CN112208506B - Intelligent fault detection method for air brake device of railway wagon - Google Patents
Intelligent fault detection method for air brake device of railway wagon Download PDFInfo
- Publication number
- CN112208506B CN112208506B CN202011011009.4A CN202011011009A CN112208506B CN 112208506 B CN112208506 B CN 112208506B CN 202011011009 A CN202011011009 A CN 202011011009A CN 112208506 B CN112208506 B CN 112208506B
- Authority
- CN
- China
- Prior art keywords
- data
- value
- railway wagon
- calculating
- difference
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T17/00—Component parts, details, or accessories of power brake systems not covered by groups B60T8/00, B60T13/00 or B60T15/00, or presenting other characteristic features
- B60T17/18—Safety devices; Monitoring
- B60T17/22—Devices for monitoring or checking brake systems; Signal devices
- B60T17/228—Devices for monitoring or checking brake systems; Signal devices for railway vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Regulating Braking Force (AREA)
- Valves And Accessory Devices For Braking Systems (AREA)
Abstract
The invention relates to the technical field of railway wagon brake detection, in particular to an intelligent fault detection method for an air brake device of a railway wagon, which comprises the following steps: acquiring air pressure measurement data for a sensor under normal and different damage degree working conditions of each component to form a training data set; dividing accumulated pressure measurement data acquired by a sensor into single different braking state data, wherein different braking states comprise braking, pressure maintaining and relieving; thirdly, judging a braking state according to the accumulated difference function, and extracting data characteristics; and fourthly, building a random forest, and inputting the extracted data characteristics to train model parameters. The invention can better detect the fault type and the damage degree of the air brake device of the railway wagon.
Description
Technical Field
The invention relates to the technical field of railway wagon brake detection, in particular to an intelligent fault detection method for an air brake device of a railway wagon.
Background
The railway wagon air brake system components comprise a brake main pipe, an auxiliary air cylinder, a brake cylinder, a 120-valve and the like. The brake main pipe provides compressed air for each vehicle, and when air pressure in the pipe is reduced, the train has a braking effect, the air pressure is increased, and the train has a relieving effect. The 120 valve is the core of the brake device, the brake main pipe, the brake cylinder and the auxiliary reservoir are all connected with the 120 valve, and the positions of the internal slide valve and the check valve are adjusted according to the change of the air pressure of the brake main pipe to open or block a relevant channel, so that the control of the braking or the relieving of the train is realized. The auxiliary reservoir is used for storing compressed air and providing the compressed air to the brake cylinder when the train brakes. Brake cylinders are the components of a vehicle air brake system that perform braking or mitigation commands. With the continuous improvement of the loading capacity of the railway freight car and the running speed of the train, high requirements are provided for the performance of the air brake system, and the structure of the brake system is designed to be more complex so as to achieve higher performance. Due to the dependence and coupling relation of each part in the air brake system, faults are difficult to find, and strong ambiguity exists between fault symptoms and reasons, so that great difficulty is brought to fault diagnosis and positioning.
At present, an overrun alarm mechanism is generally adopted by a detection system to monitor a railway wagon air brake system, and the faults are generally caused by serious leakage, so that the performance is reduced due to the degradation of state components of the brake system, and the faults cannot be well monitored. Although fault detection methods such as a fault tree theory, an expert system, a fuzzy theory and the like can achieve the effect of brake fault diagnosis to a certain extent, the fault diagnosis precision, the fault positioning accuracy and the fault diagnosis efficiency are low, and the interpretability of fault occurrence is not strong. The traditional model is generally modeled based on expert knowledge and experience, the fact cannot be accurately reflected, and meanwhile, the model is not easy to change after being built and is not easy to accept external information and man-machine interaction.
Disclosure of Invention
It is an object of the present invention to provide an intelligent fault detection method for a railway wagon air brake device that overcomes some or all of the disadvantages of the prior art.
The intelligent fault detection method for the air brake device of the railway wagon comprises the following steps of:
acquiring air pressure measurement data for a sensor under normal and different damage degree working conditions of each component to form a training data set;
dividing accumulated pressure measurement data acquired by a sensor into single different braking state data, wherein different braking states comprise braking, pressure maintaining and relieving;
judging the braking state according to the accumulated difference function, and extracting data characteristics, wherein the data characteristics comprise a peak value, a slope, pressure increase and decrease duration time, pressure increase and decrease quantity, pressure difference between a brake cylinder, an auxiliary air cylinder and a main pipe, and variance characteristics of adjacent vehicles;
fourthly, building a random forest, inputting extracted data characteristics to train model parameters, wherein the training contents comprise: load state judgment, normal and abnormal classification, fault part classification, fault degree classification and fault degree prediction.
Preferably, in the first step, the method for obtaining the air pressure measurement data comprises: and installing an air pressure sensor on the air brake device of the railway wagon to be detected, and digitizing the acquired signals to obtain air pressure measurement data.
Preferably, in the second step, modeling is carried out by utilizing pressure measurement data generated by the railway wagon in braking, relieving and pressure maintaining states, and the modeling corresponds to pressurization, depressurization and stabilization measures respectively; after the data acquisition instrument collects data of a certain time, firstly, the data is subjected to sliding smooth filtering treatment:
searching an extreme point of the air pressure in real-time measurement by simply comparing adjacent values, and taking the extreme point as a brake state division end point; counting the collected normal braking and release duration time, and removing sections with duration time obviously not meeting the requirements due to data noise and the like; finally, using the accumulated differenceAnd judging the braking state.
Preferably, in step three, the data feature extraction step is as follows:
3.1, calculating the pressure increase and decrease value of the monitored vehicle and the characteristic value of the steady state before and after the occurrence, correspondingly calculating the characteristic value of the data of the adjacent vehicles and calculating the average value;
3.2, taking the characteristic value of the sensor at the same position of the adjacent vehicle as a reference, and solving a difference value between the adjacent vehicle and the monitored vehicle;
3.3, calculating difference values of the sensor difference values, the initial values and the final values of different parts of the monitored vehicle, correspondingly calculating difference values of adjacent vehicles and calculating an average value;
3.4, measuring the sum of variance between data variables by a sensor at the same position of the monitored vehicle and an adjacent vehicle as an IVV characteristic; increasing and decreasing the duration as DVV characteristics, correspondingly calculating characteristic values of adjacent vehicle data and averaging;
3.5, taking the characteristic value of the sensor at the same position of the adjacent vehicle as a reference, and solving a difference value between the adjacent vehicle and the monitored vehicle;
3.6, taking the pressure increasing and decreasing efficiency as the SVV slope characteristic;
3.7, introducing some statistical indexes of maximum value, minimum value and standard deviation.
Preferably, the steady state characteristic values are a difference value, a start value and an end value;
the initial value calculation method comprises the following steps: extracting the first 20 points of the measured data, calculating the standard deviation, taking the average value as an initial value if the standard deviation does not exceed the threshold, and halving the data volume if the standard deviation exceeds the threshold until the threshold is met;
the method for calculating the termination value comprises the following steps: after 20 points of the measurement data are extracted, calculating the standard deviation, if the standard deviation does not exceed the threshold, taking the average value as a termination value, and if the standard deviation exceeds the threshold, halving the data volume until the threshold is met;
the difference value calculation method comprises the following steps: difference-end value-start value;
DVV calculation method: firstly, measuring data difference and data smoothing, and calculating a starting point and an end point which exceed a threshold value, wherein the length between the two points is a duration characteristic;
the SVV calculation method comprises the following steps: difference value DVV.
Preferably, in the fourth step, the training step is: firstly, judging the load state of the railway wagon; then, the air brake system of each vehicle is respectively judged to be abnormal in braking, and whether a fault occurs is judged; secondly, performing secondary data scanning on the abnormal data information, detecting the position and the type of the fault and giving out the recognition probability; and finally, classifying the fault degree and predicting the degradation trend.
The invention has the following beneficial effects:
1. the installation of the multi-position sensor and the air pressure detection can represent the internal association between equipment faults by using data characteristics conveniently and visually;
2. the characteristics of the measured data are modeled by comparing the single-vehicle characteristics with multiple vehicles of the sample, so that the fault can be detected and separated in time, and meanwhile, the modeling is carried out aiming at abnormal components or pipelines, so that the method can be used for predicting serious faults and providing certain reference for timely maintenance and replacement of the components;
3. the random forest adopts an integrated learning idea, overfitting is not easy to generate, structure learning and parameter learning can be carried out through a large amount of fault sample data, and a network structure and node probability distribution which can reflect the objective fact of the fault can be obtained. The random forest has low calculation cost, does not need to rely on a GPU to complete training, flexibly updates fault characteristics, tree structures and node parameters, and simultaneously learns the contribution of different characteristic variables to fault detection through different interpretations of a decision tree. In practical application, fault sample data are continuously accumulated for continuous change of the working state and the working environment of equipment, and a random forest can learn new fault samples to improve detection performance.
Drawings
Fig. 1 is a flowchart of an intelligent fault detection method for an air brake device of a railway wagon in embodiment 1;
FIG. 2 is a flowchart of data feature extraction in example 1;
fig. 3 is a flow chart of the model parameter training in example 1.
FIG. 4 is a schematic view showing the structure of an air brake system and the installation position of a sensor in embodiment 1;
fig. 5 is a schematic view of brake state data of air pressure of a brake main pipe in the air brake system of the railway freight car in embodiment 1;
fig. 6 is a schematic view of a fault diagnosis result and a trend prediction result of the air brake system of the railway freight car in embodiment 1.
Detailed Description
For a further understanding of the invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings and examples. It is to be understood that the examples are illustrative of the invention and not limiting.
Example 1
As shown in fig. 1, the present embodiment provides an intelligent fault detection method for an air brake device of a railway wagon, which includes the following steps:
acquiring air pressure measurement data for a sensor under normal and different damage degree working conditions of each component to form a training data set;
dividing accumulated pressure measurement data acquired by a sensor into single different braking state (braking, pressure maintaining and relieving) data;
judging the braking state according to the accumulated difference function, and extracting data characteristics, wherein the data characteristics comprise a peak value, a slope, pressure increase and decrease duration time, pressure increase and decrease quantity, pressure difference between a brake cylinder, an auxiliary air cylinder and a main pipe, and variance characteristics of adjacent vehicles;
fourthly, building a random forest, inputting extracted data characteristics to train model parameters, wherein the training contents comprise: load state judgment, normal and abnormal classification, fault part classification, fault degree classification and fault degree prediction.
In the first step, the method for obtaining the air pressure measurement data comprises the following steps: and installing an air pressure sensor on the air brake device of the railway wagon to be detected, and digitizing the acquired signals to obtain air pressure measurement data. As shown in fig. 4, the schematic diagram of the structure of the air brake system and the installation position of the sensor is shown, and the sensor can preferably measure the air pressure measurement data of the brake main pipe, the valve-in of 120, the auxiliary reservoir, the valve-out of 120 and the brake cylinder.
In the second step, modeling is carried out by utilizing pressure measurement data generated by the railway wagon in braking, relieving and pressure maintaining states, and the pressure measurement data correspond to pressurization, depressurization and stabilization measures respectively; after the data acquisition instrument collects data of a certain time, firstly, the data is subjected to sliding smooth filtering treatment:
searching an extreme point of the air pressure in real-time measurement by simply comparing adjacent values, and taking the extreme point as a brake state division end point; counting the collected normal braking and release duration time, and removing sections with duration time obviously not meeting the requirements due to data noise and the like; finally, using the accumulated differenceAnd judging the braking state.
Fig. 5 is a schematic diagram showing the braking state data of the air pressure of the brake main pipe in the air brake system of the railway wagon.
As shown in fig. 2, in the third step, the data feature extraction steps are as follows:
3.1, calculating the pressure increase and decrease value of the monitored vehicle and the characteristic value of the steady state before and after the occurrence, correspondingly calculating the characteristic value of the data of the adjacent vehicles and calculating the average value;
3.2, taking the characteristic value of the sensor at the same position of the adjacent vehicle as a reference, and solving a difference value between the adjacent vehicle and the monitored vehicle;
3.3, calculating difference values of the difference values, the initial values and the final values of the sensors at different parts of the monitored vehicle by considering the connectivity of each sensing part and the change of the air pressure caused by the fault, correspondingly calculating the difference values of adjacent vehicles and calculating the average value;
3.4, measuring the sum of variance between data variables by a sensor at the same position of the monitored vehicle and an adjacent vehicle as an IVV characteristic; increasing and decreasing the duration as DVV characteristics, correspondingly calculating characteristic values of adjacent vehicle data and averaging;
3.5, taking the characteristic value of the sensor at the same position of the adjacent vehicle as a reference, and solving a difference value between the adjacent vehicle and the monitored vehicle;
3.6, taking the pressure increasing and decreasing efficiency as the SVV slope characteristic;
3.7, introducing some statistical indexes of maximum value, minimum value and standard deviation.
The steady state characteristic values are a difference value, a starting value and an ending value;
the initial value calculation method comprises the following steps: extracting the first 20 points of the measured data, calculating the standard deviation, taking the average value as an initial value if the standard deviation does not exceed the threshold, and halving the data volume if the standard deviation exceeds the threshold until the threshold is met;
the method for calculating the termination value comprises the following steps: after 20 points of the measurement data are extracted, calculating the standard deviation, if the standard deviation does not exceed the threshold, taking the average value as a termination value, and if the standard deviation exceeds the threshold, halving the data volume until the threshold is met;
the difference value calculation method comprises the following steps: difference-end value-start value;
DVV calculation method: firstly, measuring data difference and data smoothing, and calculating a starting point and an end point which exceed a threshold value, wherein the length between the two points is a duration characteristic;
the SVV calculation method comprises the following steps: difference value DVV.
As shown in fig. 3, in the fourth step, the training step is: firstly, judging the load state of the railway wagon; then, the air brake system of each vehicle is respectively judged to be abnormal in braking, and whether a fault occurs is judged; secondly, performing secondary data scanning on the abnormal data information, detecting the position and the type of the fault and giving out the recognition probability; and finally, classifying the fault degree and predicting the degradation trend.
The random forest is a typical Bagging integration algorithm, the base classifiers are decision trees, the number of the base classifiers is set to be 1000, the number of features randomly selected by each decision tree is 7, and the optimal parameter for optimization is set to be 7. The braking conditions corresponding to different load states are greatly different, so that the load state of the railway wagon is judged firstly; identifying the running state of the current air brake system, and classifying the running state into 11 types of common faults, such as a main brake pipe, a cylinder branch pipe, an auxiliary air cylinder, a leather cup, a plug, a brake adjuster, a slide valve, a check valve, a plate perforation, a diaphragm plate, a dust cover and the like; the failure degree mainly comprises the steps of punching 1, 2, 3 and 45mm holes in the plate, punching 3, 4 and 5mm holes in the leather cups, punching 1, 3, 4 and 5mm holes in the plate, sorting 1, 2, 4, 5 and 8mm holes in the auxiliary air cylinders, sorting 1, 2, 4, 5 and 8mm holes in the main pipes, sorting 1, 2, 4, 5 and 8mm holes in the cylinder branch pipes, sorting 105, 205, 255 and 257mm holes in the brake adjuster, drawing and predicting a failure development curve according to the material characteristics of different braking parts, and displaying a failure diagnosis result and a trend prediction result as shown in FIG. 6.
The multi-site sensor installation and air pressure detection of the present embodiment may be used to visually characterize the intrinsic association between equipment failures from a convenient perspective.
The embodiment models the characteristics of the measured data by comparing the characteristics of a single vehicle with multiple vehicles of a sample, can detect and separate faults in time, and simultaneously models abnormal parts or pipelines, so that the method can be used for predicting serious faults and providing certain reference for timely maintenance and replacement of the parts.
The random forest of the embodiment adopts an integrated learning idea, overfitting is not easy to generate, structure learning and parameter learning can be carried out through a large amount of fault sample data, and a network structure and node probability distribution which can reflect the objective fact of the fault can be obtained. The random forest has low calculation cost, does not need to rely on a GPU to complete training, flexibly updates fault characteristics, tree structures and node parameters, and simultaneously learns the contribution of different characteristic variables to fault detection through different interpretations of a decision tree. In practical application, fault sample data are continuously accumulated for continuous change of the working state and the working environment of equipment, and a random forest can learn new fault samples to improve detection performance.
The present invention and its embodiments have been described above schematically, without limitation, and what is shown in the drawings is only one of the embodiments of the present invention, and the actual structure is not limited thereto. Therefore, if the person skilled in the art receives the teaching, without departing from the spirit of the invention, the person skilled in the art shall not inventively design the similar structural modes and embodiments to the technical solution, but shall fall within the scope of the invention.
Claims (6)
1. The intelligent fault detection method for the air brake device of the railway wagon is characterized by comprising the following steps of: the method comprises the following steps:
acquiring air pressure measurement data for a sensor under normal and different damage degree working conditions of each component to form a training data set;
dividing accumulated pressure measurement data acquired by a sensor into single different braking state data, wherein different braking states comprise braking, pressure maintaining and relieving;
judging the braking state according to the accumulated difference function, and extracting data characteristics, wherein the data characteristics comprise a peak value, a slope, pressure increase and decrease duration time, pressure increase and decrease quantity, pressure difference between a brake cylinder, an auxiliary air cylinder and a main pipe, and variance characteristics of adjacent vehicles;
fourthly, building a random forest, inputting extracted data characteristics to train model parameters, wherein the training contents comprise: load state judgment, normal and abnormal classification, fault part classification, fault degree classification and fault degree prediction.
2. The intelligent fault detection method for a railway wagon air brake device as claimed in claim 1, wherein: in the first step, the method for obtaining the air pressure measurement data comprises the following steps: and installing an air pressure sensor on the air brake device of the railway wagon to be detected, and digitizing the acquired signals to obtain air pressure measurement data.
3. The intelligent fault detection method for a railway wagon air brake device as claimed in claim 1, wherein: in the second step, modeling is carried out by utilizing pressure measurement data generated by the railway wagon in braking, relieving and pressure maintaining states, and the pressure measurement data correspond to pressurization, depressurization and stabilization measures respectively; after the data acquisition instrument collects data of a certain time, firstly, the data is subjected to sliding smooth filtering treatment:
searching an extreme point of the air pressure in real-time measurement by simply comparing adjacent values, and taking the extreme point as a brake state division end point; counting the collected normal braking and release duration time, and removing sections with duration time obviously not meeting the requirements due to data noise; finally, using the accumulated differenceAnd judging the braking state.
4. The intelligent fault detection method for a railway wagon air brake device as claimed in claim 1, wherein: in the third step, the data feature extraction steps are as follows:
3.1, calculating the pressure increase and decrease value of the monitored vehicle and the characteristic value of the steady state before and after the occurrence, correspondingly calculating the characteristic value of the data of the adjacent vehicles and calculating the average value;
3.2, taking the characteristic value of the sensor at the same position of the adjacent vehicle as a reference, and solving a difference value between the adjacent vehicle and the monitored vehicle;
3.3, calculating difference values of the sensor difference values, the initial values and the final values of different parts of the monitored vehicle, correspondingly calculating difference values of adjacent vehicles and calculating an average value;
3.4, measuring the sum of variance between data variables by a sensor at the same position of the monitored vehicle and an adjacent vehicle as an IVV characteristic; increasing and decreasing the duration as DVV characteristics, correspondingly calculating characteristic values of adjacent vehicle data and averaging;
3.5, taking the characteristic value of the sensor at the same position of the adjacent vehicle as a reference, and solving a difference value between the adjacent vehicle and the monitored vehicle;
3.6, taking the pressure increasing and decreasing efficiency as the SVV slope characteristic;
3.7, introducing some statistical indexes of maximum value, minimum value and standard deviation.
5. The intelligent fault detection method for a railway wagon air brake device as claimed in claim 4, wherein: the steady state characteristic values are a difference value, a starting value and an ending value;
the initial value calculation method comprises the following steps: extracting the first 20 points of the measured data, calculating the standard deviation, taking the average value as an initial value if the standard deviation does not exceed the threshold, and halving the data volume if the standard deviation exceeds the threshold until the threshold is met;
the method for calculating the termination value comprises the following steps: after 20 points of the measurement data are extracted, calculating the standard deviation, if the standard deviation does not exceed the threshold, taking the average value as a termination value, and if the standard deviation exceeds the threshold, halving the data volume until the threshold is met;
the difference value calculation method comprises the following steps: difference-end value-start value;
DVV calculation method: firstly, measuring data difference and data smoothing, and calculating a starting point and an end point which exceed a threshold value, wherein the length between the two points is a duration characteristic;
the SVV calculation method comprises the following steps: difference value DVV.
6. The intelligent fault detection method for a railway wagon air brake device as claimed in claim 1, wherein: in the fourth step, the training step is as follows: firstly, judging the load state of the railway wagon; then, the air brake system of each vehicle is respectively judged to be abnormal in braking, and whether a fault occurs is judged; secondly, performing secondary data scanning on the abnormal data information, detecting the position and the type of the fault and giving out the recognition probability; and finally, classifying the fault degree and predicting the degradation trend.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011011009.4A CN112208506B (en) | 2020-09-23 | 2020-09-23 | Intelligent fault detection method for air brake device of railway wagon |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011011009.4A CN112208506B (en) | 2020-09-23 | 2020-09-23 | Intelligent fault detection method for air brake device of railway wagon |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112208506A CN112208506A (en) | 2021-01-12 |
CN112208506B true CN112208506B (en) | 2021-07-06 |
Family
ID=74051290
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011011009.4A Active CN112208506B (en) | 2020-09-23 | 2020-09-23 | Intelligent fault detection method for air brake device of railway wagon |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112208506B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113022520B (en) * | 2021-05-27 | 2021-08-13 | 天津所托瑞安汽车科技有限公司 | Adaptive braking method, apparatus and storage medium |
CN113534774B (en) * | 2021-06-28 | 2022-07-01 | 长沙理工大学 | Fault prediction method, system and medium for subway brake system |
CN113804469B (en) * | 2021-09-30 | 2024-01-12 | 中车青岛四方机车车辆股份有限公司 | Rail vehicle braking state judging system, method, storage medium and equipment |
CN115158272B (en) * | 2022-06-30 | 2023-05-30 | 东风商用车有限公司 | Leakage prediction method and device for air brake system |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105730431B (en) * | 2016-01-29 | 2018-04-20 | 清华大学 | EMU checking cylinder fault monitoring method and fault monitoring system |
CN108802525A (en) * | 2018-06-06 | 2018-11-13 | 浙江宇天科技股份有限公司 | Equipment fault intelligent Forecasting based on small sample |
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 |
CN110293949B (en) * | 2019-06-06 | 2021-09-24 | 山东科技大学 | Method for detecting tiny faults of air brake system of high-speed train |
US11551488B2 (en) * | 2019-08-22 | 2023-01-10 | GM Global Technology Operations LLC | Adaptive fault diagnostic system for motor vehicles |
CN111114519B (en) * | 2020-01-03 | 2022-04-08 | 中国铁路郑州局集团有限公司科学技术研究所 | Railway vehicle brake fault prediction method and health management system |
CN111177939B (en) * | 2020-01-03 | 2023-04-18 | 中国铁路郑州局集团有限公司科学技术研究所 | Method for predicting brake cylinder pressure of train air brake system based on deep learning |
-
2020
- 2020-09-23 CN CN202011011009.4A patent/CN112208506B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN112208506A (en) | 2021-01-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112208506B (en) | Intelligent fault detection method for air brake device of railway wagon | |
CN106338406B (en) | The on-line monitoring of train traction electric drive system and fault early warning system and method | |
CN112505549B (en) | New energy automobile battery abnormity detection method based on isolated forest algorithm | |
CN109213121B (en) | Method for diagnosing clamping cylinder fault of fan braking system | |
CN109949823B (en) | DWPT-MFCC and GMM-based in-vehicle abnormal sound identification method | |
CN108069308A (en) | A kind of electric staircase failure diagnosis method based on sequential probability | |
CN112393906B (en) | Method for diagnosing, classifying and evaluating health of weak signal fault of bogie bearing of metro vehicle | |
CN116881745B (en) | Pressure transmitter abnormality monitoring method based on big data | |
CN110580492A (en) | Track circuit fault precursor discovery method based on small fluctuation detection | |
CN112182912B (en) | Manufacturing equipment spindle bearing health assessment method based on probability description and spectrum analysis | |
CN104677997A (en) | Transformer oil chromatography online monitoring differential early warning method | |
CN110486350B (en) | Electro-hydraulic servo valve fault diagnosis method and device, storage medium and electronic equipment | |
CN110553789A (en) | state detection method and device of piezoresistive pressure sensor and brake system | |
CN109214522B (en) | Equipment performance degradation evaluation method based on multiple attributes | |
CN208207843U (en) | A kind of slush pump trouble-shooter | |
CN115409131A (en) | Production line abnormity detection method based on SPC process control system | |
CN114266197A (en) | Method for diagnosing equipment fault of hydropower station | |
CN113657217A (en) | Concrete state recognition model based on improved BP neural network | |
CN109255201A (en) | A kind of ball screw assembly, health evaluating method based on SOM-MQE | |
CN117406026A (en) | Power distribution network fault detection method suitable for distributed power supply | |
KR20230104527A (en) | Partial Discharge Monitoring System And Partial Discharge Monitoring Method | |
CN117365869A (en) | Self-adaptive early warning strategy design method for wind turbine blade tower sweeping faults | |
CN114359663A (en) | Hydraulic plunger pump intelligent fault diagnosis method based on pressure signals | |
CN108716463A (en) | A kind of method for diagnosing faults of reciprocating compressor ring air flap | |
CN116951328B (en) | Intelligent drainage pipeline operation monitoring system based on big data |
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 |