CN109325310B - High-speed train intermittent fault detection method based on multiple T-square control diagram - Google Patents
High-speed train intermittent fault detection method based on multiple T-square control diagram Download PDFInfo
- Publication number
- CN109325310B CN109325310B CN201811250111.2A CN201811250111A CN109325310B CN 109325310 B CN109325310 B CN 109325310B CN 201811250111 A CN201811250111 A CN 201811250111A CN 109325310 B CN109325310 B CN 109325310B
- Authority
- CN
- China
- Prior art keywords
- time
- fault
- intermittent
- following
- steps
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 25
- 238000010586 diagram Methods 0.000 title claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000008034 disappearance Effects 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 230000001934 delay Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 5
- 230000008030 elimination Effects 0.000 abstract description 5
- 238000003379 elimination reaction Methods 0.000 abstract description 5
- 238000003745 diagnosis Methods 0.000 abstract description 4
- 238000004088 simulation Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/08—Railway vehicles
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Debugging And Monitoring (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses a high-speed train intermittent fault detection method based on a multiple T-square control diagram, belonging to the field of fault diagnosis; the method can detect intermittent faults with smaller amplitude and limited duration, deduce the occurrence time and disappearance time of the intermittent faults, and remarkably improve the detection effect of the data driving method on the intermittent faults; the method comprises the following steps: training data is collected and preprocessed, intermittent fault parameters of a key system of the high-speed train are analyzed, on-line fault detection and false alarm elimination are carried out, and time window selection and time of occurrence and disappearance of the intermittent fault are deduced in real time. The invention effectively guarantees the actual application requirements of the intermittent fault detection of the key system of the high-speed train.
Description
Technical Field
The invention belongs to the field of fault diagnosis, and particularly relates to a high-speed train intermittent fault detection method based on a multiple T-square control diagram.
Background
In recent years, fault detection of a critical system of a high-speed train has become a subject of hot research. However, for many years, people mostly focus on the problem of detecting continuous faults only, and the problem of detecting intermittent faults is less studied. On the other hand, with the rapid development of electronic, information and other technologies, a special type of fault, i.e., intermittent fault, different from the conventional continuous fault form, is becoming more important. The high-speed train brake control system operates in a complex environment and is easy to generate intermittent faults. On the one hand, the electronic brake control unit is composed of a complex electronic circuit, and the intermittent fault of the controller is caused by loosening of the control circuit due to cold joint aging and the like. In addition, the running environment of the high-speed train is complex, and the vehicle-mounted sensor is easily affected by vibration, electromagnetic interference and the like, so that the sensor is in intermittent failure.
Intermittent faults are faults of a type that have a finite duration and that can automatically disappear without external compensation measures, allowing the system to resume acceptable performance. Compared with continuous faults, intermittent faults occur randomly and repeatedly, and have obvious cumulative effects, namely, the occurrence frequency and the duration of the faults gradually increase with the passage of time, and the intermittent faults finally evolve into continuous faults. Because the occurrence and disappearance of intermittent faults have randomness, the diagnosis standard of the intermittent faults requires that the occurrence and disappearance time of the faults are detected simultaneously, the diagnosis performance meets the requirements that the occurrence time of the faults is determined before the current fault disappears, and the disappearance time of the faults is determined before the next fault occurs. The detection result is more beneficial to analyzing the running state of the system, making reasonable maintenance and repair strategies and designing fault-tolerant control laws aiming at intermittent faults.
The characteristics of small intermittent fault amplitude and limited duration make it difficult for the conventional data-driven fault detection method to be directly applied to intermittent fault detection
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides a high-speed train intermittent fault detection method based on a multiple T-square control diagram, which is reasonable in design, overcomes the defects in the prior art and has good effect.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
the intermittent fault detection method of the high-speed train based on the multiple T-square control diagram comprises the following steps:
step 1: offline training; the method specifically comprises the following steps:
step 1.1: constructing a normal working condition measurement matrix:
[X 1 ,X 2 ,…,X k ,…,X N ]∈R m×N ;
wherein X is k ∈R m×1 Is a column vector; m is the detectedThe number of sensors included in the system; n is the number of independent samples included in each sensor;
Step 1.3: according to experience or by analyzing historical fault data, the intermittent fault direction xi of the key system of the high-speed train is given q Lower boundary of fault amplitudeFault duration lower bound->And failure vanishing time lower bound->And calculates the following formula:
wherein W is * And W is # Two time windows;W=1,2,…,W # ;/>is an upper bound for theoretically detecting delays to faults; />Is the upper bound of the detection delay for the disappearance of a fault in theory;
step 1.4: time window w=1, 2, …, W according to a given confidence level α # Respectively calculateI.e.
Wherein F is a (p, N-p) is the upper score when the F distribution confidence level of the degree of freedom is p and the N-p is alpha, and is recorded
Step 2: detecting on-line faults; the method specifically comprises the following steps:
step 2.1: on-line collection of new measurement samplesAnd time window w=1, 2, …, W # Calculating statistics of each T-square control diagram in real time>I.e.
Step 2.2: time window w=1, 2, …, W # The alarm time of each T-party control chart is updated in real time: alarm for fault occurrenceAnd failure vanishing alarm->I.e.
Step 2.3: the method for eliminating false alarms on line specifically comprises the following steps:
step 2.3.1: if for W=W # The following is not true
Restarting step 2; otherwise, executing the step 2.4;
step 2.3.2: if W epsilon [ W ] exists * ,W # ]If the formula (1) is not established, restarting the step (2); otherwise, executing the step 2.4;
step 2.3.3: if W exists, W' E [ W ] * ,W # ]So that the following is true
Wherein the method comprises the steps ofIf the set is empty, restarting the step 2, otherwise executing the step 2.4;
step 2.4: the time window is selected, and the method specifically comprises the following steps:
step 2.4.1: sequentially checking w= { W * …,1}, if the formula (1) is not established, setting W o =w+1; otherwise wo=1;
step 2.4.2: sequentially checking w= { W * …, wo if W' ∈ (W, W) # ]So that (2) is established, then wo=w+1 is reset;
step 2.4.3: for W= { W * ,…,W o ' MeterCalculation of
Step 2.4.4: sequentially checking w= { W * ,…,W o If W' ∈ (W, W) # ]So that the following formula is established, reset W o =W+1;
Step 2.5: the intermittent fault occurrence and disappearance time deducing specifically comprises the following steps:
step 2.5.1: final inference for calculating intermittent fault occurrence and disappearance timeAnd->I.e.
Step 2.5.2: deducing that the intermittent fault is inOccurs within the moment>And vanishes in time.
The invention has the beneficial technical effects that:
the invention provides a new method for detecting intermittent faults, which provides a selection criterion of multiple time windows, can detect intermittent faults with smaller amplitude and limited duration, deduces the occurrence time and disappearance time of the intermittent faults, and obviously improves the detection effect of the data driving method on the intermittent faults.
Drawings
Fig. 1 is a schematic diagram of simulation results of intermittent fault recurrence and disappearance.
Fig. 2 is a schematic diagram of an original T-party control detection result when false alarm elimination is not performed.
Fig. 3 is a schematic diagram of intermittent fault detection results after false alarm elimination and time window selection.
Fig. 4 is a flow chart of the method of the present invention.
Detailed Description
The invention is described in further detail below with reference to the attached drawings and detailed description:
1. the intermittent fault detection method of the high-speed train based on the multiple T-party control diagram has a flow shown in fig. 4 and comprises the following steps:
step 1: offline training; the method specifically comprises the following steps:
step 1.1: constructing a normal working condition measurement matrix:
[X 1 ,X 2 ,…,X k ,…,X N ]∈R m×N ;
wherein X is k ∈R m×1 Is a column vector; m is the number of sensors contained in the detected system; n is the number of independent samples included in each sensor;
Step 1.3: according to experience or by analyzing historical fault data, the intermittent fault direction xi of the key system of the high-speed train is given q Lower boundary of fault amplitudeFault duration lower bound->And failure vanishing time lower bound->And calculates the following formula:
wherein W is * And W is # Two time windows;W=1,2,…,W # ;/>is an upper bound for theoretically detecting delays to faults; />Is the upper bound of the detection delay for the disappearance of a fault in theory;
step 1.4: time window w=1, 2, …, W according to a given confidence level α # Respectively calculateI.e.
Wherein F is a (p, N-p) is the degree of freedom pCounting the upper score when the F distribution confidence level of N-p is alpha
Step 2: detecting on-line faults; the method specifically comprises the following steps:
step 2.1: on-line collection of new measurement samplesAnd time window w=1, 2, …, W # Calculating statistics of each T-square control diagram in real time>I.e.
Step 2.2: time window w=1, 2, …, W # The alarm time of each T-party control chart is updated in real time: alarm for fault occurrenceAnd failure vanishing alarm->I.e.
Step 2.3: the method for eliminating false alarms on line specifically comprises the following steps:
step 2.3.1: if for W=W # The following is not true
Restarting step 2; otherwise, executing the step 2.4;
step 2.3.2: if W epsilon [ W ] exists * ,W # ]If the formula (1) is not established, restarting the step (2); otherwise, executing the step 2.4;
step 2.3.3: if W exists, W' E [ W ] * ,W # ]So that the following is true
Wherein the method comprises the steps ofIf the set is empty, restarting the step 2, otherwise executing the step 2.4; />
Step 2.4: the time window is selected, and the method specifically comprises the following steps:
step 2.4.1: sequentially checking w= { W * …,1}, if the formula (1) is not established, setting W o =w+1; otherwise W o =1;
Step 2.4.2: sequentially checking w= { W * ,…,W o If W' ∈ (W, W) # ]So that the expression (2) is established, reset W o =W+1;
Step 2.4.3: for W= { W * ,…,W o Calculation of
Step 2.4.4: sequentially checking w= { W * …, wo if W' ∈ (W, W) # ]So that the following formula holds, then wo=w+1 is reset;
step 2.5: the intermittent fault occurrence and disappearance time deducing specifically comprises the following steps:
step 2.5.1: final inference for calculating intermittent fault occurrence and disappearance timeAnd->I.e.
Step 2.5.2: deducing that the intermittent fault is inOccurs within the moment>And vanishes in time.
2. Simulation study
The simulation model is selected as follows:
firstly, 5000 training data are generated according to the model, and the training data represent observation data under normal working conditions; then generating 500 test data according to the model, and adding direction xi from 201 st data q =[0.2425,0.9701] T Intermittent failure of (a); intermittent fault amplitude lower bound isThe duration of intermittent faults and the lower limit of the disappearance time are +.>The simulation results are shown in fig. 1,2 and 3.
The left ordinate in fig. 1 shows the true value of the 1/4 failure amplitude, and the solid line in the figure represents the recurrence and disappearance of intermittent failure. The ordinate on the right side of fig. 2 represents the detection result of the 1/10 window length, and the dashed line in the figure represents the original T-square control detection result when false alarm elimination is not performed. The broken line in fig. 3 shows the intermittent fault detection result after false alarm elimination and time window selection, if the corresponding time window has no broken line when the fault occurs, it represents that we do not select the time window when predicting the occurrence and disappearance time of the fault in real time.
Table 1 gives our results of the inference of time to occurrence and disappearance of each intermittent fault, where μ q Representing the actual occurrence time of intermittent faults, the two following two columns represent the deduction of our method on the occurrence time, namely in the algorithm stepSimilarly, v q The real vanishing time representing intermittent faults, the two latter columns are the deduction of our method on the vanishing time, namely +.>
TABLE 1
It is easy to see that the method can effectively reduce false alarm and infer the occurrence and disappearance time of intermittent faults, and the method can detect intermittent faults with smaller amplitude and limited duration, thereby remarkably improving the detection effect of the data driving method on the intermittent faults.
It should be understood that the above description is not intended to limit the invention to the particular embodiments disclosed, but to limit the invention to the particular embodiments disclosed, and that the invention is not limited to the particular embodiments disclosed, but is intended to cover modifications, adaptations, additions and alternatives falling within the spirit and scope of the invention.
Claims (1)
1. A high-speed train intermittent fault detection method based on a multiple T-party control diagram is characterized in that: the method comprises the following steps:
step 1: offline training; the method specifically comprises the following steps:
step 1.1: constructing a normal working condition measurement matrix:
[X 1 ,X 2 ,…,X k ,…,X N ]∈R m×N ;
wherein X is k ∈R m×1 Is a column vector; m is the number of sensors contained in the detected system; n is the number of independent samples included in each sensor;
Step 1.3: according to experience or by analyzing historical fault data, the intermittent fault direction xi of the key system of the high-speed train is given q Lower boundary of fault amplitudeFault duration lower bound->And failure vanishing time lower bound->And calculates the following formula:
wherein W is * And W is # Two time windows;W=1,2,…,W # ;/>is an upper bound for theoretically detecting delays to faults; />Is the upper bound of the detection delay for the disappearance of a fault in theory;
step 1.4: time window w=1, 2, …, W according to a given confidence level α # Respectively calculateI.e.
Wherein F is a (p, N-p) is the upper score when the F distribution confidence level of the degree of freedom is p and N-p is alpha, and delta is recorded 2 =δ 1 2 ;
Step 2: detecting on-line faults; the method specifically comprises the following steps:
step 2.1: on-line collection of new measurement samplesAnd time window w=1, 2, …, W # Calculating statistics of each T-square control diagram in real time>I.e.
Step 2.2: time window w=1, 2, …, W # The alarm time of each T-party control chart is updated in real time: alarm for fault occurrenceAnd failure vanishing alarm->I.e.
Step 2.3: the method for eliminating false alarms on line specifically comprises the following steps:
step 2.3.1: if for W=W # The following is not true
Restarting step 2; otherwise, executing the step 2.4;
step 2.3.2: if W epsilon [ W ] exists * ,W # ]If the formula (1) is not established, restarting the step (2); otherwise, executing the step 2.4; step 2.3.3: if W exists, W' E [ W ] * ,W # ]So that the following is true
Wherein the method comprises the steps ofIf the set is empty, restarting the step 2, otherwise executing the step 2.4;
step 2.4: the time window is selected, and the method specifically comprises the following steps:
step 2.4.1: sequentially checking w= { W * …,1}, if the formula (1) is not established, setting W o =w+1; otherwise W o =1;
Step 2.4.2: sequentially checking w= { W * ,…,W o If W' ∈ (W, W) # ]So that the expression (2) is established, reset W o =w+1; step 2.4.3: for W= { W * ,…,W o Calculation of
Step 2.4.4: sequentially checking w= { W * ,…,W o If W' ∈ (W, W) # ]So that the following formula is established, reset W o =W+1;
Step 2.5: the intermittent fault occurrence and disappearance time deducing specifically comprises the following steps:
step 2.5.1: final estimation mu of intermittent fault occurrence and disappearance time q ,And v q ,/>I.e.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811250111.2A CN109325310B (en) | 2018-10-25 | 2018-10-25 | High-speed train intermittent fault detection method based on multiple T-square control diagram |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811250111.2A CN109325310B (en) | 2018-10-25 | 2018-10-25 | High-speed train intermittent fault detection method based on multiple T-square control diagram |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109325310A CN109325310A (en) | 2019-02-12 |
CN109325310B true CN109325310B (en) | 2023-05-05 |
Family
ID=65261847
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811250111.2A Active CN109325310B (en) | 2018-10-25 | 2018-10-25 | High-speed train intermittent fault detection method based on multiple T-square control diagram |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109325310B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109855855B (en) * | 2019-03-13 | 2020-10-09 | 山东科技大学 | Intermittent fault detection method for closed-loop brake system of high-speed train |
CN114664058B (en) * | 2022-01-29 | 2023-08-18 | 上海至冕伟业科技有限公司 | Overall fault early warning system and method for fire fighting water system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0479718A (en) * | 1990-07-20 | 1992-03-13 | Kansai Electric Power Co Inc:The | Detecting method of intermittent ground fault |
WO2010115474A1 (en) * | 2009-04-10 | 2010-10-14 | Areva T&D Uk Ltd | Method and system for transient and intermittent earth fault detection and direction determination in a three-phase median voltage electric power distribution system |
CN104697804A (en) * | 2015-03-24 | 2015-06-10 | 清华大学 | Method and system for detecting and separating intermittent faults of train active suspension system |
CN107703740A (en) * | 2017-07-10 | 2018-02-16 | 山东科技大学 | A kind of robust interval sensor fault diagnosis method of bullet train critical system |
-
2018
- 2018-10-25 CN CN201811250111.2A patent/CN109325310B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH0479718A (en) * | 1990-07-20 | 1992-03-13 | Kansai Electric Power Co Inc:The | Detecting method of intermittent ground fault |
WO2010115474A1 (en) * | 2009-04-10 | 2010-10-14 | Areva T&D Uk Ltd | Method and system for transient and intermittent earth fault detection and direction determination in a three-phase median voltage electric power distribution system |
CN104697804A (en) * | 2015-03-24 | 2015-06-10 | 清华大学 | Method and system for detecting and separating intermittent faults of train active suspension system |
CN107703740A (en) * | 2017-07-10 | 2018-02-16 | 山东科技大学 | A kind of robust interval sensor fault diagnosis method of bullet train critical system |
Non-Patent Citations (1)
Title |
---|
鄢镕易 ; 何潇 ; 周东华.一类存在参数摄动的线性随机系统的鲁棒间歇故障诊断方法.第26届中国过程控制会议.2015,全文. * |
Also Published As
Publication number | Publication date |
---|---|
CN109325310A (en) | 2019-02-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103776654B (en) | The method for diagnosing faults of multi-sensor information fusion | |
CN102829967A (en) | Time-domain fault identifying method based on coefficient variation of regression model | |
Tcherniak et al. | Vibration-based SHM system: application to wind turbine blades | |
CN105787561A (en) | Recurrent neural network model construction method and gearbox fault detection method and device | |
CN109325310B (en) | High-speed train intermittent fault detection method based on multiple T-square control diagram | |
CN106199412A (en) | A kind of permanent magnet mechanism high-pressure vacuum breaker method of fault pattern recognition | |
CN112193959A (en) | Method and system for detecting abnormal sound of elevator | |
CN108444696A (en) | A kind of gearbox fault analysis method | |
CN112067701B (en) | Fan blade remote auscultation method based on acoustic diagnosis | |
CN112632845B (en) | Data-based mini-reactor online fault diagnosis method, medium and equipment | |
CN106429689A (en) | Elevator maintenance system based on Internet-of-things big data support | |
CN104864985A (en) | Train axle temperature sensor fault detection method and apparatus | |
CN109489931A (en) | A kind of abnormal impact real-time detection method | |
CN106679847A (en) | Electric power equipment fault diagnosing method and apparatus | |
CN108287964B (en) | Gray cloud reasoning structure damage identification method based on acceleration inner product vector | |
CN114738205A (en) | Method, device, equipment and medium for monitoring state of floating fan foundation | |
CN112711850A (en) | Unit online monitoring method based on big data | |
CN110887899B (en) | Turbine blade water erosion defect monitoring and identifying method | |
CN112664410B (en) | Big data-based modeling method for unit online monitoring system | |
CN113344275B (en) | Floating platform wave climbing online forecasting method based on LSTM model | |
Ma et al. | Two-stage damage identification based on modal strain energy and revised particle swarm optimization | |
JP3394817B2 (en) | Plant diagnostic equipment | |
CN109872511B (en) | Self-adaptive two-stage alarm method for monitoring axial displacement sudden change | |
EP2956751B1 (en) | Method and monitoring device for monitoring a structure | |
CN110262447A (en) | A kind of ACS closed-loop system Fault Locating Method based on ANN |
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 |