CN106406295B - Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state - Google Patents
Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state Download PDFInfo
- Publication number
- CN106406295B CN106406295B CN201611094444.1A CN201611094444A CN106406295B CN 106406295 B CN106406295 B CN 106406295B CN 201611094444 A CN201611094444 A CN 201611094444A CN 106406295 B CN106406295 B CN 106406295B
- Authority
- CN
- China
- Prior art keywords
- data
- door system
- rail traffic
- traffic vehicles
- health
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0243—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults model based detection method, e.g. first-principles knowledge model
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Train Traffic Observation, Control, And Security (AREA)
Abstract
The invention discloses a kind of rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state, initialization normal data, daily normal data and real-time running data are divided into for the operation data of door system, it can more accurately realize diagnosis and early warning, real-time monitoring parameter is monitored using motor and gating device I/O signal diagnoses door system typical fault, enrich the diagnosable fault type of door system, and by refinement fault signature, diagnostic accuracy is improved；Real time health degree model is established, Sub-health is diagnosed by the residual error feature of real-time parameter and model, provides new method for initial failure early warning；The inferior health problem slowly degenerated for a long time for door system, traditional diagnosis method is unable to monitor door system parameter substantially, and this method is by establishing degradation model, the Secular Variation Tendency of comparative analysis state parameter characteristic value, it can identify long-term degradation sub-health state, and realize early warning, it has a good application prospect.
Description
Technical field
The present invention relates to urban rail transit technology fields, and in particular to a kind of rail traffic vehicles door based on multi-state
System fault diagnosis and method for early warning.
Background technique
With the increasingly increasing of city size, urban track traffic carries huge traffic passenger flow pressure, for city
The real time monitoring and reliability of rail traffic vehicles state, it was also proposed that increasingly higher demands.
During urban rail transit vehicles operation, door system needs often to open and close, in addition human factor
It influences, will lead to Train door Frequent Troubles, frequent occurrence proteges of the powerful who stay with their benefactions like parasites's event.Therefore, the reliability and safety of door system, will be straight
Connect the operation for affecting the normal safety of urban rail transit vehicles.
The diagnosis logic of existing door system is relatively simple, and door system is run in the early stage and later period operation, peak period
When running in operation and library, different operating conditions analyze the state parameter of door system and diagnosis has more artificial or system interference, nothing
Method meets the positioning of failure and the demand of failure cause investigation.In addition, door system uses planned maintenance at present, when frame overhaul by
In lacking diagnostic means, whole updating is carried out to critical component, maintenance cost has been significantly greatly increased.Therefore, it is handed over for city rail
The research of the intelligent comprehensive diagnosis and method for early warning of the door system being open to traffic, for the job stability of urban rail transit vehicles
Promotion is of great significance.
Currently, the diagnosis and operation maintenance of existing door system, there are mainly two types of modes: first is that the regular inspection of maintenance personnel
Look into maintenance, time-consuming and laborious, poor reliability；Second is that foundation reliability analysis technology diagnoses fault type and reason after failure occurs,
Including reliability block diagram method, Bayesian network method, Fault Tree etc., but this kind of diagnostic techniques needs to rely on a large amount of sufficient experts
Based on knowledge base, does not have real-time, be not suitable for feature mining and initial failure early warning yet.
Research in terms of monitoring with the status safety of door system is constantly promoting and perfect, the door system of big data quantity
Real time data can be acquired, wherein including door system status information abundant.So for the door of urban rail transit vehicles
System, it is necessary to the resultant fault diagnosis and method for early warning that a kind of multi-state data with online real time execution are driving are invented,
To realize urban rail transit vehicles door from time-based checking maintenance mode to the checking maintenance mould based on state and risk
Formula transformation, is current urgent problem.
Summary of the invention
The purpose of the invention is to overcome the problems of existing rail traffic vehicles door system.Base of the invention
It, can be to the typical case of rail traffic vehicles door system in the rail traffic vehicles door system fault diagnosis and method for early warning of multi-state
Fault type is accurately identified, carries out early warning to door system inferior health and degradation trend, and rail traffic vehicles door is improved
System reliability of operation, has a good application prospect.
In order to achieve the above object, the technical scheme adopted by the invention is that:
Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state, it is characterised in that: including following
Step,
Step (A) acquires each car door by sensor in rail traffic vehicles door system driving motor and gating device
Operation data, and store into database；
The operation data includes corner, revolving speed, electric current, torque, temperature, I/O signal and the critical CAP value of switch triggering；
Step (B), the operation data of storage is pre-processed, is classified, specific as follows:
It will be new online and adjusted the operation data of normal, empty wagons running-in rail traffic vehicles door system and be defined as initially
Change normal data A, for the original state as rail traffic vehicles door system, and the ginseng as long-term degradation inferior health parameter
Examine benchmark；
By daily rail traffic vehicles it is online before, in factory library check when multiple switching door normal data be defined as daily
Normal data B, for the normalcy daily as rail traffic vehicles door system；
By daily rail traffic vehicles it is online after, the sensor and gating device of rail traffic vehicles door system acquire respectively in real time
The operation data of car door is defined as real time data C, as Real-Time Switch gate-shaped of the rail traffic vehicles door system in operation
State discriminates whether that current door system breaks down or inferior health in real time is abnormal；
The simulation test data that rail traffic vehicles are carried out in test-bed, for assist formulate typical fault rule and
Inferior health rule；
Step (C), it is online to daily rail traffic vehicles after rail traffic vehicles door system carry out Representative Faults Diagnosis,
Population characteristic value extraction is carried out to real time data C, determines whether real time data C belongs to failure classes exception, if preliminary judgement result
Be it is yes, then using typical fault decision rule determine fault type, and issue failure warning and fault parameter；If preliminary judgement knot
Fruit be it is no, then enter step (D)；
Step (D), it is online to daily rail traffic vehicles after rail traffic vehicles door system carry out real-time inferior health shape
State diagnosis, compares and analyzes real time data C and daily normal data B, according to daily normal data B building based on feature point
The data statistics model of cloth parameter is as real time health degree model, by the kinematic feature factor and in real time for calculating real time data C
The residual error of health degree model compares, switchs critical triggering CAP is compared with switch standard amount of over-pressurization range, sets health degree threshold
Value realizes the real-time inferior health quantization of rail traffic vehicles door system, judges whether fiducial value exceeds health degree threshold value, if exceeding
Health degree threshold value, according to the inferior health exception class of points and location determination rail traffic vehicles door system beyond health degree threshold value
Type, and continue the Frequency and front-rear switch door state of the monitoring and statistics inferior health Exception Type, finally combine expertise
Library diagnoses to obtain the current sub-health state of rail traffic vehicles door system, and according to the sub-health state, carries out initial failure
Early warning；If entering step (E) without departing from health degree threshold value；
Step (E), the inferior health for carrying out long-term degradation type to rail traffic vehicles door system diagnoses, to daily criterion numeral
It compares and analyzes according to B and initialization normal data A, using the typical characteristics for initializing normal data A as reference standard, establishes
Degradation model, it is pre- by least square method by analyzing the Secular Variation Tendency of daily normal data B critical eigenvalue
The critical eigenvalue in next period is surveyed, if the critical eigenvalue of prediction exceeds the range of degradation model, output track is handed over
Logical train-door system degeneration inferior health early warning；Otherwise, rail traffic vehicles door system operates normally.
Rail traffic vehicles door system fault diagnosis and method for early warning above-mentioned based on multi-state, it is characterised in that: step
Suddenly (C) carries out population characteristic value extraction to real time data C, determines whether real time data C belongs to failure classes exception, if preliminary judgement
As a result be it is yes, then using typical fault decision rule determine fault type, and issue failure warning and fault parameter, specifically include
Following steps,
(C1), the pretreatment of redundancy acquisition is aligned, removed to real time data C, and the source of an allusion is distinguished by global feature value
Type fault data；
(C2), the corresponding refinement global feature value of typical fault data, including opening-closing door time, total kilometres, maximum are extracted
Stroke, maximum current, effective current, motor stall number, motor rotation blockage number, motor stall position, motor rotation blockage position
It sets, enabling direction times of exercise and shutdown direction times of exercise；
(C3), refined global feature value is compared with typical fault rule, according to the cloth model of critical eigenvalue point
The correspondence fault type of determining current failure data is enclosed, the critical eigenvalue is opening-closing door time, total kilometres；
(C4), the correspondence fault type that will determine current failure data, by the failure warning of rail traffic vehicles door system
It is pushed to live customer service and related maintenance personnel with fault parameter information through mail, realizes that the real-time online of door system failure is examined
It is disconnected.
Rail traffic vehicles door system fault diagnosis and method for early warning above-mentioned based on multi-state, it is characterised in that: step
Suddenly (D) constructs the data statistics model based on feature distribution parameter as real time health degree model according to daily normal data B,
Specifically include following steps,
(D1) pretreatment of redundancy acquisition is aligned, removed to daily normal data B；Reality is distinguished by global feature value
When inferior health data；
(D2) according to motion feature respectively by corner, revolving speed and current data be divided into starting section, raising speed section, at the uniform velocity section,
Braking section and jogging section, and extract each section of time, maximum value, minimum value, mean value, standard deviation, the degree of bias, kurtosis feature；
(D3) the critical triggering CAP value tag of switch of switch gate process is acquired by gating device；
(D4) in order to measure real time data characteristic value, corner revolving speed, current parameters variation degree, with same door system
Daily normal data B and its segmentation feature value construct real time health degree model as training sample, such as public if training sample database X
Shown in formula (1),
Wherein, { [xi1,xi2,...,xim], i=1,2 ..., n } it is segmentation feature value, corner timing by i-th group of data
The characteristic quantity that data, revolving speed time series data and electric current time series data are constituted, n represent total sample group number, the number of m representative feature
Amount；
(D5) according to training sample database X, the average statistical of each characteristic quantity of its population sample is calculatedAnd standard deviationSuch as public affairs
Shown in formula (2) and formula (3),
(D6) 3sigma criterion is combined, the envelope range of real time health degree model is constructed: withAs early warning envelope
Boundary,As grave warning envelope boundary, real time health degree model is obtained.
Rail traffic vehicles door system fault diagnosis and method for early warning above-mentioned based on multi-state, it is characterised in that: step
Suddenly (D) switchs the range of critical triggering CAP between 6-23；The switch standard amount of over-pressurization range is between 1.5mm-6mm.
Rail traffic vehicles door system fault diagnosis and method for early warning above-mentioned based on multi-state, it is characterised in that: step
Suddenly (D), the value for switching critical triggering CAP is 18；The switch standard amount of over-pressurization is 4mm.
Rail traffic vehicles door system fault diagnosis and method for early warning above-mentioned based on multi-state, it is characterised in that: step
Suddenly (E) establishes degradation model using the typical characteristics for initializing normal data A as reference standard, specifically includes following step
Suddenly,
(E1) typical characteristics with significant change trend in initialization normal data A are filtered out；
(E2) enabling all typical characteristics Z is training sample database, obtains degradation model in conjunction with 3sigma criteria construction,
WithAs degeneration inferior health threshold value of warning,As fault pre-alarming threshold value, whereinFor training sample database Z's
The average statistical of each characteristic quantity of population sample；For the standard deviation of each characteristic quantity of population sample of training sample database Z.
The beneficial effects of the present invention are: rail traffic vehicles door system fault diagnosis of the invention based on multi-state and pre-
Alarm method is divided into initialization normal data, daily normal data and real-time running data for the operation data of door system, has
Effect reduce the influence that human factor and system degradation analyzes lasting door system data, can more accurately realize diagnosis with
Early warning monitors real-time monitoring parameter using motor and gating device I/O signal diagnoses door system typical fault, and enriching door system can
Type is diagnosed fault, and by refinement fault signature, improves diagnostic accuracy；Real time health degree model is established, by joining in real time
Several residual error features with model diagnose Sub-health, provide new method for initial failure early warning；Door system is delayed for a long time
Slowly the inferior health problem degenerated, traditional diagnosis method is unable to monitor door system parameter substantially, and this method is by establishing degeneration threshold
It is worth model, the Secular Variation Tendency of comparative analysis state parameter characteristic value can identify long-term degradation sub-health state, and realize
Early warning has a good application prospect.
Detailed description of the invention
Fig. 1 is the process of the rail traffic vehicles door system fault diagnosis and method for early warning of the invention based on multi-state
Figure.
Fig. 2 is the flow chart of Representative Faults Diagnosis of the invention.
Fig. 3 is the flow chart of real-time sub-health state diagnosis of the invention.
Fig. 4 is the flow chart of the inferior health diagnosis of long-term degradation type of the invention.
Specific embodiment
Below in conjunction with Figure of description, the present invention is further illustrated.
Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state of the invention, for door system
Operation data is divided into initialization normal data, daily normal data and real-time running data, effectively reduce human factor and
The influence that system degradation analyzes lasting door system data, can more accurately be realized diagnosis and early warning, be monitored using motor
Real-time monitoring parameter and gating device I/O signal diagnose door system typical fault, enrich the diagnosable fault type of door system, and logical
Refinement fault signature is crossed, diagnostic accuracy is improved；Real time health degree model is established, the residual error feature of real-time parameter and model is passed through
Sub-health is diagnosed, provides new method for initial failure early warning；For the inferior health problem that door system is slowly degenerated for a long time,
Traditional diagnosis method is unable to monitor door system parameter substantially, and this method is by establishing degradation model, comparative analysis state
The Secular Variation Tendency of parameter attribute value can identify long-term degradation sub-health state, and realize early warning, as shown in Figure 1,
Include the following steps,
Step (A) acquires each car door by sensor in rail traffic vehicles door system driving motor and gating device
Operation data, and store into database；
The operation data includes corner, revolving speed, electric current, torque, temperature, I/O signal and the critical CAP value of switch triggering；
Step (B), the operation data of storage is pre-processed, is classified, specific as follows:
It will be new online and adjusted the operation data of normal, empty wagons running-in rail traffic vehicles door system and be defined as initially
Change normal data A, for the original state as rail traffic vehicles door system, and the ginseng as long-term degradation inferior health parameter
Examine benchmark；
By daily rail traffic vehicles it is online before, in factory library check when multiple switching door normal data be defined as daily
Normal data B, for the normalcy daily as rail traffic vehicles door system；
By daily rail traffic vehicles it is online after, the sensor and gating device of rail traffic vehicles door system acquire respectively in real time
The operation data of car door is defined as real time data C, as Real-Time Switch gate-shaped of the rail traffic vehicles door system in operation
State discriminates whether that current door system breaks down or inferior health in real time is abnormal；
The simulation test data that rail traffic vehicles are carried out in test-bed, for assist formulate typical fault rule and
Inferior health rule；
Here initialization normal data A, daily normal data B, real time data C at least 30 groups or more virtual values；
Step (C), it is online to daily rail traffic vehicles after rail traffic vehicles door system carry out Representative Faults Diagnosis,
For process as shown in Fig. 2, carrying out population characteristic value extraction to real time data C, whether judgement real time data C belongs to failure classes exception,
If preliminary judgement result be it is yes, using typical fault decision rule determine fault type, and issue failure warning and failure ginseng
Number, specifically includes following steps,
(C1), the pretreatment of redundancy acquisition is aligned, removed to real time data C, and the source of an allusion is distinguished by global feature value
Type fault data；
(C2), the corresponding refinement global feature value of typical fault data, including opening-closing door time, total kilometres, maximum are extracted
Stroke, maximum current, effective current, motor stall number, motor rotation blockage number, motor stall position, motor rotation blockage position
It sets, enabling direction times of exercise and shutdown direction times of exercise；
(C3), refined global feature value is compared with typical fault rule, according to the cloth model of critical eigenvalue point
The correspondence fault type of determining current failure data is enclosed, the critical eigenvalue is opening-closing door time, total kilometres；
(C4), the correspondence fault type that will determine current failure data, by the failure warning of rail traffic vehicles door system
It is pushed to live customer service and related maintenance personnel with fault parameter information through mail, realizes that the real-time online of door system failure is examined
It is disconnected；
If preliminary judgement result be it is no, enter step (D)；
Step (D), it is online to daily rail traffic vehicles after rail traffic vehicles door system carry out real-time inferior health shape
State diagnosis, process is as shown in figure 3, compare and analyze real time data C and daily normal data B, according to daily normal data B
The data statistics model based on feature distribution parameter is constructed as real time health degree model, by the movement for calculating real time data C
Characteristic parameter compared with the residual error of real time health degree model, switch critical triggering CAP with switch standard amount of over-pressurization range compared
Compared with setting health degree threshold value realizes the real-time inferior health quantization of rail traffic vehicles door system, and it is strong to judge whether fiducial value exceeds
Kang Du threshold value, if exceeding health degree threshold value, according to the points and location determination rail traffic vehicles door system for exceeding health degree threshold value
The inferior health Exception Type of system, and continue the Frequency and front-rear switch door state of the monitoring and statistics inferior health Exception Type,
Finally expert knowledge library is combined to diagnose to obtain the current sub-health state of rail traffic vehicles door system, and according to the inferior health shape
State carries out initial failure early warning；If entering step (E) without departing from health degree threshold value；
The data statistics model based on feature distribution parameter is constructed as real time health degree mould according to daily normal data B
Type specifically includes following steps,
(D1) pretreatment of redundancy acquisition is aligned, removed to daily normal data B；Reality is distinguished by global feature value
When inferior health data；
(D2) according to motion feature respectively by corner, revolving speed and current data be divided into starting section, raising speed section, at the uniform velocity section,
Braking section and jogging section, and extract each section of time, maximum value, minimum value, mean value, standard deviation, the degree of bias, kurtosis feature；
(D3) the critical triggering CAP value tag of switch of switch gate process is acquired by gating device；
(D4) in order to measure real time data characteristic value, corner revolving speed, current parameters variation degree, with same door system
Daily normal data B and its segmentation feature value construct real time health degree model as training sample, such as public if training sample database X
Shown in formula (1),
Wherein, { [xi1,xi2,...,xim], i=1,2 ..., n } it is segmentation feature value, corner timing by i-th group of data
The characteristic quantity that data, revolving speed time series data and electric current time series data are constituted, n represent total sample group number, the number of m representative feature
Amount；
(D5) according to training sample database X, the average statistical of each characteristic quantity of its population sample is calculatedAnd standard deviationSuch as public affairs
Shown in formula (2) and formula (3),
(D6) 3sigma criterion is combined, the envelope range of real time health degree model is constructed: withAs early warning envelope
Boundary,As grave warning envelope boundary, real time health degree model is obtained.
The range of the critical triggering CAP of switch is between 6-23, and preferable 18, and can be adjusted according to the actual situation
It is whole；The switch standard amount of over-pressurization range is between 1.5mm-6mm, preferable 4mm, and can be adjusted according to the actual situation；
Directly it is determined as amount of over-pressurization exception beyond critical range, one it is found that be determined as door system inferior health, and provide corresponding etc. immediately
Grade Forewarning Measures；After the completion of door system real-time inferior health problem processing, need to reset the threshold value of warning of door system, it will be real-time
Inferior health model relearns data after adjustment；
Step (E), the inferior health for carrying out long-term degradation type to rail traffic vehicles door system diagnose, process such as Fig. 4 institute
Show, daily normal data B and initialization normal data A is compared and analyzed, to initialize the typical characteristics of normal data A
For reference standard, degradation model is established, by analyzing the Secular Variation Tendency of daily normal data B critical eigenvalue, is led to
The critical eigenvalue that least square method predicts the next period is crossed, if the critical eigenvalue of prediction exceeds the model of degradation model
It encloses, then output track vehicular traffic door system degeneration inferior health early warning；Otherwise, rail traffic vehicles door system operates normally, with
The typical characteristics for initializing normal data A are reference standard, establish degradation model, specifically include following steps,
(E1) typical characteristics with significant change trend in initialization normal data A are filtered out；
(E2) enabling all typical characteristics Z is training sample database, obtains degradation model in conjunction with 3sigma criteria construction,
WithAs degeneration inferior health threshold value of warning,As fault pre-alarming threshold value, whereinFor training sample database Z's
The average statistical of each characteristic quantity of population sample；For the standard deviation of each characteristic quantity of population sample of training sample database Z.
Above-mentioned 3sigma criterion is modeled according to mean value ± 3/6* standard deviation, and 3sigma criterion is this field skill
The art personnel conventional means used in modeling.
Since each door system on-line time does not have periodically, according to the data volume of daily normal data B and in advance
The period is surveyed, handling averagely at equal intervals is carried out to daily normal data B characteristic value, all kinds of characteristic value variation tendencies are carried out later
Least square fitting, and according to fitted trend predicted value and door system degradation model threshold value comparison, it is moved back according to the door system of prediction
Change severity and provides corresponding early warning, by the customer service of mail reminder scene and the timely inspection door system mode of related maintenance personnel, and
Early replacement degeneration components.
In conclusion the rail traffic vehicles door system fault diagnosis and method for early warning of the invention based on multi-state, energy
It is enough that the typical fault type of rail traffic vehicles door system is accurately identified, door system inferior health and degradation trend are carried out
Early warning improves rail traffic vehicles door system reliability of operation, has a good application prospect.
The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention.The technology of the industry
Personnel are it should be appreciated that the present invention is not limited to the above embodiments, and the above embodiments and description only describe this
The principle of invention, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes
Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appended claims and its
Equivalent thereof.
Claims (4)
1. rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state, it is characterised in that: including following step
Suddenly,
Step (A) acquires the operation of each car door by sensor in rail traffic vehicles door system driving motor and gating device
Data, and store into database；
The operation data includes corner, revolving speed, electric current, torque, temperature, I/O signal and the critical CAP value of switch triggering；
Step (B), the operation data of storage is pre-processed, is classified, specific as follows:
It will be new online and adjusted the operation data of normal, empty wagons running-in rail traffic vehicles door system and be defined as initialization mark
Quasi- data A, for the original state as rail traffic vehicles door system, and the reference base as long-term degradation inferior health parameter
It is quasi-；
By daily rail traffic vehicles it is online before, in factory library check when multiple switching door normal data be defined as daily standard
Data B, for the normalcy daily as rail traffic vehicles door system；
By daily rail traffic vehicles it is online after, the sensor and gating device of rail traffic vehicles door system acquire each car door in real time
Operation data be defined as real time data C, as Real-Time Switch door state of the rail traffic vehicles door system in operation, sentence
Not not whether current door system breaks down or inferior health in real time is abnormal；
The simulation test data that rail traffic vehicles are carried out in test-bed, for assisting formulation typical fault rule and Asia strong
Health rule；
Step (C), it is online to daily rail traffic vehicles after rail traffic vehicles door system carry out Representative Faults Diagnosis, to reality
When data C carry out population characteristic value extraction, determine real time data C whether belong to failure classes exception, if preliminary judgement result be it is yes,
Fault type is then determined using typical fault decision rule, and issues failure warning and fault parameter；If preliminary judgement result is
It is no, then enter step (D)；
Step (D), it is online to daily rail traffic vehicles after rail traffic vehicles door system carry out real-time sub-health state and examine
It is disconnected, real time data C and daily normal data B are compared and analyzed, joined according to daily normal data B building based on feature distribution
Several data statistics models is as real time health degree model, by the kinematic feature factor and real time health that calculate real time data C
Degree model residual error compare, switch critical triggering CAP with switch standard amount of over-pressurization range be compared, setting health degree threshold value reality
The real-time inferior health quantization of existing rail traffic vehicles door system, judges whether fiducial value exceeds health degree threshold value, if beyond health
Spend threshold value, according to beyond health degree threshold value points and location determination rail traffic vehicles door system inferior health Exception Type,
And continue the Frequency and front-rear switch door state of the monitoring and statistics inferior health Exception Type, finally examined in conjunction with expert knowledge library
It is disconnected to obtain the current sub-health state of rail traffic vehicles door system, and according to the sub-health state, carry out initial failure early warning；
If entering step (E) without departing from health degree threshold value；In step (D), according to daily normal data B building based on feature point
The data statistics model of cloth parameter specifically includes following steps as real time health degree model,
(D1) pretreatment of redundancy acquisition is aligned, removed to daily normal data B；Real-time Asia is distinguished by global feature value
Health data；
(D2) corner, revolving speed and current data are divided into respectively by starting section, raising speed section, at the uniform velocity section, deceleration according to motion feature
Section and jogging section, and extract each section of time, maximum value, minimum value, mean value, standard deviation, the degree of bias, kurtosis feature；
(D3) the critical triggering CAP value tag of switch of switch gate process is acquired by gating device；
(D4) in order to measure real time data characteristic value, corner revolving speed, current parameters variation degree, with the daily of same door system
Normal data B and its segmentation feature value construct real time health degree model, if training sample database X, such as formula as training sample
(1) shown in,
Wherein, { [xi1,xi2,...,xim], i=1,2 ..., n } ordinal number when being segmentation feature value, the corner by i-th group of data
According to the characteristic quantity that, revolving speed time series data and electric current time series data are constituted, n represents total sample group number, the quantity of m representative feature；
(D5) according to training sample database X, the average statistical of each characteristic quantity of its population sample is calculatedAnd standard deviationSuch as formula
(2) and shown in formula (3),
(D6) 3sigma criterion is combined, the envelope range of real time health degree model is constructed: withAs early warning envelope boundary,As grave warning envelope boundary, real time health degree model is obtained；
Step (E), the inferior health for carrying out long-term degradation type to rail traffic vehicles door system diagnoses, to daily normal data B
It is compared and analyzed with initialization normal data A, to initialize the typical characteristics of normal data A as reference standard, foundation is moved back
Change threshold model, by analyzing the Secular Variation Tendency of daily normal data B critical eigenvalue, is predicted by least square method
The critical eigenvalue in next period, if the critical eigenvalue of prediction exceeds the range of degradation model, output track traffic
Train-door system degeneration inferior health early warning；Otherwise, rail traffic vehicles door system operates normally；In step (E), to initialize mark
The typical characteristics of quasi- data A are reference standard, establish degradation model, specifically include following steps,
(E1) typical characteristics with significant change trend in initialization normal data A are filtered out；
(E2) enabling all typical characteristics Z is training sample database, obtains degradation model in conjunction with 3sigma criteria construction, withAs degeneration inferior health threshold value of warning,As fault pre-alarming threshold value, whereinFor the total of training sample database Z
The average statistical of each characteristic quantity of body sample；For the standard deviation of each characteristic quantity of population sample of training sample database Z.
2. the rail traffic vehicles door system fault diagnosis and method for early warning according to claim 1 based on multi-state,
Be characterized in that: step (C) carries out population characteristic value extraction to real time data C, and it is different to determine whether real time data C belongs to failure classes
Often, if preliminary judgement result be it is yes, using typical fault decision rule determine fault type, and issue failure warning and failure
Parameter specifically includes following steps,
(C1), the pretreatment of redundancy acquisition is aligned, removed to real time data C, and typical event is distinguished by global feature value
Hinder data；
(C2), the corresponding refinement global feature value of typical fault data, including opening-closing door time, total kilometres, maximum row are extracted
Journey, maximum current, effective current, motor stalling number, motor rotation blockage number, motor stall position, motor rotation blockage position,
Enabling direction times of exercise and shutdown direction times of exercise；
(C3), refined global feature value is compared with typical fault rule, the cloth range according to critical eigenvalue point is true
The correspondence fault type of settled prior fault data, the critical eigenvalue are opening-closing door time, total kilometres；
(C4), the correspondence fault type that will determine current failure data, by the failure warning of rail traffic vehicles door system and event
Hinder parameter information by mail push to live customer service and related maintenance personnel, realizes the real-time online diagnosis of door system failure.
3. the rail traffic vehicles door system fault diagnosis and method for early warning according to claim 1 based on multi-state,
Be characterized in that: step (D) switchs the range of critical triggering CAP between 6-23；The switch standard amount of over-pressurization range exists
Between 1.5mm-6mm.
4. the rail traffic vehicles door system fault diagnosis and method for early warning according to claim 1 based on multi-state,
Be characterized in that: step (D), the value for switching critical triggering CAP is 18；The switch standard amount of over-pressurization is 4mm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611094444.1A CN106406295B (en) | 2016-12-02 | 2016-12-02 | Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611094444.1A CN106406295B (en) | 2016-12-02 | 2016-12-02 | Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106406295A CN106406295A (en) | 2017-02-15 |
| CN106406295B true CN106406295B (en) | 2019-02-26 |
Family
ID=58083507
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611094444.1A Active CN106406295B (en) | 2016-12-02 | 2016-12-02 | Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106406295B (en) |
Families Citing this family (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107153841B (en) * | 2017-04-24 | 2020-02-18 | 南京康尼机电股份有限公司 | Sub-health prediction method for urban rail transit vehicle door system |
| CN107192565B (en) * | 2017-05-25 | 2019-05-28 | 南京康尼机电股份有限公司 | A kind of synchronization detecting method of subway vehicle door system exception operating condition and component degradation |
| JP6298562B1 (en) * | 2017-05-31 | 2018-03-20 | 伸和コントロールズ株式会社 | Status monitoring device, status monitoring method and program |
| CN109145328A (en) * | 2017-06-27 | 2019-01-04 | 中车株洲电力机车研究所有限公司 | Product degradation parameter identification method, device and properties of product appraisal procedure, system |
| CN107345860B (en) * | 2017-07-11 | 2019-05-31 | 南京康尼机电股份有限公司 | Rail vehicle door sub-health state recognition methods based on Time Series Data Mining |
| CN107860587B (en) * | 2017-11-08 | 2019-07-02 | 南京康尼机电股份有限公司 | Train-door system sub-health state early warning based on multi-feature fusion and diagnostic method |
| CN108107865B (en) * | 2017-11-13 | 2020-05-19 | 中车青岛四方机车车辆股份有限公司 | Fault diagnosis method and equipment for inner end door of motor train unit |
| DE102017130002A1 (en) * | 2017-12-14 | 2019-06-19 | Gebr. Bode Gmbh & Co. Kg | Method for condition-based maintenance of an access device |
| CN109034010B (en) * | 2018-07-06 | 2021-05-14 | 北京天泽智云科技有限公司 | On-line prediction method for lubrication failure of automatic door system |
| CN109625025B (en) * | 2018-12-13 | 2021-02-09 | 北京交大思诺科技股份有限公司 | BTM equipment early warning system |
| CN110515781B (en) * | 2019-07-03 | 2021-06-22 | 北京交通大学 | Complex system state monitoring and fault diagnosis method |
| CN112394703B (en) * | 2019-08-14 | 2022-06-10 | 中车时代电动汽车股份有限公司 | Vehicle fault management system |
| CN110717379A (en) * | 2019-08-28 | 2020-01-21 | 南京康尼机电股份有限公司 | Health assessment method for subway car door key components based on feature fusion |
| EP3792138A1 (en) * | 2019-09-13 | 2021-03-17 | Knorr-Bremse Gesellschaft mit beschränkter Haftung | Door system for a vehicle and method and device for providing maintenance information for a door system for a vehicle |
| CN110658807A (en) * | 2019-10-16 | 2020-01-07 | 上海仁童电子科技有限公司 | Vehicle fault diagnosis method, device and system |
| CN111082968A (en) * | 2019-11-13 | 2020-04-28 | 广西电网有限责任公司防城港供电局 | Network equipment security configuration compliance batch inspection method |
| CN111260822B (en) * | 2019-12-31 | 2022-07-26 | 杭州安脉盛智能技术有限公司 | Rail transit vehicle health state analysis method and terminal based on big data |
| CN111260823B (en) * | 2020-01-16 | 2021-12-24 | 南京康尼机电股份有限公司 | Fault diagnosis method based on I/O (input/output) measuring point fault dependency matrix |
| CN111426343A (en) * | 2020-03-17 | 2020-07-17 | 东华大学 | Automatic door fault detection early warning method |
| CN112560165A (en) * | 2020-06-11 | 2021-03-26 | 中车青岛四方机车车辆股份有限公司 | Urban rail vehicle and passenger room door fault diagnosis method thereof |
| CN112001532B (en) * | 2020-08-04 | 2024-03-01 | 交控科技股份有限公司 | Switch fault prediction method and device, electronic equipment and storage medium |
| CN111797944A (en) * | 2020-08-04 | 2020-10-20 | 上海仁童电子科技有限公司 | Vehicle door abnormity diagnosis method and device |
| CN111999580B (en) * | 2020-08-14 | 2023-12-01 | 佳都科技集团股份有限公司 | Method and device for detecting subway platform gate, computer equipment and storage medium |
| CN112127718B (en) * | 2020-09-25 | 2022-01-28 | 牛玉涛 | Method for detecting running state of sliding door lock of platform door system |
| CN112348078A (en) * | 2020-11-09 | 2021-02-09 | 南京工程学院 | Gate machine controller with sub-health pre-diagnosis and fault type clustering functions |
| CN112801558B (en) * | 2021-04-07 | 2021-07-30 | 北京瑞莱智慧科技有限公司 | Optimization method and device of process parameter adjustment action decision model |
| CN113377090B (en) * | 2021-08-12 | 2021-10-29 | 新誉轨道交通科技有限公司 | Pressure change model for motor train unit, and method, system and device for diagnosing air door fault |
| CN113806969B (en) * | 2021-10-26 | 2022-12-27 | 国家石油天然气管网集团有限公司 | Compressor unit health prediction method based on time domain data correlation modeling |
| CN114296105A (en) * | 2021-12-27 | 2022-04-08 | 中国第一汽车股份有限公司 | Method, device, equipment and storage medium for determining positioning fault reason |
| CN115688493A (en) * | 2023-01-03 | 2023-02-03 | 深圳市信润富联数字科技有限公司 | Punching abnormity monitoring method and device, electronic equipment and storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102270271B (en) * | 2011-05-03 | 2014-03-19 | 北京中瑞泰科技有限公司 | Equipment failure early warning and optimizing method and system based on similarity curve |
| CN102829967B (en) * | 2012-08-27 | 2015-12-16 | 中国舰船研究设计中心 | A kind of time domain fault recognition method based on regression model index variation |
| CN103345207B (en) * | 2013-05-31 | 2015-06-24 | 北京泰乐德信息技术有限公司 | Mining analyzing and fault diagnosis system of rail transit monitoring data |
| CN103699698B (en) * | 2014-01-16 | 2017-03-29 | 北京泰乐德信息技术有限公司 | A kind of being based on improves Bayesian rail transit fault identification method and system |
| CN104091070B (en) * | 2014-07-07 | 2017-05-17 | 北京泰乐德信息技术有限公司 | Rail transit fault diagnosis method and system based on time series analysis |
| US9984513B2 (en) * | 2014-12-23 | 2018-05-29 | Palo Alto Resarch Center Incorporated | System and method for determining vehicle component conditions |
| CN105045256B (en) * | 2015-07-08 | 2018-11-20 | 北京泰乐德信息技术有限公司 | Rail traffic real-time fault diagnosis method and system based on date comprision |
-
2016
- 2016-12-02 CN CN201611094444.1A patent/CN106406295B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106406295A (en) | 2017-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106406295B (en) | Rail traffic vehicles door system fault diagnosis and method for early warning based on multi-state | |
| CN110262463B (en) | Rail transit platform door fault diagnosis system based on deep learning | |
| CN108569607B (en) | Elevator fault early warning method based on bidirectional gating cyclic neural network | |
| CN104091070B (en) | Rail transit fault diagnosis method and system based on time series analysis | |
| CN111274737A (en) | Method and system for predicting remaining service life of mechanical equipment | |
| CN110110870A (en) | A kind of equipment fault intelligent control method based on event graphical spectrum technology | |
| CN104809878A (en) | Method for detecting abnormal condition of urban road traffic by utilizing GPS (Global Positioning System) data of public buses | |
| CN104134010B (en) | Satellite fault diagnosis method for discrete type data based on Naive Bayes | |
| CN110414154B (en) | Fan component temperature abnormity detection and alarm method with double measuring points | |
| CN110865924B (en) | Health degree diagnosis method and health diagnosis framework for internal server of power information system | |
| WO2018049695A1 (en) | Door controller with integrated data collection and transmission device and transmission processing method thereof | |
| CN110209999A (en) | A kind of mobile unit failure trend prediction method | |
| CN107844067B (en) | A kind of gate of hydropower station on-line condition monitoring control method and monitoring system | |
| CN104268381A (en) | Satellite fault diagnosing method based on AdaBoost algorithm | |
| CN108709744B (en) | Motor bearings method for diagnosing faults under a kind of varying load operating condition | |
| CN107153841A (en) | A kind of inferior health Forecasting Methodology of urban rail transit vehicles door system | |
| CN108776452B (en) | Special equipment field maintenance monitoring method and system | |
| CN103359137A (en) | Turnout fault early warning method | |
| CN113581253A (en) | Method and device for determining state of electric empty switch machine | |
| CN105930957A (en) | Risk early warning method for electric energy meter automatic verification line | |
| CN108919104A (en) | A kind of circuit breaker failure diagnostic method based on Fisher identification and classification method | |
| CN105302476B (en) | A kind of reliability data online acquisition for nuclear power plant equipment analyzes storage system and its storage method | |
| Asada et al. | Development of an effective condition monitoring system for AC point machines | |
| CN110610016A (en) | Method for predicting rail transit stopping problem based on big data machine learning | |
| CN110411686A (en) | The quiet dynamic image holography condition health monitoring diagnostic method of bridge and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |