CN109635008B - Equipment fault detection method based on machine learning - Google Patents
Equipment fault detection method based on machine learning Download PDFInfo
- Publication number
- CN109635008B CN109635008B CN201811572223.XA CN201811572223A CN109635008B CN 109635008 B CN109635008 B CN 109635008B CN 201811572223 A CN201811572223 A CN 201811572223A CN 109635008 B CN109635008 B CN 109635008B
- Authority
- CN
- China
- Prior art keywords
- fault
- data
- initial
- initial data
- equipment
- 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
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention belongs to the technical field of industrial equipment data acquisition and analysis, and discloses an equipment fault detection method based on machine learning. The technical scheme of the invention is as follows: s1, acquiring initial data of the acquired equipment with the fault; s2, carrying out cleaning operation and secondary screening on the initial data; s3, optimizing the pre-data and outputting a retrieval result; s4, classifying the data to obtain an initial result set; s5, classifying the initial result set successively; s6, judging whether the equipment fault represented by the initial data belongs to a known fault or not by comparing the weighted average result with the initial result set; and S7, outputting fault information through a human-computer interface or adding the initial data into a fault database. The invention can realize the self-learning function of the machine, establish the fault database of the equipment, accurately and timely determine the equipment fault and solve the problem, reduce the maintenance times of the mechanical equipment, obviously improve the maintenance efficiency and is suitable for popularization and use.
Description
Technical Field
The invention belongs to the technical field of industrial equipment data acquisition and analysis, and particularly relates to an equipment fault detection method based on machine learning.
Background
In industrial production processes, equipment maintenance is an indispensable task. However, in the process of maintaining the equipment at present, the industrial equipment is usually checked and verified in real time manually after the equipment reports a fault or finds abnormal work, and the manual detection method cannot automatically detect the fault of the industrial equipment, and has low automation degree.
In the prior art, machine learning-based rolling bearing fault detection is adopted, but the technology mainly aims at analyzing various characteristic data and represented information of the rolling bearing in the production process, so that the technology cannot be widely applied to data types of equipment such as various processing machines and the like, and further ensures the safe production of the equipment.
Disclosure of Invention
In order to solve the problems in the prior art, the invention aims to provide an equipment fault detection method based on machine learning, which can realize the self-learning function of a machine, establish a fault database of the equipment, accurately and timely determine the equipment fault and solve the equipment fault, prevent the occurrence of mechanical faults during the detection work of the equipment, reduce the maintenance frequency of the mechanical equipment, and obviously improve the maintenance efficiency, thereby improving the production efficiency of enterprises greatly, improving the economic benefit of the enterprises and having very important significance for the production practice of the enterprises.
The technical scheme adopted by the invention is as follows:
a device fault detection method based on machine learning comprises the following steps:
s1, acquiring initial data of the acquired equipment with the fault;
s2, cleaning the initial data, and performing secondary screening on the initial data after the cleaning operation is completed to obtain pre-data;
s3, optimizing the pre-data through a genetic algorithm, randomly generating a plurality of initial points through the analyzed pre-data, then retrieving the initial points at the same time, and outputting a retrieval result;
s4, weighting and retrieving results through an AdaBoost element algorithm, and classifying data to obtain an initial result set;
s5, classifying the initial result set successively according to the set weight of the samples in each classifier so as to obtain a weighted average result;
s6, judging whether the equipment fault represented by the initial data belongs to a known fault or not by comparing the weighted average result with the initial result set;
and S7, if the judgment result in the step S6 is positive, calling the fault information corresponding to the initial data from the fault database, and outputting the fault information through a human-computer interface, and if the judgment result in the step S6 is negative, adding the initial data into the fault database, and outputting an emergency fault prompt through the human-computer interface.
Preferably, in step S1, the initial data is from an existing data interface of the current device and/or an external sensor of the current device.
Preferably, in step S2, the cleansing operation includes column data editing, cell editing, grouping item editing, and checking the data storage format, length, encoding format, and reference field.
Preferably, in step S3, the specific steps of the optimization operation are as follows:
s301, randomly selecting an initial population P _0 when the initial time t is equal to 0;
s302, calculating an individual fitness function value F (P _ t);
s303, if the fitness function value corresponding to the optimal individual in the population is large enough or the algorithm is continuously operated for multiple generations and the optimal fitness of the individual is not obviously improved, turning to the step S306;
s304, selecting P _ t from P _ (t-1) by applying a selection operator method when the current time t' ═ t + 1;
s305, performing crossing and mutation operations on the P _ t, and then turning to the step S302;
s306, giving the optimal kernel function parameters and penalty factors C, and training a data set by using the optimal kernel function parameters and penalty factors C to obtain a plurality of starting points.
Preferably, in step S3, a Support Vector Machine (SVM) classifier is used to construct an optimal classification surface according to the principle of minimizing structural risk, so that the classification error of the initial data is minimized.
Preferably, in step S5, after each classifier is passed, the weighted classification calculation result is updated with the corresponding weight value.
Preferably, in step S5, the weighted results output by each classifier are summed in a circle to obtain a final output result.
Preferably, the weight value is calculated as follows:
Hfinal=sign=(∑ai·Hi),
wherein HiFor the basic classifier based on a threshold value at which the error rate is lowest, aiRepresents HiImportance in the final classifier.
The invention has the beneficial effects that:
1) the invention can realize the self-learning function of the machine, establish a fault database of the equipment, accurately and timely determine the equipment fault and solve the problem, completely eradicate the occurrence of the mechanical fault during the detection of the mechanical equipment, reduce the maintenance frequency of the mechanical equipment and obviously improve the maintenance efficiency, thereby improving the production efficiency of enterprises greatly and improving the economic benefit of the enterprises, thereby having very important significance for the production practice of the enterprises;
2) in the process of model construction, the method replaces the traditional single model fault detection method and the artificial fault detection method, utilizes the combination of a genetic algorithm and an AdaBoost element algorithm, uses the genetic algorithm to adjust parameters to replace the traditional parameter adjustment method, can improve the accuracy of the algorithm and improve the efficiency, utilizes the AdaBoost element algorithm to support the self-defined addition of various classifiers (a support vector machine, a random forest and a BP neural network) to detect and train the fault, and ensures that the method has the capability of autonomously diagnosing the fault without manually judging the condition;
3) in the machine learning process, the collected fault information is stored in a fault database corresponding to the current equipment, so that the related condition record is carried out on the fault in the past, the machine learning process is realized, when similar same fault occurs for the second time, the fault generation reason and the solution can be prompted on a human-computer interface at the first time, and a manager can sort and update the information in the fault database through the human-computer interface;
4) after training, the invention can have a certain degree of universal applicability to various machine tool processing equipment and production equipment, can be widely applied to various industrial production application scenes with various possible mechanical production faults, and is suitable for popularization and use.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a block flow diagram of the present invention.
Detailed Description
The pedestrian traffic statistical method based on deep learning and multi-target tracking provided by the invention will be described in detail by way of embodiments with reference to the accompanying drawings. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto.
The term "and/or" herein is merely an association describing an associated object, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, B exists alone, and A and B exist at the same time, and the term "/and" is used herein to describe another association object relationship, which means that two relationships may exist, for example, A/and B, may mean: a alone, and both a and B alone, and further, the character "/" in this document generally means that the former and latter associated objects are in an "or" relationship.
Example 1:
as shown in fig. 1, the present embodiment provides a device fault detection method based on machine learning, including the following steps:
s1, acquiring initial data of the acquired equipment with the fault;
s2, cleaning the initial data, and performing secondary screening on the initial data after the cleaning operation is completed to obtain pre-data;
s3, optimizing the pre-data through a genetic algorithm, randomly generating a plurality of initial points through the analyzed pre-data, then retrieving the initial points at the same time, and outputting a retrieval result;
s4, weighting and retrieving results through an AdaBoost element algorithm, and classifying data to obtain an initial result set;
s5, classifying the initial result set successively according to the set weight of the samples in each classifier so as to obtain a weighted average result;
s6, judging whether the equipment fault represented by the initial data belongs to a known fault or not by comparing the weighted average result with the initial result set;
and S7, if the judgment result in the step S6 is positive, calling the fault information corresponding to the initial data from the fault database, and outputting the fault information through a human-computer interface, and if the judgment result in the step S6 is negative, adding the initial data into the fault database, and outputting an emergency fault prompt through the human-computer interface.
In this embodiment, in step S1, the initial data comes from the existing data interface of the current device and/or the external sensor of the current device.
Example 2
On the basis of embodiment 1, as shown in fig. 1, this embodiment provides a device fault detection method based on machine learning, including the following steps:
s1, acquiring initial data of the acquired equipment with the fault;
s2, cleaning the initial data, and performing secondary screening on the initial data after the cleaning operation is completed to obtain pre-data;
s3, optimizing the pre-data through a genetic algorithm, randomly generating a plurality of initial points through the analyzed pre-data, then retrieving the initial points at the same time, and outputting a retrieval result;
s4, weighting and retrieving results through an AdaBoost element algorithm, and classifying data to obtain an initial result set;
s5, classifying the initial result set successively according to the set weight of the samples in each classifier so as to obtain a weighted average result;
s6, judging whether the equipment fault represented by the initial data belongs to a known fault or not by comparing the weighted average result with the initial result set;
and S7, if the judgment result in the step S6 is positive, calling the fault information corresponding to the initial data from the fault database, and outputting the fault information through a human-computer interface, and if the judgment result in the step S6 is negative, adding the initial data into the fault database, and outputting an emergency fault prompt through the human-computer interface.
In this embodiment, in step S2, the cleaning operation includes column data editing, cell editing, and grouping item editing, and further includes checking the data storage format, length, encoding format, and reference field.
In this embodiment, in step S3, a random forest is used to obtain a classification result of the initial data. The specific operation is as follows: first, defining the data of each occurrence, namely N: training initial data number, M: the number of features m: the specified feature number is used for determining the decision result (M is less than or equal to M) of one node on the decision tree; for each classification item, N training initial data are extracted from the training set at random and put back to serve as the training set of the classification item, and the non-extracted initial data are used for prediction and error evaluation. For the node where each branch appears, m features are randomly selected, and each node on the decision branch item is decided by the features. The initial data in each branch item that violates the abstraction is selected, defined and classified as oob initial data. For each initial data, the classification of the classification item is calculated and analyzed as oob initial data. Simple majority vote is used as the classification result of the initial data, and the ratio of the number of misclassifications to the total number of the initial data is used as the oob misclassifications of the random forest. And calculating the optimal branch occurrence mode according to the m characteristics. For each initial datum, it is computed oob as a branch entry for the initial datum. And for the classification condition of the initial data, taking simple majority vote as the classification result of the initial data.
Example 3
On the basis of embodiment 1, as shown in fig. 1, this embodiment provides a device fault detection method based on machine learning, including the following steps:
s1, acquiring initial data of the acquired equipment with the fault;
s2, cleaning the initial data, and performing secondary screening on the initial data after the cleaning operation is completed to obtain pre-data;
s3, optimizing the pre-data through a genetic algorithm, randomly generating a plurality of initial points through the analyzed pre-data, then retrieving the initial points at the same time, and outputting a retrieval result;
s4, weighting and retrieving results through an AdaBoost element algorithm, and classifying data to obtain an initial result set;
s5, classifying the initial result set successively according to the set weight of the samples in each classifier so as to obtain a weighted average result;
s6, judging whether the equipment fault represented by the initial data belongs to a known fault or not by comparing the weighted average result with the initial result set;
and S7, if the judgment result in the step S6 is positive, calling the fault information corresponding to the initial data from the fault database, and outputting the fault information through a human-computer interface, and if the judgment result in the step S6 is negative, adding the initial data into the fault database, and outputting an emergency fault prompt through the human-computer interface.
In this embodiment, the specific steps of the optimization operation are as follows:
s301, randomly selecting an initial population P _0 when the initial time t is equal to 0;
s302, calculating an individual fitness function value F (P _ t);
s303, if the fitness function value corresponding to the optimal individual in the population is large enough or the algorithm is continuously operated for multiple generations and the optimal fitness of the individual is not obviously improved, turning to the step S306;
s304, selecting P _ t from P _ (t-1) by applying a selection operator method when the current time t' ═ t + 1;
s305, performing crossing and mutation operations on the P _ t, and then turning to the step S302;
s306, giving the optimal kernel function parameters and penalty factors C, and training a data set by using the optimal kernel function parameters and penalty factors C to obtain a plurality of starting points.
Example 4
On the basis of embodiment 1, as shown in fig. 1, this embodiment provides a device fault detection method based on machine learning, including the following steps:
s1, acquiring initial data of the acquired equipment with the fault;
s2, cleaning the initial data, and performing secondary screening on the initial data after the cleaning operation is completed to obtain pre-data;
s3, optimizing the pre-data through a genetic algorithm, randomly generating a plurality of initial points through the analyzed pre-data, then retrieving the initial points at the same time, and outputting a retrieval result;
s4, weighting and retrieving results through an AdaBoost element algorithm, and classifying data to obtain an initial result set;
s5, classifying the initial result set successively according to the set weight of the samples in each classifier so as to obtain a weighted average result;
s6, judging whether the equipment fault represented by the initial data belongs to a known fault or not by comparing the weighted average result with the initial result set;
and S7, if the judgment result in the step S6 is positive, calling the fault information corresponding to the initial data from the fault database, and outputting the fault information through a human-computer interface, and if the judgment result in the step S6 is negative, adding the initial data into the fault database, and outputting an emergency fault prompt through the human-computer interface.
In this embodiment, in step S3, a Support Vector Machine (SVM) classifier is used to construct an optimal classification surface according to the principle of minimizing structural risk, so that the classification error of the initial data is minimized. Such as: converting the nonlinear separable data into linear separable data in a feature space through nonlinear mapping, and classifying the linear separable data by finding a linear classifier if a group of initial data is linearly separable. At this time, the feature vector of the input initial data is a low-dimensional vector and is linearly inseparable in the low-dimensional space, and then the feature vector can be converted into a high-dimensional vector by using a kernel function, and the mapping can convert two types of points which are linearly inseparable in the low-dimensional space into linearly separable points.
In the process of model construction, the method replaces the traditional single model fault detection method and the artificial fault detection method, utilizes the combination of a genetic algorithm and an AdaBoost element algorithm, uses the genetic algorithm to adjust parameters to replace the traditional parameter adjustment method, can improve the accuracy of the algorithm and the efficiency, and can support the custom addition of various classifiers (a support vector machine, a random forest and a BP neural network) by utilizing the AdaBoost element algorithm to detect and train the fault, so that the method has the capability of autonomously diagnosing the fault without manually judging the condition.
Example 5
On the basis of embodiment 1, as shown in fig. 1, this embodiment provides a device fault detection method based on machine learning, including the following steps:
s1, acquiring initial data of the acquired equipment with the fault;
s2, cleaning the initial data, and performing secondary screening on the initial data after the cleaning operation is completed to obtain pre-data;
s3, optimizing the pre-data through a genetic algorithm, randomly generating a plurality of initial points through the analyzed pre-data, then retrieving the initial points at the same time, and outputting a retrieval result;
s4, weighting and retrieving results through an AdaBoost element algorithm, and classifying data to obtain an initial result set;
s5, classifying the initial result set successively according to the set weight of the samples in each classifier so as to obtain a weighted average result;
s6, judging whether the equipment fault represented by the initial data belongs to a known fault or not by comparing the weighted average result with the initial result set;
and S7, if the judgment result in the step S6 is positive, calling the fault information corresponding to the initial data from the fault database, and outputting the fault information through a human-computer interface, and if the judgment result in the step S6 is negative, adding the initial data into the fault database, and outputting an emergency fault prompt through the human-computer interface.
In this embodiment, in step S5, after each classifier is passed, the weighted classification calculation result is updated with the corresponding weight value.
In this embodiment, in step S5, the weighted results output from each classifier are summed in a circle to obtain a final output result.
In this embodiment, the formula for calculating the weight value is as follows:
Hfinal=sign=(∑ai·Hi),
wherein HiFor the basic classifier based on a threshold value at which the error rate is lowest, aiRepresents HiImportance in the final classifier.
The invention can realize the self-learning function of the machine, establish a fault database of the equipment, accurately and timely determine the equipment fault and solve the problem, completely eradicate the occurrence of the mechanical fault during the detection of the mechanical equipment, reduce the maintenance frequency of the mechanical equipment and obviously improve the maintenance efficiency, thereby improving the production efficiency of enterprises greatly and improving the economic benefit of the enterprises, thereby having very important significance for the production practice of the enterprises; after training, the invention can have a certain degree of universal applicability to various machine tool processing equipment and production equipment, can be widely applied to various industrial production application scenes with various possible mechanical production faults, and is suitable for popularization and use.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other similar elements in a process, method, article, or apparatus that comprises the element.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, and the program can be stored in a computer readable storage medium, and when the program is executed, the steps comprising the method embodiments are executed.
Finally, it is to be noted that: the above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes performed by the present specification and drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (7)
1. A device fault detection method based on machine learning is characterized in that: the method comprises the following steps:
s1, acquiring initial data of the acquired equipment with the fault;
s2, cleaning the initial data, and performing secondary screening on the initial data after the cleaning operation is completed to obtain pre-data;
s3, optimizing the pre-data through a genetic algorithm, randomly generating a plurality of initial points through the analyzed pre-data, then retrieving the initial points at the same time, and outputting a retrieval result;
s4, weighting and retrieving results through an AdaBoost element algorithm, and classifying data to obtain an initial result set;
s5, classifying the initial result set successively according to the set weight of the samples in each classifier so as to obtain a weighted average result;
s6, judging whether the equipment fault represented by the initial data belongs to a known fault or not by comparing the weighted average result with the initial result set;
s7, if the judgment result in the step S6 is yes, calling the fault information corresponding to the initial data from the fault database, and outputting the fault information through a human-computer interface, if the judgment result in the step S6 is no, adding the initial data into the fault database, and outputting an emergency fault prompt through the human-computer interface;
in step S3, the specific steps of the optimization operation are as follows:
s301, randomly selecting an initial population P _0 when the initial time t is equal to 0;
s302, calculating an individual fitness function value F (P _ t);
s303, if the fitness function value corresponding to the optimal individual in the population is large enough or the algorithm is continuously operated for multiple generations and the optimal fitness of the individual is not obviously improved, turning to the step S306;
s304, selecting P _ t from P _ (t-1) by applying a selection operator method when the current time t' ═ t + 1;
s305, performing crossing and mutation operations on the P _ t, and then turning to the step S302;
s306, giving the optimal kernel function parameters and penalty factors C, and training a data set by using the optimal kernel function parameters and penalty factors C to obtain a plurality of starting points.
2. The machine learning-based device failure detection method of claim 1, wherein: in step S1, the initial data is from the existing data interface of the current device and/or the external sensor of the current device.
3. The machine learning-based device failure detection method of claim 1, wherein: in step S2, the cleaning operation includes column data editing, cell editing, and grouping item editing, and further includes checking a data storage format, a length, an encoding format, and a reference field, respectively; and during secondary screening, the initial data reference system after cleaning is kept consistent, and the uniqueness of the codes is ensured.
4. The machine learning-based device failure detection method of claim 1, wherein: in step S3, a Support Vector Machine (SVM) classifier is used to construct an optimal classification surface according to the principle of minimizing structural risk, so that the classification error of the initial data is minimized.
5. The machine learning-based device failure detection method of claim 1, wherein: in step S5, after each classifier is passed, the weighted classification calculation result is updated with a corresponding weight value.
6. The machine learning-based device failure detection method of claim 5, wherein: in step S5, the weighted results output from each classifier are summed in a circle to obtain the final output result.
7. The machine learning-based device failure detection method of claim 5, wherein: the weighted value is calculated as follows:
Hfinal=sign(∑ai·Hi)
wherein HiFor the basic classifier based on a threshold value at which the error rate is lowest, aiRepresents HiImportance in the final classifier.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811572223.XA CN109635008B (en) | 2018-12-21 | 2018-12-21 | Equipment fault detection method based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811572223.XA CN109635008B (en) | 2018-12-21 | 2018-12-21 | Equipment fault detection method based on machine learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109635008A CN109635008A (en) | 2019-04-16 |
CN109635008B true CN109635008B (en) | 2020-11-24 |
Family
ID=66076229
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811572223.XA Active CN109635008B (en) | 2018-12-21 | 2018-12-21 | Equipment fault detection method based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109635008B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110837898A (en) * | 2019-10-25 | 2020-02-25 | 袁茂银 | Underground cable fault repairing method and device based on model selection |
CN113240140A (en) * | 2021-06-04 | 2021-08-10 | 海尔数字科技(青岛)有限公司 | Fault detection method, device, equipment and storage medium of physical equipment |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103617576B (en) * | 2013-11-21 | 2015-03-04 | 南京大学 | General fault detecting and maintenance method for equipment |
CN108520080B (en) * | 2018-05-11 | 2020-05-05 | 武汉理工大学 | Ship diesel generator fault prediction and health state online evaluation system and method |
CN108681747B (en) * | 2018-05-11 | 2021-05-04 | 武汉理工大学 | Rotary machine fault diagnosis and state monitoring system and method based on deep learning |
-
2018
- 2018-12-21 CN CN201811572223.XA patent/CN109635008B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN109635008A (en) | 2019-04-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110636066B (en) | Network security threat situation assessment method based on unsupervised generative reasoning | |
CN109492748B (en) | Method for establishing medium-and-long-term load prediction model of power system based on convolutional neural network | |
CN113449919B (en) | Power consumption prediction method and system based on feature and trend perception | |
CN112363896A (en) | Log anomaly detection system | |
CN109635008B (en) | Equipment fault detection method based on machine learning | |
CN112084240B (en) | Intelligent identification and linkage treatment method and system for group renting | |
CN110826237A (en) | Bayesian belief network-based wind power equipment reliability analysis method and device | |
CN116503118A (en) | Waste household appliance value evaluation system based on classification selection reinforcement prediction model | |
CN116702090A (en) | Multi-mode data fusion and uncertain estimation water level prediction method and system | |
CN104537383A (en) | Massive organizational structure data classification method and system based on particle swarm | |
CN117435992A (en) | Fault prediction method and system for hydraulic propulsion system of shield tunneling machine | |
CN111984514A (en) | Prophet-bLSTM-DTW-based log anomaly detection method | |
CN115438190B (en) | Power distribution network fault auxiliary decision knowledge extraction method and system | |
CN116545679A (en) | Industrial situation security basic framework and network attack behavior feature analysis method | |
CN116452335A (en) | Financial big data management system based on time sequence neural network | |
CN115293827A (en) | Novel model interpretability analysis method for assisting fine operation of enterprise | |
CN111143774B (en) | Power load prediction method and device based on influence factor multi-state model | |
CN111680572B (en) | Dynamic judgment method and system for power grid operation scene | |
CN114820074A (en) | Target user group prediction model construction method based on machine learning | |
CN113850483A (en) | Enterprise credit risk rating system | |
CN112348275A (en) | Regional ecological environment change prediction method based on online incremental learning | |
CN116861286B (en) | Multi-mode landslide displacement intelligent prediction method based on fuzzy calculation | |
WO2022059183A1 (en) | Information processing device, information processing method, and information processing program | |
CN117057777A (en) | Comprehensive management method for operation and maintenance of power distribution network | |
Escovedo et al. | Using ensembles for adaptive learning: A comparative approach |
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 |