CN112906915B - Rail transit system fault diagnosis method based on deep learning - Google Patents
Rail transit system fault diagnosis method based on deep learning Download PDFInfo
- Publication number
- CN112906915B CN112906915B CN202110087150.0A CN202110087150A CN112906915B CN 112906915 B CN112906915 B CN 112906915B CN 202110087150 A CN202110087150 A CN 202110087150A CN 112906915 B CN112906915 B CN 112906915B
- Authority
- CN
- China
- Prior art keywords
- data
- traffic data
- rail
- neural network
- network model
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000013135 deep learning Methods 0.000 title claims abstract description 16
- 238000003062 neural network model Methods 0.000 claims abstract description 26
- 125000004122 cyclic group Chemical group 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000010606 normalization Methods 0.000 claims abstract description 9
- 230000006870 function Effects 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000012423 maintenance Methods 0.000 abstract description 13
- 206010063385 Intellectualisation Diseases 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 230000001010 compromised effect Effects 0.000 description 2
- 238000013138 pruning Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- 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
- G06Q10/20—Administration of product repair or maintenance
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Primary Health Care (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses a fault diagnosis method of an rail transit system based on deep learning, which comprises the steps of integrating rail transit data and carrying out normalization pretreatment; extracting features of the preprocessed rail traffic data; adjusting the distribution of the rail traffic data to form a data set, and establishing a circulating neural network model; inputting the data set into the cyclic neural network model for training, and outputting the data set meeting the requirements; the rail transit fault diagnosis is carried out by utilizing the trained data set, and an intelligent means is used for replacing manpower in the rail transit fault diagnosis, so that the intellectualization, the datamation and the informatization of the maintenance system are improved, the personnel investment is greatly reduced, the labor cost is reduced, and the reliability, the effectiveness and the safety of the operation maintenance of the whole line are further improved.
Description
Technical Field
The invention relates to the technical field of rail transit operation and maintenance, in particular to a rail transit system fault diagnosis method based on deep learning.
Background
Through the development of the last ten years, china has become the country with the most rapid development of worldwide track traffic, new routes of railway and urban track traffic are far and ahead each year, and the operation mileage is continuously increased. The rail transit is a backbone network for pulse and traffic transportation of national economy in China, not only bears most national strategy and economic material transportation, but also bears passenger transportation function, and plays a great role in promoting the transportation of resources in China, enhancing the communication of economic areas, solving urban traffic congestion and the like. Along with the formation and development of the track traffic network in China, the track traffic industry at present gradually enters into construction, operation and maintenance and is in a new stage, and the maintenance of the track traffic is particularly important.
At present, the work aiming at fault diagnosis of the rail transit system is also based on field operation of a large number of personnel, on one hand, along with the increase of the capacity of the rail transit system, a large number of personnel investment is required to be equipped, and the labor cost is continuously increased; on the other hand, on-site staff cannot always ensure the reliability of system fault diagnosis after heavy work, so that the effectiveness, safety and reliability of the whole line operation maintenance are greatly compromised.
Disclosure of Invention
This section is intended to outline some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description summary and in the title of the application, to avoid obscuring the purpose of this section, the description summary and the title of the invention, which should not be used to limit the scope of the invention.
The present invention has been made in view of the above-mentioned problems associated with the conventional rail transit system fault diagnosis method.
Therefore, the technical problems solved by the invention are as follows: the method solves the problems that the existing fault diagnosis work for the rail transit system is mainly based on the increase of cost caused by field operation of a large number of personnel and the reliability of fault diagnosis cannot be always ensured.
In order to solve the technical problems, the invention provides the following technical scheme: a fault diagnosis method of an rail transit system based on deep learning comprises the steps of integrating rail transit data and carrying out normalization pretreatment; extracting features of the preprocessed rail traffic data; adjusting the distribution of the rail traffic data to form a data set, and establishing a circulating neural network model; inputting the data set into the cyclic neural network model for training, and outputting the data set meeting the requirements; diagnosis of rail transit system faults is performed by using the trained data set.
As a preferable scheme of the deep learning-based rail transit system fault diagnosis method, the invention comprises the following steps: integrating the rail traffic data and performing normalization preprocessing comprises the steps of obtaining the rail traffic data with different sources and different characteristics; sequentially carrying out reduction pretreatment according to the selected characteristic values corresponding to each item of track traffic data to obtain each item of reduced track traffic data; and sequentially carrying out normalization processing on each item of the reduced rail traffic data.
As a preferable scheme of the deep learning-based rail transit system fault diagnosis method, the invention comprises the following steps: performing reduction preprocessing on the track traffic data according to the following formula to acquire each item of reduced track traffic data,
wherein t is the acquired track crossing data, t' is each item of reduced track crossing data, and delta is the selected characteristic value corresponding to each item of track crossing data t.
As a preferable scheme of the deep learning-based rail transit system fault diagnosis method, the invention comprises the following steps: the extraction reference quantity when the feature extraction is carried out on the preprocessed rail traffic data is as follows,
wherein, delta' is the extraction reference quantity when extracting the characteristics, delta is the selected characteristic value corresponding to each track traffic data t.
As a preferable scheme of the deep learning-based rail transit system fault diagnosis method, the invention comprises the following steps: adjusting the distribution of the track traffic data comprises establishing a topological structure and inputting the extracted characteristics into the topological structure; determining a reference sequence; acquiring the comprehensive association degree between the extracted different features; arranging the data from low to high outer ring according to the comprehensive association degree between the features; a balanced dataset is formed.
As a preferable scheme of the deep learning-based rail transit system fault diagnosis method, the invention comprises the following steps: defining the minimum extracted reference quantity as a reference sequence.
As a preferable scheme of the deep learning-based rail transit system fault diagnosis method, the invention comprises the following steps: inputting the data set into the cyclic neural network model for training, outputting the data set meeting the requirements, and then verifying and adjusting the super parameters of the cyclic neural network model.
As a preferable scheme of the deep learning-based rail transit system fault diagnosis method, the invention comprises the following steps: the cyclic neural network model function is,
E＝t′·∑P·H(δ)
wherein t' is the reduced rail intersection data, delta is the selected characteristic value corresponding to the rail intersection data t, t is the rail intersection data, P is the comprehensive association function value, H is the selected characteristic value delta function value corresponding to the rail intersection data t, and E is the output quantity of the circulating neural network model.
As a preferable scheme of the deep learning-based rail transit system fault diagnosis method, the invention comprises the following steps: and when the E value output of the cyclic neural network model is larger than the corresponding t', defining that the corresponding data in the data set passes through the training of the cyclic neural network model, and meeting the requirements.
The invention has the beneficial effects that: the invention replaces manual work with an intelligent means in fault diagnosis of the rail transit system, the intellectualization, the datamation and the informatization of the maintenance system are improved, the investment of personnel is greatly reduced, the labor cost is reduced, the reliability, the effectiveness and the safety of the operation maintenance of the whole line are further improved, and the new world of the intelligent operation maintenance of the rail transit is opened.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein:
FIG. 1 is a flow chart of a method provided by the present invention;
FIG. 2 is a diagram illustrating the operation of a code interface during database operation pruning operation according to the present invention;
FIG. 3 is a schematic diagram of a sparse coding linear model according to the present invention;
FIG. 4 is a schematic diagram of a topology employed in the present invention;
FIG. 5 is a schematic diagram of the present invention after the data is arranged from low to high outer loop according to the overall correlation between features.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
Example 1
At present, the work aiming at fault diagnosis of the rail transit system is also based on field operation of a large number of personnel, on one hand, along with the increase of the capacity of the rail transit system, a large number of personnel investment is required to be equipped, and the labor cost is continuously increased; on the other hand, on-site staff cannot always ensure the reliability of system fault diagnosis after heavy work, so that the effectiveness, safety and reliability of the whole line operation maintenance are greatly compromised.
Accordingly, referring to fig. 1 to 5, the present invention provides a fault diagnosis method for an rail transit system based on deep learning, which includes:
integrating rail traffic data and carrying out normalization pretreatment;
extracting features of the preprocessed rail traffic data;
adjusting the distribution of the rail traffic data to form a data set, and establishing a circulating neural network model;
inputting the data set into a cyclic neural network model for training, and outputting the data set meeting the requirements;
diagnosis of rail transit system faults is performed by using the trained data set.
Further, integrating the rail traffic data and performing normalization preprocessing includes:
acquiring rail traffic data with different sources and different characteristics;
sequentially carrying out reduction pretreatment according to the selected characteristic values corresponding to each item of rail traffic data to obtain each item of reduced rail traffic data;
and sequentially carrying out normalization processing on each item of reduced rail traffic data.
It should be noted that:
(1) and acquiring rail traffic data with different sources and different characteristics through the sensor. And after corresponding data are acquired by different sensors, the data are uniformly connected into a database system for classification processing.
It is noted that, different track traffic data are transmitted after being integrated evenly according to the principle of 'large storage capacity + small storage capacity' (namely maximum + minimum, second maximum + second minimum, third maximum + third minimum … …) in the process of being transmitted to the database, so that the data pressure of transmission is reduced, and the occurrence of data turbulence is prevented.
(2) The acquired rail traffic data comprise anchor segments, positioning points and the like. In consideration of the fact that the amount of data of the track traffic acquired by the sensor is large, the excessive amount of data can increase the operation pressure of the database, and the operation accuracy can be reduced, so that a certain reduction process is carried out on a large amount of data transmitted to the database.
Wherein, the track traffic data is reduced and preprocessed according to the following formula to obtain each reduced track traffic data,
wherein t is the acquired rail traffic data, t' is the reduced rail traffic data, and delta is the selected characteristic value corresponding to the rail traffic data t.
Specifically, aiming at anchor segment rail traffic data of different sources, delta is selected to be 1.2-1.3, and a 1.2-bit best selected characteristic value is generally selected; for the positioning point rail traffic data of different sources, delta is selected to be 0.8-1.1, and a characteristic value is selected to be 0.9 best.
Additionally, as shown in fig. 2, a code running chart is used for running a corresponding pruning operation for the database, and a program algorithm for performing the reduction processing on the database is as follows:
Spring.datasource.url＝jdbc:mysql://localhost:3307/springboot-crud-mysql-vuejsserverTimezone＝UTC&useSSL＝false
Spring.datasource.username＝root
Spring.datasource.password＝(δ,δ＇)
Spring.datasource.driver-class-name＝com.mysql.jdbc.Driver
Spring.jpa.hibernate.ddl-auto＝create
Spring.jpa.database-platform＝org.hiberate.dialect.MySQL1.2Dialect
Spring.jpa.database-platform＝org.hiberate.dialect.MySQL0.9Dialect
Spring.jpa.generate-ddl＝true
Spring.jpa.show-sql＝true
Spring.freemarker.suffix＝.html
furthermore, the self-encoder technology such as sparse coding technology is adopted to extract the characteristics of the adjusted data.
Referring to fig. 3, specifically, the sparse coding cost function model is:
the detailed algorithm is as follows:
input: signal f (t), dictionary.
And (3) outputting: list of coefficients (an, g) rn ).
Initializing:
R 1 ——f(t)；
n——1；
repeating:
findg rn ∈Dwith maximum inner product∣＜Rn,g rn ＞∣；
a n ——＜Rn,g rn ＞；
R n+1 ——R n -g rn ；
n——n+1；
until a sparse stop condition is reached, for example: ||R n ∣∣＜threshold.
Wherein the extraction reference quantity when the feature extraction is carried out on the preprocessed rail traffic data is as follows,
wherein, delta' is the extraction reference quantity when extracting the characteristics, delta is the selected characteristic value corresponding to each item of track traffic data t.
Further, considering that the distribution of the collected fault data may be very unbalanced due to the difference of the type and model of the device, for better prediction capability of the generalization system, balancing the unbalanced data set, and adjusting the distribution of the track traffic data includes:
establishing a topological structure, and inputting the extracted characteristics into the topological structure;
determining a reference sequence;
acquiring the comprehensive association degree between the extracted different features;
arranging the data from low to high outer ring according to the comprehensive association degree between the features;
a balanced dataset is formed.
Wherein, the minimum extraction reference quantity is defined as a reference sequence.
Referring to fig. 4, a schematic diagram of the topology is adopted.
Referring to fig. 5, a schematic diagram of the data after being arranged according to the outer loop from low to high of the integrated association degree between the features is shown.
The following table 1 shows the comparison of the performance of the prediction results of the present invention with the data distribution of the track traffic without adjustment:
table 1: predicted outcome performance comparison table
| Database rate of operation (bytes/s) | Prediction accuracy (100%) | |
| Adjusting distribution | 3062.8 | 94.87 |
| Unadjusting the distribution | 1022 | 73.25 |
As shown in table 1 above, the performance after adjustment of the distribution is significantly better than the case without adjustment of the distribution in terms of the database calculation rate and the accuracy of prediction.
Additionally, the data set is input into the cyclic neural network model for training, and the super parameters of the cyclic neural network model are verified and adjusted after the data set meeting the requirements is output.
The conventional hidden Markov process, conditional random field and the like are realized by adopting scikit-learn and other technologies, training is carried out on a data set by adopting a deep learning method such as a cyclic neural network and the like realized based on Tensorflow, and the super parameters of a model are adjusted by cross verification.
Specifically, the cyclic neural network model function is,
E＝t′·∑P·H(δ)
wherein t' is the reduced rail intersection data, delta is the selected characteristic value corresponding to the rail intersection data t, t is the rail intersection data, P is the comprehensive association function value, H is the selected characteristic value delta function value corresponding to the rail intersection data t, and E is the output quantity of the circulating neural network model.
When the E value output of the cyclic neural network model is larger than the corresponding t', corresponding data in the data set are defined to meet the requirements through training of the cyclic neural network model.
The following table 2 shows a comparison of the performance of the present invention and conventional methods in the diagnosis of an on-track traffic system:
table 2: performance comparison table 2
As shown in table 2 above, based on the ROS platform, the corresponding results were counted.
Because the traditional technology only detects 8 times in each quarter by a manual mode, the failure rate is not detected, and the absolute accuracy of prediction reaches 100% of deficiency, the intelligent detection is realized, and the statistics times and the relative accuracy of prediction are obviously higher than those of the traditional technology.
The invention replaces manual work with an intelligent means in fault diagnosis of the rail transit system, the intellectualization, the datamation and the informatization of the maintenance system are improved, the investment of personnel is greatly reduced, the labor cost is reduced, the reliability, the effectiveness and the safety of the operation maintenance of the whole line are further improved, and the new world of the intelligent operation maintenance of the rail transit is opened.
It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.
Claims (1)
1. A fault diagnosis method of an rail transit system based on deep learning is characterized by comprising the following steps of: comprising the steps of (a) a step of,
integrating rail traffic data and carrying out normalization pretreatment;
extracting features of the preprocessed rail traffic data;
adjusting the distribution of the rail traffic data to form a data set, and establishing a circulating neural network model;
inputting the data set into the cyclic neural network model for training, and outputting the data set meeting the requirements;
diagnosing rail transit system faults by utilizing the trained data set;
wherein the integrating the rail traffic data and the normalizing preprocessing comprises,
acquiring the rail traffic data with different sources and different characteristics;
sequentially carrying out reduction pretreatment according to the selected characteristic values corresponding to each item of track traffic data to obtain each item of reduced track traffic data;
sequentially carrying out normalization processing on each item of reduced rail traffic data;
wherein, the track traffic data is reduced and preprocessed according to the following formula to obtain each item of reduced track traffic data,
wherein t is the acquired track traffic data, t' is each item of reduced track traffic data, and delta is a selected characteristic value corresponding to each item of track traffic data t;
wherein the extraction reference quantity when the feature extraction is carried out on the preprocessed rail traffic data is as follows,
wherein, delta' is an extraction reference quantity when extracting the characteristics, delta is a selected characteristic value corresponding to each track traffic data t;
wherein adjusting the distribution of the track traffic data comprises,
establishing a topological structure, and inputting the extracted characteristics into the topological structure;
determining a reference sequence;
acquiring the comprehensive association degree between the extracted different features;
arranging the data from low to high outer ring according to the comprehensive association degree between the features;
forming a balanced dataset;
defining the minimum extracted reference quantity as a reference sequence;
inputting the data set into the cyclic neural network model for training, outputting the data set meeting the requirements, and then verifying and adjusting the super parameters of the cyclic neural network model;
wherein the cyclic neural network model function is as follows,
E＝t′·∑P·H(δ)
wherein t' is the reduced rail intersection data, delta is the selected characteristic value corresponding to each rail intersection data t, t is the rail intersection data, P is the comprehensive association function value, H is the selected characteristic value delta function value corresponding to each rail intersection data t, and E is the output quantity of the circulating neural network model;
when the E value output of the cyclic neural network model is larger than the corresponding t', defining that the corresponding data in the data set passes through the training of the cyclic neural network model, and meeting the requirements.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110087150.0A CN112906915B (en) | 2021-01-22 | 2021-01-22 | Rail transit system fault diagnosis method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110087150.0A CN112906915B (en) | 2021-01-22 | 2021-01-22 | Rail transit system fault diagnosis method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112906915A CN112906915A (en) | 2021-06-04 |
| CN112906915B true CN112906915B (en) | 2024-03-22 |
Family
ID=76117517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110087150.0A Active CN112906915B (en) | 2021-01-22 | 2021-01-22 | Rail transit system fault diagnosis method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112906915B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114996258A (en) * | 2022-06-23 | 2022-09-02 | 中铁第四勘察设计院集团有限公司 | Contact network fault diagnosis method based on data warehouse |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107769972A (en) * | 2017-10-25 | 2018-03-06 | 武汉大学 | A kind of power telecom network equipment fault Forecasting Methodology based on improved LSTM |
| CN108960077A (en) * | 2018-06-12 | 2018-12-07 | 南京航空航天大学 | A kind of intelligent failure diagnosis method based on Recognition with Recurrent Neural Network |
| CN108985279A (en) * | 2018-08-28 | 2018-12-11 | 上海仁童电子科技有限公司 | The method for diagnosing faults and device of double-unit traction controller waveform |
| CN109959122A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of Energy consumption forecast for air conditioning method based on shot and long term memory Recognition with Recurrent Neural Network |
| CN110033135A (en) * | 2019-04-15 | 2019-07-19 | 北京交通大学 | The train braking system failure prediction method that Multivariate Time Series feature is reinforced |
| CN110222844A (en) * | 2019-05-30 | 2019-09-10 | 西安交通大学 | A kind of compressor performance prediction technique based on artificial neural network |
| CN110348513A (en) * | 2019-07-10 | 2019-10-18 | 北京华电天仁电力控制技术有限公司 | A kind of Wind turbines failure prediction method based on deep learning |
| CN110659779A (en) * | 2019-09-26 | 2020-01-07 | 国网湖南省电力有限公司 | Power distribution system network loss prediction method based on long-term and short-term memory network |
| CN111060779A (en) * | 2019-11-20 | 2020-04-24 | 安徽大学 | Power grid partition fault diagnosis method and system based on probabilistic neural network |
| CN111505543A (en) * | 2020-03-18 | 2020-08-07 | 中国电力科学研究院有限公司 | Method and system for compensating dynamic hysteresis based on recurrent neural network |
| CN111813084A (en) * | 2020-07-10 | 2020-10-23 | 重庆大学 | Mechanical equipment fault diagnosis method based on deep learning |
| CN111898802A (en) * | 2020-07-08 | 2020-11-06 | 贵州电网有限责任公司 | Power grid disaster situation evaluation method based on deep neural network |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102016008987B4 (en) * | 2015-07-31 | 2021-09-16 | Fanuc Corporation | Machine learning method and machine learning apparatus for learning failure conditions, and failure prediction apparatus and failure prediction system including the machine learning apparatus |
-
2021
- 2021-01-22 CN CN202110087150.0A patent/CN112906915B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107769972A (en) * | 2017-10-25 | 2018-03-06 | 武汉大学 | A kind of power telecom network equipment fault Forecasting Methodology based on improved LSTM |
| CN108960077A (en) * | 2018-06-12 | 2018-12-07 | 南京航空航天大学 | A kind of intelligent failure diagnosis method based on Recognition with Recurrent Neural Network |
| CN108985279A (en) * | 2018-08-28 | 2018-12-11 | 上海仁童电子科技有限公司 | The method for diagnosing faults and device of double-unit traction controller waveform |
| CN109959122A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of Energy consumption forecast for air conditioning method based on shot and long term memory Recognition with Recurrent Neural Network |
| CN110033135A (en) * | 2019-04-15 | 2019-07-19 | 北京交通大学 | The train braking system failure prediction method that Multivariate Time Series feature is reinforced |
| CN110222844A (en) * | 2019-05-30 | 2019-09-10 | 西安交通大学 | A kind of compressor performance prediction technique based on artificial neural network |
| CN110348513A (en) * | 2019-07-10 | 2019-10-18 | 北京华电天仁电力控制技术有限公司 | A kind of Wind turbines failure prediction method based on deep learning |
| CN110659779A (en) * | 2019-09-26 | 2020-01-07 | 国网湖南省电力有限公司 | Power distribution system network loss prediction method based on long-term and short-term memory network |
| CN111060779A (en) * | 2019-11-20 | 2020-04-24 | 安徽大学 | Power grid partition fault diagnosis method and system based on probabilistic neural network |
| CN111505543A (en) * | 2020-03-18 | 2020-08-07 | 中国电力科学研究院有限公司 | Method and system for compensating dynamic hysteresis based on recurrent neural network |
| CN111898802A (en) * | 2020-07-08 | 2020-11-06 | 贵州电网有限责任公司 | Power grid disaster situation evaluation method based on deep neural network |
| CN111813084A (en) * | 2020-07-10 | 2020-10-23 | 重庆大学 | Mechanical equipment fault diagnosis method based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112906915A (en) | 2021-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110867196B (en) | Machine equipment state monitoring system based on deep learning and voice recognition | |
| CN111080032A (en) | Load prediction method based on Transformer structure | |
| CN112529341B (en) | Drilling well leakage probability prediction method based on naive Bayesian algorithm | |
| CN106600140A (en) | Improved support vector machine-based gas pipeline fault prediction and early warning system and method | |
| CN106446208A (en) | Method for identifying traveling mode of smart phone by considering integrating degree of road network | |
| CN102495919B (en) | Extraction method for influence factors of carbon exchange of ecosystem and system | |
| CN106772695B (en) | A kind of Along Railway wind speed forecasting method for merging many air measuring station measured datas | |
| CN111709244B (en) | Deep learning method for identifying cause and effect relationship of contradictory dispute | |
| CN112036490B (en) | Identification reconstruction method for railway longitudinal section line shape | |
| CN102063641A (en) | Method for forecasting wind speed of high speed railway line | |
| CN112906915B (en) | Rail transit system fault diagnosis method based on deep learning | |
| CN112612997B (en) | Oil-water relative permeability curve calculation method based on deep learning | |
| CN105574489A (en) | Layered stack based violent group behavior detection method | |
| CN109191828B (en) | Traffic participant accident risk prediction method based on ensemble learning | |
| CN114757365A (en) | High-speed railway roadbed settlement prediction and early warning method based on deep learning | |
| CN113640380A (en) | Multi-stage classification method and system for rail damage detection | |
| CN109086926A (en) | A kind of track traffic for passenger flow prediction technique in short-term based on combination neural net structure | |
| CN114964781B (en) | Intelligent diagnosis method for train bearing faults | |
| CN116542170A (en) | Drainage pipeline siltation disease dynamic diagnosis method based on SSAE and MLSTM | |
| CN111340516A (en) | Satisfaction evaluation system and method based on information entropy and variation coefficient fusion algorithm | |
| CN103955622B (en) | System and method for identifying targets by airborne multiple sensors | |
| CN112651176A (en) | GA-BP algorithm-based shielded door motor output force prediction method | |
| CN116186562B (en) | Encoder-based long text matching method | |
| CN114252706B (en) | Lightning early warning method and system | |
| CN105427583A (en) | LZW-coding-based road traffic data compression method |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230704 Address after: Room 603a, building 7, 266 Chuangyan Road, Qilin hi tech Industrial Development Zone, Nanjing, Jiangsu, 211100 Applicant after: Jiangsu Anshi Intelligent Technology Co.,Ltd. Address before: Room 603a, building 7, 266 Chuangyan Road, Qilin hi tech Industrial Development Zone, Nanjing, Jiangsu, 211100 Applicant before: Jiangsu Anshi Intelligent Technology Co.,Ltd. Applicant before: Fang Jian |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |