CN117435992A - Fault prediction method and system for hydraulic propulsion system of shield tunneling machine - Google Patents
Fault prediction method and system for hydraulic propulsion system of shield tunneling machine Download PDFInfo
- Publication number
- CN117435992A CN117435992A CN202311407956.9A CN202311407956A CN117435992A CN 117435992 A CN117435992 A CN 117435992A CN 202311407956 A CN202311407956 A CN 202311407956A CN 117435992 A CN117435992 A CN 117435992A
- Authority
- CN
- China
- Prior art keywords
- fault
- fault prediction
- model
- tunneling machine
- data set
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000005641 tunneling Effects 0.000 title claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000010276 construction Methods 0.000 claims abstract description 25
- 238000001514 detection method Methods 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims description 8
- 238000002372 labelling Methods 0.000 claims description 6
- 238000010606 normalization Methods 0.000 claims description 5
- 230000009467 reduction Effects 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000006073 displacement reaction Methods 0.000 claims description 3
- 230000001105 regulatory effect Effects 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
Classifications
-
- 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
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
-
- 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
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Mathematical Physics (AREA)
- Business, Economics & Management (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Molecular Biology (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Databases & Information Systems (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Development Economics (AREA)
- Computational Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Algebra (AREA)
- Probability & Statistics with Applications (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
Abstract
The invention discloses a fault prediction method for a hydraulic propulsion system of a shield tunneling machine, which comprises the following steps: constructing a shield tunneling machine data set; constructing a fault prediction initial model; performing regression tree iterative training on the fault prediction initial model based on the shield machine data set to obtain a plurality of weak learners; integrating all weak learners and combining the failure prediction initial model to obtain a failure prediction model; and inputting the data to be detected into the fault prediction model to obtain a corresponding detection result. Correspondingly, the invention discloses a fault prediction system for a hydraulic propulsion system of a shield tunneling machine, which comprises the following components: the system comprises a data set construction module, an initial model construction module, an iterative training module, a fault model construction module and a result output module. On the basis of short training period, the accurate fault prediction of the hydraulic propulsion system of the shield machine is realized.
Description
Technical Field
The invention relates to the technical field of fault prediction, in particular to a fault prediction method and a fault prediction system for a hydraulic propulsion system of a shield machine.
Background
The practical application of the fault prediction technology of the hydraulic propulsion system of the shield machine in engineering is beneficial to the timely maintenance of the hydraulic propulsion system of the shield machine by constructors, ensures the construction safety and the construction efficiency of the shield machine, and has remarkable social benefit and economic benefit. The existing fault prediction method comprises a fault diagnosis method based on a neural network and a fault diagnosis method based on reverse feature elimination.
However, most of the existing fault prediction methods are completed based on the neural network, and have the problems of long training period and difficult algorithm convergence. The actual theory and method for realizing the fault prediction model of the hydraulic propulsion system of the shield machine are lacked.
Therefore, how to realize accurate fault prediction of the hydraulic propulsion system of the shield machine on the basis of short training period is a problem to be solved by the technicians in the field.
Disclosure of Invention
In view of the above, the invention provides a fault prediction method and a fault prediction system for a hydraulic propulsion system of a shield machine, which solve the problem of realizing accurate fault prediction of the hydraulic propulsion system of the shield machine on the basis of short training period.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a fault prediction method for a hydraulic propulsion system of a shield tunneling machine comprises the following steps:
constructing a shield tunneling machine data set;
constructing a fault prediction initial model;
performing regression tree iterative training on the fault prediction initial model based on the shield tunneling machine data set to obtain a plurality of weak learners;
integrating all the weak learners and combining the initial fault prediction model to obtain a fault prediction model;
and inputting the data to be detected into the fault prediction model to obtain a corresponding detection result.
Preferably, the construction of the shield tunneling machine data set specifically includes:
collecting normal sample data and fault sample data of the shield machine, and correspondingly labeling to obtain labeled sample data; the features corresponding to the labeling sample data at least comprise: hydraulic cylinder displacement, hydraulic cylinder speed, hydraulic cylinder rodless cavity pressure, speed regulating valve output flow, load force, system pressure and rod cavity pressure; the fault type of the fault sample data at least comprises: hydraulic cylinder leakage, overflow valve damage, reversing valve leakage and proportional speed valve spring failure;
and sequentially performing dimension reduction treatment and normalization treatment on the marked sample data to obtain a shield tunneling machine data set.
Preferably, the failure prediction initial model f 0 (x) The method specifically comprises the following steps:
wherein N represents the number of sample tags, c represents an uncertainty parameter, L (·) represents a loss function, y i Representing the value of the i-th sample tag.
Preferably, the regression tree iterative training is performed on the fault prediction initial model, and the specific process is as follows:
calculating a shield tunneling machine data set sample residual error r mk ：
Wherein m=1, 2,3, …, M represents the number of regression trees, f (x) k ) A model representing the m-1 st tree, x k Represents the kth sample in the shield tunneling machine data set, f m-1 (x) Representing a model of the tree obtained in the last iteration;
data (x) k ，r mk ) Training as training data to obtain a new regression tree, and calculating the best fitting value c of all leaf nodes of the regression tree mj ：
Wherein j=1, 2,3, …, J represents the number of regression leaf nodes, R mj Representing leaf node regions corresponding to the regression tree.
Preferably, the specific generation process of the regression tree is as follows:
s1, calculating a coefficient Cini (D) of the existing characteristic corresponding values of the current node:
s2, selecting a feature A with the minimum coefficient of the radix and a corresponding value a as an optimal feature and an optimal dividing point;
s3, dividing the node data set into two node data sets D1 and D2 based on the optimal characteristics and the optimal segmentation points;
s4, recursively executing S1-S3 on the sub-nodes in the D1 and the D2 respectively to generate a regression tree;
s5, repeating the S1-S4 until the regression value meets the requirement, and obtaining a plurality of weak classifiers.
6. The method for predicting faults of a hydraulic propulsion system of a shield tunneling machine according to claim 5, wherein a calculation formula of a coefficient Cini (D) of foundation is as follows:
wherein D represents a sample set of the current node, n represents the number of fault categories in the sample set D, s h Represents the h fault class, p (s h ) Representing fault class s h Probability of occurrence.
Preferably, the regression value is up to the requirement, specifically:
and when any condition that the number of samples is smaller than a threshold value, the samples have no characteristics or the coefficient of the foundation is smaller than the threshold value is met, returning to the decision tree subtree, and stopping recursion by the current node.
Preferably, the obtaining a fault prediction model specifically includes:
integrating all the weak learners and combining the failure prediction initial model f 0 (x) Obtaining a fault prediction model f M (x)：
A fault prediction system for a shield machine hydraulic propulsion system, comprising: the system comprises a data set construction module, an initial model construction module, an iterative training module, a fault model construction module and a result output module;
the data set construction module is used for constructing a shield machine data set;
the initial model construction module is used for constructing a fault prediction initial model;
the iterative training module is used for carrying out regression tree iterative training on the fault prediction initial model based on the shield tunneling machine data set to obtain a plurality of weak learners;
the fault model construction module is used for integrating all the weak learners and combining the fault prediction initial model to obtain a fault prediction model;
and the result output module is used for inputting the data to be detected into the fault prediction model to obtain a corresponding detection result.
Compared with the prior art, the invention discloses a fault prediction method and a fault prediction system for a hydraulic propulsion system of a shield machine, which realize the prior data processing by constructing a priori fault database of faults and functions of the shield machine, and perform the fault prediction by constructing a fault prediction model of the shield machine based on GBDT algorithm according to the database.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a fault prediction method provided by the invention.
FIG. 2 is a training flow chart of a failure prediction model provided by the invention.
Fig. 3 is a flowchart of regression tree generation provided by the present invention.
Fig. 4 is a schematic structural diagram of a fault prediction system provided by the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
As shown in fig. 1, the embodiment of the invention discloses a fault prediction method for a hydraulic propulsion system of a shield tunneling machine, which comprises the following steps:
constructing a shield tunneling machine data set;
constructing a fault prediction initial model;
performing regression tree iterative training on the fault prediction initial model based on the shield machine data set to obtain a plurality of weak learners;
integrating all weak learners and combining the failure prediction initial model to obtain a failure prediction model;
and inputting the data to be detected into the fault prediction model to obtain a corresponding detection result.
Example 2
The invention discloses a fault prediction method for a hydraulic propulsion system of a shield tunneling machine, which comprises the following steps:
and constructing a shield tunneling machine data set.
Preferably, the construction of the shield tunneling machine data set specifically comprises:
collecting normal sample data and fault sample data of the shield machine, dividing the normal sample data and the fault sample data into a normal sample set and a fault sample set, and correspondingly labeling to obtain labeled sample data; the features corresponding to the labeling sample data at least comprise: hydraulic cylinder displacement, hydraulic cylinder speed, hydraulic cylinder rodless cavity pressure, speed regulating valve output flow, load force, system pressure and rod cavity pressure; the fault type of the fault sample data at least comprises: hydraulic cylinder leakage, overflow valve damage, reversing valve leakage and proportional speed valve spring failure; and sequentially performing dimension reduction treatment and normalization treatment on the marked sample data to obtain a shield tunneling machine data set.
Preferably, the dimension reduction processing is to reduce the dimension of the high-dimension data into low-dimension data, and only the linearly related attributes or the invariable attributes in the data are combined in the dimension reduction process of the high-dimension data, so that the fault prediction deviation caused by information loss can be avoided; for the linearly related attributes, only one attribute is selected to remain and participate in training; for a constant attribute, the attribute is directly removed from the dataset.
Preferably, the data normalization processing can avoid the mutual influence of data with different dimensionalities, improve the convergence rate of an algorithm, and the normalization function is a mapmin max function.
And constructing a fault prediction initial model.
Preferably, the failure prediction initial model f 0 (x) The method specifically comprises the following steps:
taking the loss function as square loss, taking the square loss as convex function, directly deriving the uncertain parameter c to enable the derivative to be zero, and obtaining the value of c by solving for:
wherein N represents the number of sample tags, L (-) represents the loss function, y i Representing the value of the i-th sample tag.
As shown in fig. 2, performing regression tree iterative training on the fault prediction initial model based on the shield machine data set to obtain a plurality of weak learners; .
Preferably, in FIG. 2, the weight D (m) represents the best fit value c mj Weak classifier m corresponds to update learner f m (x) The learning rate alpha m is a value customized according to actual needs, and the finally obtained strong classifier is a fault prediction model f M (x) The method comprises the steps of carrying out a first treatment on the surface of the The prediction accuracy of the weak classifier is lower, the prediction accuracy of the strong classifier is higher, the generalization performance is better, and the strong learner is obtained by integrating a plurality of weak learners together.
Preferably, the shield tunneling machine data set is divided into a training set and a verification set, wherein the training set and the verification set both contain normal sample data and fault sample data, the verification set is used for checking the prediction accuracy and generalization capability of the fault diagnosis model in the training process and is used for adjusting parameters, and the training set is used for training the fault prediction initial model.
Preferably, the specific process of the initial model training of the fault prediction is as follows:
calculating a shield tunneling machine data set sample residual error r mk ：
Wherein m=1, 2,3, …, M represents the number of regression trees, f (x) k ) A model representing the m-1 st tree, x k Represents the kth sample in the shield tunneling machine data set, f m-1 (x) Representing a model of the tree obtained in the last iteration;
data (x) k ，r mk ) Training to obtain a new regression tree as training data, and calculating the best fitting value c of all leaf nodes of the regression tree mj ：
Wherein j=1, 2,3, …, J represents the number of regression leaf nodes, R mj Representing leaf node regions corresponding to the regression tree.
Preferably, as shown in fig. 3, the regression tree specifically generating process is:
s1, calculating a coefficient Cini (D) of the existing characteristic corresponding values of the current node:
s2, selecting a feature A with the minimum coefficient of the radix and a corresponding value a as an optimal feature and an optimal dividing point;
s3, dividing the data set of the node into two node data sets D1 and D2 based on the optimal characteristics and the optimal segmentation point;
s4, recursively executing S1-S3 on the sub-nodes in the D1 and the D2 respectively to generate a regression tree;
s5 the processes of S1-S4 are repeated,until the regression value meets the requirement, a plurality of weak learners f are obtained m (x)：
Preferably, if the characteristic a when the coefficient of the base is minimum is the system pressure and the corresponding optimal division point a is 3.50×102, samples with the system pressure less than 3.50×102 in the data set D are all divided into D1, and samples with the system pressure greater than 3.50×102 are all divided into D2.
Preferably, the calculated formula of the coefficient Cini (D) is as follows:
wherein D represents a sample set of the current node, n represents the number of fault categories in the sample set D, s h Represents the h fault class, p (s h ) Representing fault class s h Probability of occurrence.
Preferably, the sample set D represents the entire training set, and for each node thereafter, represents a subset of the training set after being sliced after recursion.
Preferably, until the regression value meets the requirement, specifically: and when any condition that the number of samples is smaller than a threshold value, the samples have no characteristics or the coefficient of the foundation is smaller than the threshold value is met, returning to the decision tree subtree, and stopping recursion by the current node.
All weak learners f m (x) Integrating and combining the fault prediction initial model f 0 (x) Obtaining a fault prediction model f M (x)：
And inputting the data to be detected into the fault prediction model to obtain a corresponding detection result.
Example 3
As shown in fig. 4, a fault prediction system for a hydraulic propulsion system of a shield tunneling machine is characterized by comprising: the system comprises a data set construction module, an initial model construction module, an iterative training module, a fault model construction module and a result output module;
the data set construction module is used for constructing a shield tunneling machine data set;
the initial model construction module is used for constructing a fault prediction initial model;
the iterative training module is used for carrying out regression tree iterative training on the fault prediction initial model based on the shield machine data set to obtain a plurality of weak learners;
the fault model construction module is used for integrating all the weak learners and combining the fault prediction initial model to obtain a fault prediction model;
and the result output module is used for inputting the data to be detected into the fault prediction model to obtain a corresponding detection result.
Compared with the prior art, the invention discloses a fault prediction method and a fault prediction system for a hydraulic propulsion system of a shield machine, which realize the prior data processing by constructing a priori fault database of faults and functions of the shield machine, and perform the fault prediction by constructing a fault prediction model of the shield machine based on GBDT algorithm according to the database.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
1. A fault prediction method for a hydraulic propulsion system of a shield tunneling machine, comprising:
constructing a shield tunneling machine data set;
constructing a fault prediction initial model;
performing regression tree iterative training on the fault prediction initial model based on the shield tunneling machine data set to obtain a plurality of weak learners;
integrating all the weak learners and combining the initial fault prediction model to obtain a fault prediction model;
and inputting the data to be detected into the fault prediction model to obtain a corresponding detection result.
2. The method for predicting faults of a hydraulic propulsion system of a shield machine according to claim 1, wherein the constructing a data set of the shield machine specifically comprises:
collecting normal sample data and fault sample data of the shield machine, and correspondingly labeling to obtain labeled sample data; the features corresponding to the labeling sample data at least comprise: hydraulic cylinder displacement, hydraulic cylinder speed, hydraulic cylinder rodless cavity pressure, speed regulating valve output flow, load force, system pressure and rod cavity pressure; the fault type of the fault sample data at least comprises: hydraulic cylinder leakage, overflow valve damage, reversing valve leakage and proportional speed valve spring failure;
and sequentially performing dimension reduction treatment and normalization treatment on the marked sample data to obtain a shield tunneling machine data set.
3. The fault prediction method for a hydraulic propulsion system of a shield machine according to claim 2, wherein the fault prediction initial model f 0 (x) The method specifically comprises the following steps:
wherein N represents the number of sample tags, c represents an uncertainty parameter, L (·) represents a loss function, y i Representing the value of the i-th sample tag.
4. The fault prediction method for the hydraulic propulsion system of the shield machine according to claim 3, wherein the regression tree iterative training is performed on the fault prediction initial model, and the specific process is as follows:
calculating a shield tunneling machine data set sample residual error r mk ：
Wherein m=1, 2,3, …, M represents the number of regression trees, f (x) k ) A model representing the m-1 st tree, x k Represents the kth sample in the shield tunneling machine data set, f m-1 (x) Representing a model of the tree obtained in the last iteration;
data (x) k ，r mk ) Training as training data to obtain a new regression tree, and calculating the best fitting value c of all leaf nodes of the regression tree mj ：
Wherein j=1, 2,3, …, J represents the number of regression leaf nodes, R mj Representing leaf node regions corresponding to the regression tree.
5. The fault prediction method for a hydraulic propulsion system of a shield tunneling machine according to claim 4, wherein the regression tree specifically generating process is as follows:
s1, calculating a coefficient Cini (D) of the existing characteristic corresponding values of the current node:
s2, selecting a feature A with the minimum coefficient of the radix and a corresponding value a as an optimal feature and an optimal dividing point;
s3, dividing the node data set into two node data sets D1 and D2 based on the optimal characteristics and the optimal segmentation points;
s4, recursively executing S1-S3 on the sub-nodes in the D1 and the D2 respectively to generate a regression tree;
s5, repeating the S1-S4 until the regression value meets the requirement, and obtaining a plurality of weak classifiers.
6. The method for predicting faults of a hydraulic propulsion system of a shield tunneling machine according to claim 5, wherein a calculation formula of a coefficient Cini (D) of foundation is as follows:
wherein D represents a sample set of the current node, n represents the number of fault categories in the sample set D, s h Represents the h fault class, p (s h ) Representing fault class s h Probability of occurrence.
7. The method for predicting faults of a hydraulic propulsion system of a shield tunneling machine according to claim 5, wherein the until a regression value meets a requirement is specifically:
and when any condition that the number of samples is smaller than a threshold value, the samples have no characteristics or the coefficient of the foundation is smaller than the threshold value is met, returning to the decision tree subtree, and stopping recursion by the current node.
8. The method for predicting faults of a hydraulic propulsion system of a shield tunneling machine according to claim 5, wherein the obtaining a fault prediction model specifically comprises:
integrating all the weak learners and combining the failure prediction initial model f 0 (x) Obtaining a fault prediction model f M (x)：
9. A fault prediction system for a hydraulic propulsion system of a shield tunneling machine, comprising: the system comprises a data set construction module, an initial model construction module, an iterative training module, a fault model construction module and a result output module;
the data set construction module is used for constructing a shield machine data set;
the initial model construction module is used for constructing a fault prediction initial model;
the iterative training module is used for carrying out regression tree iterative training on the fault prediction initial model based on the shield tunneling machine data set to obtain a plurality of weak learners;
the fault model construction module is used for integrating all the weak learners and combining the fault prediction initial model to obtain a fault prediction model;
and the result output module is used for inputting the data to be detected into the fault prediction model to obtain a corresponding detection result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311407956.9A CN117435992A (en) | 2023-10-27 | 2023-10-27 | Fault prediction method and system for hydraulic propulsion system of shield tunneling machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311407956.9A CN117435992A (en) | 2023-10-27 | 2023-10-27 | Fault prediction method and system for hydraulic propulsion system of shield tunneling machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117435992A true CN117435992A (en) | 2024-01-23 |
Family
ID=89551103
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311407956.9A Pending CN117435992A (en) | 2023-10-27 | 2023-10-27 | Fault prediction method and system for hydraulic propulsion system of shield tunneling machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117435992A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117851810A (en) * | 2024-03-07 | 2024-04-09 | 山东天工岩土工程设备有限公司 | Method and system for detecting and solving faults of shield machine |
-
2023
- 2023-10-27 CN CN202311407956.9A patent/CN117435992A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117851810A (en) * | 2024-03-07 | 2024-04-09 | 山东天工岩土工程设备有限公司 | Method and system for detecting and solving faults of shield machine |
| CN117851810B (en) * | 2024-03-07 | 2024-05-14 | 山东天工岩土工程设备有限公司 | Method and system for detecting and solving faults of shield machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112149316B (en) | Aero-engine residual life prediction method based on improved CNN model | |
| CN111460728B (en) | Method and device for predicting residual life of industrial equipment, storage medium and equipment | |
| CN109297689B (en) | Large-scale hydraulic machinery intelligent diagnosis method introducing weight factors | |
| CN113283027A (en) | Mechanical fault diagnosis method based on knowledge graph and graph neural network | |
| CN117435992A (en) | Fault prediction method and system for hydraulic propulsion system of shield tunneling machine | |
| CN112860904B (en) | External knowledge-integrated biomedical relation extraction method | |
| CN111709244A (en) | Deep learning method for identifying causal relationship of contradictory dispute events | |
| CN110726898A (en) | Power distribution network fault type identification method | |
| CN115617990B (en) | Power equipment defect short text classification method and system based on deep learning algorithm | |
| CN113779882A (en) | Method, device, equipment and storage medium for predicting residual service life of equipment | |
| CN109766481A (en) | The online Hash cross-module state information retrieval method decomposed based on Harmonious Matrix | |
| CN113920379B (en) | Zero sample image classification method based on knowledge assistance | |
| CN114818826A (en) | Fault diagnosis method based on lightweight Vision Transformer module | |
| CN113268370B (en) | Root cause alarm analysis method, system, equipment and storage medium | |
| CN112559741B (en) | Nuclear power equipment defect record text classification method, system, medium and electronic equipment | |
| CN109635008B (en) | Equipment fault detection method based on machine learning | |
| CN112329716A (en) | Pedestrian age group identification method based on gait characteristics | |
| CN115982374B (en) | Multi-view learning entity alignment method and system for dam emergency response knowledge base linkage | |
| CN116702580A (en) | Fermentation process fault monitoring method based on attention convolution self-encoder | |
| CN113807027B (en) | Wind turbine generator system health state evaluation model, method and system | |
| CN114926702A (en) | Small sample image classification method based on depth attention measurement | |
| US20220138554A1 (en) | Systems and methods utilizing machine learning techniques for training neural networks to generate distributions | |
| CN114841063A (en) | Aero-engine residual life prediction method based on deep learning | |
| CN114298413A (en) | Hydroelectric generating set runout trend prediction method | |
| CN109506936A (en) | Bearing fault degree recognition methods based on flow graph and non-naive Bayesian reasoning |
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 |