuals in the population based on an inverse mapping mode;
Calculating the fitness value of each individual in the population, taking the initial parameter corresponding to the optimal fitness value as the optimal parameter, updating the initial support vector machine classifier according to the optimal parameter, and training the updated initial support vector machine classifier based on the training sample characteristics and the test sample characteristics to obtain a preset support vector machine meeting the requirements.
8. The method for detecting a fault in a main bearing of a shield machine according to claim 7, wherein after inputting the first feature vector and the second feature vector into a preset support vector machine classifier to locate the main bearing fault component of the shield machine to be detected, the method further comprises:
determining maintenance links corresponding to the main bearing fault parts, and determining a plurality of corresponding maintenance personnel according to the maintenance links;
And determining waiting time corresponding to each maintenance personnel according to the working state of each maintenance personnel, so as to screen the plurality of maintenance personnel based on the waiting time and determine designated maintenance personnel corresponding to the main bearing fault parts.
9. A fault detection device for a main bearing of a shield tunneling machine, the device comprising:
At least one processor; and
A memory communicatively coupled to the at least one processor; wherein,
The memory stores instructions executable by the at least one processor to enable the at least one processor to: performing the method of any of the preceding claims 1-8.
10. A non-volatile storage medium storing computer-executable instructions, the computer-executable instructions being capable of: performing the method of any of the preceding claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410370862.7A CN117969092B (en) | 2024-03-29 | 2024-03-29 | Fault detection method, equipment and medium for main bearing of shield tunneling machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410370862.7A CN117969092B (en) | 2024-03-29 | 2024-03-29 | Fault detection method, equipment and medium for main bearing of shield tunneling machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117969092A true CN117969092A (en) | 2024-05-03 |
| CN117969092B CN117969092B (en) | 2024-06-21 |
Family
ID=90853545
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410370862.7A Active CN117969092B (en) | 2024-03-29 | 2024-03-29 | Fault detection method, equipment and medium for main bearing of shield tunneling machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117969092B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118518358A (en) * | 2024-07-25 | 2024-08-20 | 山东轴研精密轴承有限公司 | Multi-dimensional bearing stability monitoring system |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200200648A1 (en) * | 2018-02-12 | 2020-06-25 | Dalian University Of Technology | Method for Fault Diagnosis of an Aero-engine Rolling Bearing Based on Random Forest of Power Spectrum Entropy |
| CN112033679A (en) * | 2020-10-13 | 2020-12-04 | 济南重工集团有限公司 | Nondestructive testing device and method for main bearing of shield machine |
| CN112033676A (en) * | 2020-09-03 | 2020-12-04 | 中铁隧道局集团有限公司 | Shield machine main bearing detection and repair method |
| CN114705429A (en) * | 2022-03-29 | 2022-07-05 | 中铁工程装备集团有限公司 | Method and device for monitoring main bearing of shield machine |
| CN115017950A (en) * | 2022-06-10 | 2022-09-06 | 上海交通大学 | Tunnel boring machine main bearing fault diagnosis method and system based on roller state |
| CN116070154A (en) * | 2023-03-29 | 2023-05-05 | 山东金帝精密机械科技股份有限公司 | Fault detection method, equipment and medium for large bearing retainer |
| CN116522237A (en) * | 2023-04-17 | 2023-08-01 | 西安交通大学 | Rolling bearing fault diagnosis method based on parallel convolution neural network |
| CN116698414A (en) * | 2023-06-02 | 2023-09-05 | 西南石油大学 | Slurry pump bearing fault monitoring method based on multi-source data fusion |
| CN117347053A (en) * | 2023-09-27 | 2024-01-05 | 上海交通大学 | Tunnel boring machine main bearing fault diagnosis method and system under rare sample |
| CN117629615A (en) * | 2024-01-25 | 2024-03-01 | 山东天工岩土工程设备有限公司 | Shield tunneling machine fault detection method, equipment and medium |
| US20240093733A1 (en) * | 2019-10-10 | 2024-03-21 | Luoyang Mtp Lubrication Tech Co., Ltd. | Rolling bearing with fully-filled plastic oil and manufacturing method therefor |
-
2024
- 2024-03-29 CN CN202410370862.7A patent/CN117969092B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200200648A1 (en) * | 2018-02-12 | 2020-06-25 | Dalian University Of Technology | Method for Fault Diagnosis of an Aero-engine Rolling Bearing Based on Random Forest of Power Spectrum Entropy |
| US20240093733A1 (en) * | 2019-10-10 | 2024-03-21 | Luoyang Mtp Lubrication Tech Co., Ltd. | Rolling bearing with fully-filled plastic oil and manufacturing method therefor |
| CN112033676A (en) * | 2020-09-03 | 2020-12-04 | 中铁隧道局集团有限公司 | Shield machine main bearing detection and repair method |
| CN112033679A (en) * | 2020-10-13 | 2020-12-04 | 济南重工集团有限公司 | Nondestructive testing device and method for main bearing of shield machine |
| CN114705429A (en) * | 2022-03-29 | 2022-07-05 | 中铁工程装备集团有限公司 | Method and device for monitoring main bearing of shield machine |
| CN115017950A (en) * | 2022-06-10 | 2022-09-06 | 上海交通大学 | Tunnel boring machine main bearing fault diagnosis method and system based on roller state |
| CN116070154A (en) * | 2023-03-29 | 2023-05-05 | 山东金帝精密机械科技股份有限公司 | Fault detection method, equipment and medium for large bearing retainer |
| CN116522237A (en) * | 2023-04-17 | 2023-08-01 | 西安交通大学 | Rolling bearing fault diagnosis method based on parallel convolution neural network |
| CN116698414A (en) * | 2023-06-02 | 2023-09-05 | 西南石油大学 | Slurry pump bearing fault monitoring method based on multi-source data fusion |
| CN117347053A (en) * | 2023-09-27 | 2024-01-05 | 上海交通大学 | Tunnel boring machine main bearing fault diagnosis method and system under rare sample |
| CN117629615A (en) * | 2024-01-25 | 2024-03-01 | 山东天工岩土工程设备有限公司 | Shield tunneling machine fault detection method, equipment and medium |
Non-Patent Citations (7)
| Title |
|---|
| 卢永;章文;王威;刘建新;钟龙风;谭桂斌;贺石中;: "油液及振动监测在机泵群轴承早期故障预警中的应用", 润滑与密封, no. 10, 15 October 2020 (2020-10-15) * |
| 吴朝来;: "盾构主轴承检测与评估技术应用", 建筑机械化, no. 08, 15 August 2018 (2018-08-15) * |
| 宫玮丽;梁波;王晓兰;: "基于小波包和Hilbert包络分析的隧道掘进机主轴承故障诊断方法研究", 工业仪表与自动化装置, no. 02, 15 April 2018 (2018-04-15) * |
| 李秋秋;陈闽杰;贺石中;: "油液分析在大型盾构机状态监测中的应用", 机电工程技术, no. 11, 15 November 2010 (2010-11-15) * |
| 柏鹤: "隧道掘进机TBM主轴承系统的状态检测技术", 设备管理与维修, 31 March 2015 (2015-03-31) * |
| 王龙;: "盾构机主驱动轴承无损检测应用技术", 河南科技, no. 05, 5 March 2017 (2017-03-05) * |
| 苏东: "全断面隧道掘进机主轴承故障诊断研究", 建筑机械化, 30 April 2020 (2020-04-30) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118518358A (en) * | 2024-07-25 | 2024-08-20 | 山东轴研精密轴承有限公司 | Multi-dimensional bearing stability monitoring system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117969092B (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117969092B (en) | Fault detection method, equipment and medium for main bearing of shield tunneling machine | |
| JP4175296B2 (en) | Construction machine data processing apparatus and construction machine data processing method | |
| CN110751073A (en) | Pipeline early damage mode identification method based on distributed optical fiber sensing and deep learning | |
| CN113469060A (en) | Multi-sensor fusion convolution neural network aeroengine bearing fault diagnosis method | |
| Martinez-Garcia et al. | Visually interpretable profile extraction with an autoencoder for health monitoring of industrial systems | |
| CN113947017B (en) | Method for predicting residual service life of rolling bearing | |
| CN116186946B (en) | Hydraulic system fault diagnosis method and system based on diagnosis model | |
| CN112926644A (en) | Method and system for predicting residual service life of rolling bearing | |
| CN112305388B (en) | On-line monitoring and diagnosing method for insulation partial discharge faults of generator stator winding | |
| CN104713728A (en) | Large slewing bearing residual life online prediction method based on multidimensional data drive | |
| CN111896254A (en) | Fault prediction system and method for variable-speed variable-load large rolling bearing | |
| CN111914490B (en) | Pump station unit state evaluation method based on depth convolution random forest self-coding | |
| CN106596105A (en) | Method, apparatus and system for diagnosing bearing faults | |
| CN114781450B (en) | Train rolling bearing state identification method based on parameter optimization MOMEDA-MIA-CNN | |
| CN116384224A (en) | Aero-engine life prediction method based on conditional parameter dynamic convolutional neural network | |
| CN116089813A (en) | Wind turbine generator set transmission chain fault diagnosis method based on machine learning | |
| Gao et al. | Long-term Temporal Attention Neural Network with Adaptive Stage Division for Remaining Useful Life Prediction of Rolling Bearings | |
| KR102306244B1 (en) | Method and apparatus for generating fault detecting model for device | |
| CN113850305A (en) | Generator fault diagnosis and health state assessment method and electronic equipment | |
| CN117633690A (en) | Rotary machine health state monitoring method and equipment based on data driving | |
| CN117218602A (en) | Structural health monitoring data abnormality diagnosis method and system | |
| CN116541771A (en) | Unbalanced sample bearing fault diagnosis method based on multi-scale feature fusion | |
| CN116611523A (en) | Method and system for predicting interpretable faults of turbofan engine | |
| Yacout | Logical analysis of maintenance and performance data of physical assets, ID34 | |
| CN115788846A (en) | Method for rapidly identifying leakage fault of hydraulic end of reciprocating pump |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |