CN111275136B - Fault prediction system based on small sample and early warning method thereof - Google Patents
Fault prediction system based on small sample and early warning method thereof Download PDFInfo
- Publication number
- CN111275136B CN111275136B CN202010176870.XA CN202010176870A CN111275136B CN 111275136 B CN111275136 B CN 111275136B CN 202010176870 A CN202010176870 A CN 202010176870A CN 111275136 B CN111275136 B CN 111275136B
- Authority
- CN
- China
- Prior art keywords
- data
- module
- feature
- early warning
- fault
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—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
- G06F18/21—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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a fault prediction system based on a small sample and an early warning method thereof, wherein a data cleaning module is used for carrying out deletion filling, smoothing treatment and redundant sample elimination on system data; the feature engineering module is used for constructing first-order difference features of fault occurrence time, abnormal value identification of sensor data and abnormal statistical features; the feature selection module is used for screening out key base features based on the number of the segmentation nodes in all the trees and an algorithm of the features; the regression module is used for training and adjusting parameters of a regression model by adopting a K-fold cross validation and grid search method so that the regression model achieves preset prediction precision; and the model fusion module is used for constructing a final prediction result by adopting a weighted fusion method for the multiple models trained by the regressor module. The invention can greatly improve the precision of the fault time prediction under the small sample data set.
Description
Technical Field
The invention relates to a general fault early warning system, in particular to a fault prediction system based on a small sample and an early warning method thereof.
Background
Machine learning technology has been developed in breakthrough in the fields of image, medical treatment, recommendation, etc.; in the production and manufacturing link, if the time of equipment failure occurrence can be accurately predicted, the equipment failure occurrence prediction method can be more active, the maintenance means can be ensured to be provided before the equipment failure occurrence prediction method, the equipment downtime is reduced, and the maintenance labor cost is reduced. Therefore, the equipment failure prediction is a key for ensuring the efficient operation of the production process, and is an important guarantee for realizing intelligent manufacturing.
Aiming at the problem of predicting the occurrence time of faults in the production and manufacturing links, intensive research is carried out in academia and industry. Existing prediction methods include traditional methods and modern methods: the traditional method is that the average value of the fault interval time is counted as the next fault occurrence time; modern methods such as statistical machine learning based methods, neural network based methods, etc. The traditional method has simple logic, is easy to understand, and has lower precision. Modern methods are complex in logic and high in accuracy, but prediction on small sample data sets often produces overfitting and is less robust because of the difficulty in acquiring faulty sample data. One problem that is often faced during practice is: insufficient accumulation of raw fault data results in a large degree of overfitting when modeling targets directly.
The traditional fault early warning is mainly classified into two types, namely judging whether the equipment is about to generate faults according to the running state data of the equipment, wherein the fault early warning system can only obtain a prediction result of 'fault' or 'no fault', but has higher requirements on the number of positive and negative samples, and has no prediction capability on the occurrence time of the faults; the second type is a regression problem, the occurrence time of the future fault of the equipment is predicted according to the historical fault occurrence time of the equipment and the equipment running state data before the corresponding fault, and the method can better solve the problem that the first type of method cannot predict the occurrence time of the equipment fault, but has higher requirements on the number of positive and negative samples.
Disclosure of Invention
The invention aims to provide a fault prediction system based on a small sample, and another aim of the invention is to provide an early warning method of the fault prediction system. The intelligent algorithm is adopted to discretize the fault occurrence time, so that samples are converted into data suitable for a classification algorithm, the data are input into a model to train, and the next fault occurrence time of the equipment is subjected to model prediction.
In order to achieve the above purpose, the invention adopts the following technical scheme:
the invention discloses a fault prediction system based on a small sample, which comprises the following modules:
the data cleaning module is used for carrying out deletion filling, smoothing treatment and redundant sample removal on the system data, and transmitting the cleaned data to the characteristic engineering module to extract corresponding characteristic data;
the feature engineering module is used for constructing first-order difference features of fault occurrence time, abnormal value identification of sensor data and abnormal statistical features;
the feature selection module is used for screening out key base features based on an L1 regularized feature coefficient algorithm, a weighted mean square error reduction algorithm and the number and algorithm of segmentation nodes of the features in all trees respectively, and inputting the base features into the regressor module respectively;
the regressor module adopts K-fold cross validation and grid search methods to train and adjust parameters of a regression model, so that the regression model achieves preset prediction precision, and meanwhile, the trained regressor module is utilized to predict test data;
and the model fusion module is used for constructing a final prediction result by adopting a weighted fusion method for the multiple models trained by the regressor module.
The invention discloses an early warning method of a fault prediction system based on a small sample, which comprises the following steps:
step 1, a training sample database is established; the training sample database contains historical fault data of equipment to be early-warned, and the method comprises the following steps: fault code type, fault occurrence time, equipment operation parameters, equipment state parameters, equipment number, equipment type, equipment geographic information and equipment installation time;
step 2, performing deletion filling, smoothing treatment and redundant sample elimination on system data through a data cleaning module, and transmitting the cleaned data to a characteristic engineering module to extract corresponding characteristic data;
step 3, the feature engineering module constructs a first-order difference feature of fault occurrence time, abnormal value identification of sensor data and abnormal statistical feature through the feature data;
step 4, screening out key base features by adopting a feature selection module based on an L1 regularized feature coefficient algorithm, a weighted mean square error reduction algorithm and the number and algorithm of segmentation nodes in all trees based on features, and inputting the key base features to a regressor module respectively;
step 5, the regressor module adopts K-fold cross validation and grid search methods to train and adjust parameters of a regression model, so that the regression model achieves preset prediction precision, and simultaneously, the trained regressor module is utilized to predict test data;
and 6, constructing a final prediction result by adopting a weighted fusion method to the multiple models trained by the regressor module through a model fusion module.
In the step 3, the characteristic engineering model is based on an Isolate Forest algorithm to recognize abnormal values of the sensor data, so that abnormal values of various sensors are counted and used as statistical characteristics, the abnormal value data of the sensors are scored based on a maximum entropy algorithm, comprehensive scores of all samples are obtained, and finally a sliding window method is adopted to construct first-order difference characteristics of fault occurrence time.
The IsolateForest algorithm calculates what layer each sample falls on by traversing a plurality of binary trees, and calculates the average height, so as to judge whether the sample is an abnormal sample point according to a preset threshold value.
The IsolateForest algorithm randomly selects a plurality of point samples from training data to serve as sub-samples, and places the sub-samples into the root node of the tree.
The method comprises the steps that the IsolateForest algorithm randomly designates a dimension, a cutting point p is randomly generated in current node data, and the cutting point p is generated between the maximum value and the minimum value of the designated dimension in the current node data; a hyperplane is generated according to the cut point p, and then the current node data space is divided into 2 subspaces, namely: a. data smaller than the cutting point p in the appointed dimension is placed at the left child node of the current node, and data larger than or equal to the cutting point p is placed at the right child node of the current node; b. and then recursing the step a in the child nodes, and continuously constructing new child nodes until only one data exists in the child nodes, namely the child nodes cannot be cut any more, or the child nodes reach a limited height.
The characteristic coefficient algorithm is based on L1 regularization, wherein the loss function is as follows:
wherein the method comprises the steps of
For all characteristic coefficients, x i Representing the input of the ith sample, y i Represents the output of the ith sample, +.>
Is an L1 regular term.
The feature coefficient calculation is based on a weighted mean square error reduction algorithm, and the importance of each feature is calculated, so that feature selection is performed according to the magnitude of the feature importance value:
where x represents the current node feature, y and z represent the leaf node feature of the current node, and N (x), N (y), N (z) represent the number of samples of nodes x, y, z.
The feature coefficient calculation is based on the number of the segmentation nodes of the features in all the trees and an algorithm, and the importance degree of each feature is calculated, so that feature selection is carried out according to the importance degree value of the feature:
wherein x represents the current node characteristic, tree i (x) Representing the number of times the current node feature is output on the ith tree.
In step 4, the regressor module obtains a test sample from the training sample database, and performs model training and verification of relative errors of the prediction results according to the candidate features finally selected by the feature selection module.
According to the invention, the feature engineering enriches the feature dimension of the data, the feature selection is used for rapidly and accurately screening key features, the robustness of fault prediction is enhanced through multi-model fusion, the fault time prediction accuracy under a small sample data set is greatly improved, the prediction of the fault occurrence time under a small sample by a fault prediction system is satisfied, and the interference of redundant features to a model is greatly reduced.
Drawings
Fig. 1 is a block diagram of a fault prediction system according to the present invention.
FIG. 2 is a block flow diagram of the early warning method of the present invention.
Detailed Description
The following describes embodiments of the present invention in detail with reference to the accompanying drawings, and the embodiments and specific operation procedures are given by the embodiments of the present invention under the premise of the technical solution of the present invention, but the scope of protection of the present invention is not limited to the following embodiments.
As shown in fig. 1, the fault prediction system based on small samples according to the present invention includes the following modules:
the data cleaning module is used for carrying out deletion filling, smoothing treatment and redundant sample removal on the system data, and transmitting the cleaned data to the characteristic engineering module to extract corresponding characteristic data;
the feature engineering module is used for constructing first-order difference features of fault occurrence time, abnormal value identification of sensor data and abnormal statistical features;
the feature selection module is used for screening out key base features based on an L1 regularized feature coefficient algorithm, a weighted mean square error reduction algorithm and the number and algorithm of segmentation nodes of the features in all trees respectively, and inputting the base features into the regressor module respectively;
the regressor module adopts K-fold cross validation and grid search methods to train and adjust parameters of a regression model, so that the regression model achieves preset prediction precision, and meanwhile, the trained regressor module is utilized to predict test data;
and the model fusion module is used for constructing a final prediction result by adopting a weighted fusion method for the multiple models trained by the regressor module.
As shown in fig. 2, the early warning method of the fault prediction system based on the small sample of the invention comprises the following steps:
step 1, a training sample database is established; the training sample database contains historical fault data of equipment to be early-warned, and the method comprises the following steps: fault code type (fault code 1, fault code 2..7), fault occurrence time, equipment operating parameters (pulse pressure value, ambient temperature, sensor signal state, motor speed), equipment status parameters (on time, working time, historical maintenance times, component replacement information), equipment number, equipment type, equipment geographical information (longitude and latitude, province county, geographical environment), equipment installation time;
step 2, carrying out missing filling on system data through a data cleaning module, specifically, filling by adopting a characteristic mean value aiming at numerical value characteristics, and filling by adopting a field representing a vacancy value aiming at discrete characteristics; smoothing, namely smoothing the interference of noise data in a log logarithm mode; removing redundant samples, calculating correlation coefficients among features, screening feature data according to the correlation coefficients, and transmitting the cleaned data to a feature engineering module to extract the corresponding feature data;
step 3, the characteristic engineering module constructs a first-order difference characteristic of fault occurrence time of each device through the characteristic data, and adopts an isocyanate Forest algorithm to identify abnormal values of operating parameters (pulse pressure value, ambient temperature, sensor signal state and motor rotating speed) of the device and identify abnormal statistical characteristics;
step 4, screening out key base features by adopting a feature selection module based on an L1 regularized feature coefficient algorithm, a weighted mean square error reduction algorithm and the number and algorithm of segmentation nodes in all trees based on features, and inputting the key base features to a regressor module respectively;
and 5, the regressor module respectively uses a LASSO regression model, a GBDT regression model or an XGBOOST regression model to obtain a training sample from a training sample database, and the training sample is subjected to a data cleaning module, a characteristic engineering module and a characteristic selection module to train and adjust parameters of the regression model by adopting a K-fold cross validation and grid search method so that the model achieves preset prediction precision, and simultaneously, the trained regressor module is utilized to predict the test data.
And 6, constructing a final prediction result by adopting a weighted fusion method to the multiple models trained by the regressor module through a model fusion module.
In the step 3, the characteristic engineering model is based on an Isolate Forest algorithm to recognize abnormal values of the sensor data, so that abnormal values of various sensors are counted and used as statistical characteristics, the abnormal value data of the sensors are scored based on a maximum entropy algorithm, comprehensive scores of all samples are obtained, and finally a sliding window method is adopted to construct first-order difference characteristics of fault occurrence time.
The IsolateForest algorithm calculates what layer each sample falls on by traversing a plurality of binary trees, and calculates the average height, so as to judge whether the sample is an abnormal sample point according to a preset threshold value.
The IsolateForest algorithm randomly selects a plurality of point samples from training data to serve as sub-samples, and places the sub-samples into the root node of the tree.
The method comprises the steps that the IsolateForest algorithm randomly designates a dimension, a cutting point p is randomly generated in current node data, and the cutting point p is generated between the maximum value and the minimum value of the designated dimension in the current node data; a hyperplane is generated according to the cut point p, and then the current node data space is divided into 2 subspaces, namely: a. data smaller than the cutting point p in the appointed dimension is placed at the left child node of the current node, and data larger than or equal to the cutting point p is placed at the right child node of the current node; b. and then recursing the step a in the child nodes, and continuously constructing new child nodes until only one data exists in the child nodes, namely the child nodes cannot be cut any more, or the child nodes reach a limited height.
The characteristic coefficient algorithm is based on L1 regularization, wherein the loss function is as follows:
wherein the method comprises the steps of
For all characteristic coefficients, x i Representing the input of the ith sample, y i Representing the output of the ith sample.
The feature coefficient calculation is based on a weighted mean square error reduction algorithm, and the importance of each feature is calculated, so that feature selection is performed according to the magnitude of the feature importance value:
where x represents the current node feature, y and z represent the leaf node feature of the current node, and N (x), N (y), N (z) represent the number of samples of nodes x, y, z.
The feature coefficient calculation is based on the number of the segmentation nodes of the features in all the trees and an algorithm, and the importance degree of each feature is calculated, so that feature selection is carried out according to the importance degree value of the feature:
wherein x represents the current node characteristic, tree i (x) Representing the number of times the current node feature is output on the ith tree.
In step 4, the regressor module obtains a test sample from the training sample database, and performs model training and verification of relative errors of the prediction results according to the candidate features finally selected by the feature selection module.
When the invention is applied to different equipment and different fault codes, the comparison of the performance of the model with the performance of the model without the performance of the model is shown in table 1.
TABLE 1
As can be seen from table 1, when data is input to the LASSO model, the GBDT model, and the XGBOOST model and applied to different equipment and different fault code training, after feature selection is performed on the input data, the relative error of 94% of models is smaller than that of models which do not perform feature selection on the data on the test set. This demonstrates that the feature selection approach employed by the present invention can improve the accuracy of the model.
When the invention is applied to different equipment and different sample numbers, the comparison of the model fusion results after feature selection is shown in Table 2.
TABLE 2
As can be seen from table 2, after feature selection, the relative error of 73% of the weighted fusion models is smaller on the test set for all different devices, different combinations of fault codes than the respective relative error of the corresponding combinations on the three base models (LASSO model, GBDT model, XGBOOST model). This shows that the method of the invention adopts the weighted fusion model to greatly improve the accuracy of the model to the prediction result.
Claims (10)
1. A fault prediction system based on a small sample, characterized in that: comprises the following modules:
the data cleaning module is used for carrying out deletion filling, smoothing treatment and redundant sample removal on the system data, and transmitting the cleaned data to the characteristic engineering module to extract corresponding characteristic data;
the feature engineering module is used for constructing first-order difference features of fault occurrence time, abnormal value identification of sensor data and abnormal statistical features;
the feature selection module is used for screening out key base features based on an L1 regularized feature coefficient algorithm, a weighted mean square error reduction algorithm and the number and algorithm of segmentation nodes of the features in all trees respectively, and inputting the base features into the regressor module respectively;
the regressor module adopts K-fold cross validation and grid search methods to train and adjust parameters of a regression model, so that the regression model achieves preset prediction precision, and meanwhile, the trained regressor module is utilized to predict test data;
and the model fusion module is used for constructing a final prediction result by adopting a weighted fusion method for the multiple models trained by the regressor module.
2. The early warning method of the small sample based fault prediction system as claimed in claim 1, wherein: comprising the following steps:
step 1, a training sample database is established; the training sample database contains historical fault data of equipment to be early-warned, and the method comprises the following steps: fault code type, fault occurrence time, equipment operation parameters, equipment state parameters, equipment number, equipment type, equipment geographic information and equipment installation time;
step 2, performing deletion filling, smoothing treatment and redundant sample elimination on system data through a data cleaning module, and transmitting the cleaned data to a characteristic engineering module to extract corresponding characteristic data;
step 3, the feature engineering module constructs a first-order difference feature of fault occurrence time, abnormal value identification of sensor data and abnormal statistical feature through the feature data;
step 4, screening out key base features by adopting a feature selection module based on an L1 regularized feature coefficient algorithm, a weighted mean square error reduction algorithm and the number and algorithm of segmentation nodes in all trees based on features, and inputting the key base features to a regressor module respectively;
step 5, the regressor module adopts K-fold cross validation and grid search methods to train and adjust parameters of a regression model, so that the regression model achieves preset prediction precision, and simultaneously, the trained regressor module is utilized to predict test data;
and 6, constructing a final prediction result by adopting a weighted fusion method to the multiple models trained by the regressor module through a model fusion module.
3. The early warning method of the fault prediction system based on the small sample according to claim 2, wherein the early warning method comprises the following steps: in the step 3, the characteristic engineering model is based on an Isolate Forest algorithm to recognize abnormal values of the sensor data, so that abnormal values of various sensors are counted and used as statistical characteristics, the abnormal value data of the sensors are scored based on a maximum entropy algorithm, comprehensive scores of all samples are obtained, and finally a sliding window method is adopted to construct first-order difference characteristics of fault occurrence time.
4. The early warning method of the fault prediction system based on the small sample according to claim 3, wherein the early warning method comprises the following steps: the IsolateForest algorithm calculates what layer each sample falls on by traversing a plurality of binary trees, and calculates the average height, so as to judge whether the sample is an abnormal sample point according to a preset threshold value.
5. The early warning method of the fault prediction system based on the small sample according to claim 3, wherein the early warning method comprises the following steps: the IsolateForest algorithm randomly selects a plurality of point samples from training data to serve as sub-samples, and places the sub-samples into the root node of the tree.
6. The early warning method of the fault prediction system based on the small sample according to claim 3, wherein the early warning method comprises the following steps: the method comprises the steps that the IsolateForest algorithm randomly designates a dimension, a cutting point p is randomly generated in current node data, and the cutting point p is generated between the maximum value and the minimum value of the designated dimension in the current node data; a hyperplane is generated according to the cut point p, and then the current node data space is divided into 2 subspaces, namely: a. data smaller than the cutting point p in the appointed dimension is placed at the left child node of the current node, and data larger than or equal to the cutting point p is placed at the right child node of the current node; b. and then recursing the step a in the child nodes, and continuously constructing new child nodes until only one data exists in the child nodes, namely the child nodes cannot be cut any more, or the child nodes reach a limited height.
7. The early warning method of the fault prediction system based on the small sample according to claim 2, wherein the early warning method comprises the following steps: the characteristic coefficient algorithm is based on L1 regularization, wherein the loss function is as follows:
wherein the method comprises the steps of
For all characteristic coefficients, x i Representing the input of the ith sample, y i Represents the output of the ith sample, +.>
Is an L1 regular term.
8. The early warning method of the fault prediction system based on the small sample according to claim 2, wherein the early warning method comprises the following steps: the feature coefficient calculation is based on a weighted mean square error reduction algorithm, and the importance of each feature is calculated, so that feature selection is performed according to the magnitude of the feature importance value:
where x represents the current node feature, y and z represent the leaf node feature of the current node, and N (x), N (y), N (z) represent the number of samples of nodes x, y, z.
9. The early warning method of the fault prediction system based on the small sample according to claim 2, wherein the early warning method comprises the following steps: the feature coefficient calculation is based on the number of the segmentation nodes of the features in all the trees and an algorithm, and the importance degree of each feature is calculated, so that feature selection is carried out according to the importance degree value of the feature:
wherein x represents the current node characteristic, tree i (x) Representing the number of times the current node feature is output on the ith tree.
10. The early warning method of the fault prediction system based on the small sample according to claim 2, wherein the early warning method comprises the following steps: in step 4, the regressor module obtains a test sample from the training sample database, and performs model training and verification of relative errors of the prediction results according to the candidate features finally selected by the feature selection module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010176870.XA CN111275136B (en) | 2020-03-13 | 2020-03-13 | Fault prediction system based on small sample and early warning method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010176870.XA CN111275136B (en) | 2020-03-13 | 2020-03-13 | Fault prediction system based on small sample and early warning method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111275136A CN111275136A (en) | 2020-06-12 |
| CN111275136B true CN111275136B (en) | 2023-05-02 |
Family
ID=70999638
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010176870.XA Active CN111275136B (en) | 2020-03-13 | 2020-03-13 | Fault prediction system based on small sample and early warning method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111275136B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112905572A (en) * | 2021-01-29 | 2021-06-04 | 铁道警察学院 | Data anomaly information studying and judging model and method |
| CN114004425B (en) * | 2021-12-29 | 2022-06-07 | 北京京东振世信息技术有限公司 | Article circulation information prediction model generation method, information generation method and device |
| CN114511113A (en) * | 2022-01-28 | 2022-05-17 | 国网上海市电力公司 | Fault positioning method and system for TTU (time to Unit) equipment in transformer area, electronic equipment and storage medium |
| CN117753795B (en) * | 2024-02-07 | 2024-05-31 | 东北大学 | Feedforward control method for hot rolled products with multiple steel types and specifications |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015158198A1 (en) * | 2014-04-17 | 2015-10-22 | 北京泰乐德信息技术有限公司 | Fault recognition method and system based on neural network self-learning |
| CN110417011A (en) * | 2019-07-31 | 2019-11-05 | 三峡大学 | A kind of online dynamic secure estimation method based on mutual information Yu iteration random forest |
| CN110806743A (en) * | 2019-12-05 | 2020-02-18 | 成都天玙兴科技有限公司 | Equipment fault detection and early warning system and method based on artificial intelligence |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9141915B2 (en) * | 2013-01-30 | 2015-09-22 | Siemens Aktiengesellschaft | Method and apparatus for deriving diagnostic data about a technical system |
-
2020
- 2020-03-13 CN CN202010176870.XA patent/CN111275136B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015158198A1 (en) * | 2014-04-17 | 2015-10-22 | 北京泰乐德信息技术有限公司 | Fault recognition method and system based on neural network self-learning |
| CN110417011A (en) * | 2019-07-31 | 2019-11-05 | 三峡大学 | A kind of online dynamic secure estimation method based on mutual information Yu iteration random forest |
| CN110806743A (en) * | 2019-12-05 | 2020-02-18 | 成都天玙兴科技有限公司 | Equipment fault detection and early warning system and method based on artificial intelligence |
Non-Patent Citations (1)
| Title |
|---|
| 邓森 ; 景博 ; 周宏亮 ; 朱海鹏 ; 刘小平 ; .一种改进的自适应增强-支持向量回归机的故障预测方法.兵工学报.2012,(08),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111275136A (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111275136B (en) | Fault prediction system based on small sample and early warning method thereof | |
| CN111914873B (en) | Two-stage cloud server unsupervised anomaly prediction method | |
| CN109102005B (en) | Small sample deep learning method based on shallow model knowledge migration | |
| CN106168799B (en) | A method of batteries of electric automobile predictive maintenance is carried out based on big data machine learning | |
| CN111222549B (en) | Unmanned aerial vehicle fault prediction method based on deep neural network | |
| CN111325410B (en) | Universal fault early warning system based on sample distribution and early warning method thereof | |
| CN109977624A (en) | Photovoltaic plant soft fault monitoring method based on deep neural network | |
| CN113378990B (en) | Flow data anomaly detection method based on deep learning | |
| CN113570138B (en) | Method and device for predicting residual service life of equipment of time convolution network | |
| CN111738308A (en) | Dynamic threshold detection method for monitoring index based on clustering and semi-supervised learning | |
| CN114201374A (en) | Operation and maintenance time sequence data anomaly detection method and system based on hybrid machine learning | |
| CN117421684B (en) | Abnormal data monitoring and analyzing method based on data mining and neural network | |
| CN116737510B (en) | Data analysis-based intelligent keyboard monitoring method and system | |
| Yang et al. | Remaining useful life prediction based on normalizing flow embedded sequence-to-sequence learning | |
| CN116915582A (en) | Diagnosis and analysis method and device for fault root cause of communication terminal | |
| CN114529067A (en) | Method for performing predictive maintenance on electric vehicle battery based on big data machine learning | |
| CN111934903B (en) | Docker container fault intelligent prediction method based on time sequence evolution gene | |
| CN113313365A (en) | Degradation early warning method and device for primary air fan | |
| CN115964470B (en) | Method and system for predicting service life of motorcycle accessories | |
| Pradeep et al. | Optimal Predictive Maintenance Technique for Manufacturing Semiconductors using Machine Learning | |
| CN117216713A (en) | Fault delimiting method, device, electronic equipment and storage medium | |
| CN111885146B (en) | Data transmission method for industrial data cloud service platform | |
| CN115392710A (en) | Wind turbine generator operation decision method and system based on data filtering | |
| CN113986673A (en) | Fault correlation method and device and computer readable storage medium | |
| Mbuli et al. | Root causes analysis and fault prediction in intelligent transportation systems: coupling unsupervised and supervised learning techniques |
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 |