CN114021452A - Urban rail vehicle door motor performance detection and evaluation method - Google Patents
Urban rail vehicle door motor performance detection and evaluation method Download PDFInfo
- Publication number
- CN114021452A CN114021452A CN202111288095.8A CN202111288095A CN114021452A CN 114021452 A CN114021452 A CN 114021452A CN 202111288095 A CN202111288095 A CN 202111288095A CN 114021452 A CN114021452 A CN 114021452A
- Authority
- CN
- China
- Prior art keywords
- performance
- motor
- vehicle door
- door
- urban rail
- 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.)
- Pending
Links
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
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Power-Operated Mechanisms For Wings (AREA)
Abstract
A method for detecting and evaluating the performance of a door motor of an urban rail vehicle comprises the steps of 1) defining the door motor, namely an acceleration section current peak value, a high-speed section current peak value and a deceleration section current peak value as door motor performance characteristic parameters; 2) constructing a vehicle door motor performance index by using Support Vector Data Description (SVDD); 3) defining the SVDD hypersphere radius obtained by training a motor performance characteristic parameter construction sample when a vehicle door motor has no fault and no performance degradation and the vehicle door is normally opened and closed as a vehicle door motor performance index reference value; 4) calculating the deviation of the door performance index value from the reference value, and 5) carrying out performance detection and performance evaluation based on the deviation value. Detecting abnormal operation of the system by analyzing the deviation condition of the performance index of the door motor of the urban rail vehicle during actual operation and the reference value of the performance index, and further identifying early faults; meanwhile, the performance state of the vehicle door motor can be quantitatively evaluated, and support is provided for realizing state-based maintenance decision.
Description
Technical Field
The invention relates to a method for detecting and evaluating the performance of a door motor of an urban rail vehicle based on support vector data description and fuzzy evaluation, and belongs to the technical field of urban rail vehicle health management.
Background
The vehicle door system is a key subsystem of the urban rail vehicle, and the reliability and the safety of the vehicle door system are directly related to the safety and the driving service quality of passengers, so that the vehicle door system is a key concern in the operation management process of the urban rail vehicle. The driving motor is used as a key component for providing driving force for opening and closing the vehicle door, and the performance state of the driving motor determines the running performance and the running safety of the vehicle door. Therefore, the running state of the vehicle door driving motor needs to be monitored, the fault of the vehicle door motor is detected in time, the performance state of the vehicle door driving motor is evaluated, and decision support is provided for overhauling and replacing the motor, so that the vehicle door driving motor can be processed before the motor fails, accidents such as parking, passenger clearing and the like caused by the vehicle door fault caused by the motor fault are avoided, the driving service quality is improved, and the driving safety is ensured.
At present, a large number of sensors are additionally arranged on an urban rail vehicle door system to monitor working parameters of the urban rail vehicle door system, and the urban rail vehicle door system has certain diagnostic capability and can detect partial faults. In addition, the urban rail vehicle can actively discover and prevent various faults including a vehicle door system by performing daily overhaul. However, at present, the state detection and daily maintenance of the door motor of the urban rail vehicle are mainly focused on the diagnosis and maintenance of the occurred fault, and an applicable method and technology for early detection of the fault are not available. Meanwhile, there is no feasible method for evaluating the overall performance of the door motor and analyzing and evaluating the system performance degradation process.
Disclosure of Invention
The invention aims to provide a method for detecting the performance of a door motor of an urban rail vehicle, which detects the abnormal operation of a system by analyzing the deviation condition of a performance index of the door motor of the urban rail vehicle during actual operation and a reference value of the door motor, thereby identifying the early failure of the system; meanwhile, the performance state of the vehicle door motor can be quantitatively evaluated based on the deviation condition of the expected value of the actual performance index, and support is provided for realizing state-based maintenance decision.
In order to achieve the aim, the invention provides a method for detecting and evaluating the performance of a door motor of an urban rail vehicle, which is characterized by comprising the following steps:
1) defining the vehicle door motor, namely an acceleration section current peak value, a high-speed section current peak value and a deceleration section current peak value as vehicle door motor performance characteristic parameters;
2) describing the SVDD by using the support vector data to construct the performance index of the vehicle door motor;
3) defining the SVDD hypersphere radius obtained by training a motor performance characteristic parameter construction sample when a vehicle door motor has no fault and no performance degradation and the vehicle door is normally opened and closed as a vehicle door motor performance index reference value;
4) calculating the deviation of the vehicle door performance index value and the reference value;
5) and performing performance detection and performance evaluation based on the deviation value.
Preferably, the SVDD hypersphere radius obtained by training the constructed sample in step 3) is specifically that the following optimization problem is constructed:
s.t.||Xi-a||2≤R2+εi;
εi≥0
the dual problem of the above optimization problem is obtained:
s.t.0≤αi≤C;
wherein, K (.) is a kernel function, and the invention adopts a Gaussian kernel function; training by adopting a training sample to obtain the spherical center and the radius of the hypersphere:
further, constructing a new sample X for the characteristic parameters of the vehicle door motor defined by the step 1) based on the SVDD model trained in the step 3)newCalculating the distance d between the new sample and the center of the SVDD hyperspherenew,
preferably, the performance of the door driving motor is detected based on Δ R and the time sequence formed by Δ R, and the specific method is as follows:
1) if the delta R exceeds a given threshold value delta, the performance of the driving motor of the door of the urban rail vehicle is over-limit, and the door needs to be overhauled, and faults or other factors are eliminated;
2) if the delta R time sequence has sudden change, the driving motor of the door of the urban rail vehicle is possibly in fault, and the maintenance and troubleshooting are required;
3) if the delta R time sequence generates a trend, the trend indicates that the driving motor of the door of the urban rail vehicle has a fault or the performance is stably reduced, and close attention needs to be paid and fault diagnosis needs to be carried out.
Preferably, the health quantitative evaluation is carried out on the urban rail brake vehicle braking system according to the magnitude of the delta R, and the specific method comprises the following steps: evaluating the performance of the vehicle door driving motor based on the delta R calculated in the step 4); because a sample point can be formed by opening and closing the vehicle door once, the performance evaluation can be carried out once by opening and closing the vehicle door once; considering that the degradation of the door driving motor is a gradual process, so that the performance evaluation can be performed once in a longer period, such as 1 day, the Δ R is first normalized as follows:
wherein δ is a normalization threshold, α normalization factor is between 0 and 1, and k can be specifically selected according to actual conditions; and then selecting a membership function, and evaluating the performance of the vehicle door driving motor according to the normalized delta R in the step 4).
Preferably, the performance evaluation of the vehicle door driving motor adopts a triangular membership function, the normalized delta R is calculated, the membership degree of a predefined motor grade is calculated, and the grade with the maximum membership degree is taken as the performance grade of the vehicle door driving motor.
Compared with the existing urban rail vehicle door motor fault detection and performance evaluation technology, the method can fully utilize the actual operation data analysis technology to detect the performance degradation and early faults of the urban rail vehicle door motor, thereby providing powerful support for the state maintenance of the urban rail vehicle brake system. In addition, in the data analysis modeling process, only the normal operation data of the door motor of the urban rail vehicle is needed to construct a training sample, a large number of fault samples are not needed, the data is easier to obtain, model training data do not need to be obtained through a fault injection test and the like, and the operability of the method is improved.
Drawings
FIG. 1 is a schematic view of a motor operating stage of a vehicle door;
fig. 2 is a performance detection and evaluation flow chart.
Detailed Description
Referring to fig. 1 and 2, the specific implementation process of the present invention is as follows:
1) according to the working characteristics of an urban rail vehicle door system and a driving motor, the current of the door driving motor in the door opening and closing process is divided into three data subsections of acceleration section current, high-speed section current and deceleration section current, and the current peak values in the three data subsections are respectively taken as characteristic parameters, namely the current peak value of the acceleration section, the current peak value of the high-speed section and the current peak value of the deceleration section are marked as Ipa,Iph,Ips;
2) Defining the performance index of the door driving motor of the urban rail vehicle, constructing the performance index by adopting a Support Vector Data Description (SVDD) model, and using the characteristic parameter I defined in the step 1) to construct the modelpa,Iph,IpsFormed sample points Xi={Ipa,Iph,IpsTaking the distance d between the sample point and the SVDD hypersphere center a as a performance index;
3) defining the running behavior of the vehicle door driving motor as normal behavior when the vehicle door is normally opened without failure or performance degradation, and adopting a sample set { X) formed by characteristic parameters corresponding to the normal behaviori}nTraining the SVDD model defined in the step 2), and taking the SVDD hypersphere radius as a reference performance index when the vehicle door driving motor works normallyAnd n is the number of samples formed by the characteristic parameters meeting the definition of the normal behaviors.
Specifically, the following optimization problem is constructed:
s.t.||Xi-a||2≤R2+εi;
εi≥0
the dual problem of the above optimization problem is obtained:
s.t.0≤αi≤C;
wherein, K (.) is a kernel function, and the invention adopts a Gaussian kernel function. Training by adopting a training sample to obtain the spherical center and the radius of the hypersphere:
from the above formula, each support vector XαAll correspond to a radius of the center of the hyper-sphere, therefore, the invention adopts 96 percent of statistical control limit as the final hyper-sphere radius and the reference performance indexThe statistical control limit may be adjusted according to the actual application.
4) Constructing a new sample X for the characteristic parameters of the vehicle door motor defined in the step 1) based on the SVDD model trained in the step 3)newCalculating the distance d between the new sample and the center of the SVDD hyperspherenew,
5) and (3) detecting the performance of the vehicle door driving motor based on the delta R and the time sequence formed by the delta R, wherein the performance detection flow is shown in figure 2.
The specific method comprises the following steps: analysis of Δ R and its constituent time series:
5.1) if the delta R exceeds a given threshold delta, indicating that the performance of the door driving motor of the urban rail vehicle is over-limit, and needing to be overhauled, and eliminating faults or other factors;
5.2) if the delta R time sequence has a sudden change, the driving motor of the door of the urban rail vehicle is possibly in fault, and the maintenance and fault removal are required;
and 5.3) if the time sequence of the delta R generates a trend, the trend indicates that the driving motor of the door of the urban rail vehicle has a fault or the performance is stably reduced, and close attention needs to be paid and fault diagnosis needs to be carried out.
In addition, the health quantitative evaluation can be carried out on the urban rail brake vehicle braking system according to the size of the delta R, and decision support is provided for the overhaul of urban rail vehicles.
6) Evaluating the performance of the door driving motor based on the Δ R calculated in the step 4). Since one sample point is formed every time the door is opened and closed, performance evaluation can be performed every time the door is opened and closed. Considering that the deterioration of the door driving motor is a gradual process, the performance evaluation can be performed once in a long period, for example, 1 day.
The specific method comprises the following steps: first, Δ R is normalized as follows:
wherein δ is a normalization threshold, α is a normalization factor, which is between 0 and 1, and k can be specifically selected according to actual conditions. And then selecting a membership function, and evaluating the performance of the vehicle door driving motor according to the normalized delta R in the step 4). The invention adopts a triangular membership function (in practical application, other membership functions can be selected according to the performance degradation condition of the vehicle door driving motor), calculates the normalized delta R, calculates the membership degree of the predefined motor grade, and takes the grade with the maximum membership degree as the performance grade of the vehicle door driving motor.
Claims (6)
1. A method for detecting and evaluating the performance of a door motor of an urban rail vehicle is characterized by comprising the following steps:
1) defining the vehicle door motor, namely an acceleration section current peak value, a high-speed section current peak value and a deceleration section current peak value as vehicle door motor performance characteristic parameters;
2) describing the SVDD by using the support vector data to construct the performance index of the vehicle door motor;
3) defining the SVDD hypersphere radius obtained by training a motor performance characteristic parameter construction sample when a vehicle door motor has no fault and no performance degradation and the vehicle door is normally opened and closed as a vehicle door motor performance index reference value;
4) calculating the deviation of the vehicle door performance index value and the reference value;
5) and performing performance detection and performance evaluation based on the deviation value.
2. The method for detecting and evaluating the performance of the door motor of the urban rail vehicle according to claim 1, characterized in that: step 3), specifically constructing the following optimization problem of the SVDD hypersphere radius obtained by sample construction training:
s.t.||Xi-a||2≤R2+εi;
εi≥0
the dual problem of the above optimization problem is obtained:
s.t.0≤αi≤C;
wherein, K (.) is a kernel function, and the invention adopts a Gaussian kernel function; training by adopting a training sample to obtain the spherical center and the radius of the hypersphere:
3. the method for detecting and evaluating the performance of the door motor of the urban rail vehicle according to claim 1, characterized in that: step 3) training the SVDD model, and constructing a new sample X for the characteristic parameters of the vehicle door motor defined in the step 1)newCalculating the distance d between the new sample and the center of the SVDD hyperspherenew,
4. the method for detecting and evaluating the performance of the door motor of the urban rail vehicle according to claim 3, characterized in that: the performance detection is carried out on the vehicle door driving motor based on the delta R and the time sequence formed by the delta R, and the specific method comprises the following steps:
1) if the delta R exceeds a given threshold value delta, the performance of the driving motor of the door of the urban rail vehicle is over-limit, and the door needs to be overhauled, and faults or other factors are eliminated;
2) if the delta R time sequence has sudden change, the driving motor of the door of the urban rail vehicle is possibly in fault, and the maintenance and troubleshooting are required;
3) if the delta R time sequence generates a trend, the trend indicates that the driving motor of the door of the urban rail vehicle has a fault or the performance is stably reduced, and close attention needs to be paid and fault diagnosis needs to be carried out.
5. The method for detecting and evaluating the performance of the door motor of the urban rail vehicle according to claim 3, characterized in that: the method comprises the following steps of carrying out quantitative health evaluation on the urban rail brake vehicle braking system according to the magnitude of delta R: because a sample point can be formed by opening and closing the vehicle door once, the performance evaluation can be carried out once by opening and closing the vehicle door once; considering that the degradation of the door driving motor is a gradual process, so that the performance evaluation can be performed once in a longer period, such as 1 day, the Δ R is first normalized as follows:
wherein δ is a normalization threshold, α normalization factor is between 0 and 1, and k can be specifically selected according to actual conditions; and then selecting a membership function, and evaluating the performance of the vehicle door driving motor according to the normalized delta R in the step 4).
6. The method for detecting and evaluating the performance of the door motor of the urban rail vehicle according to claim 5, wherein the method comprises the following steps: and the performance evaluation of the vehicle door driving motor adopts a triangular membership function, calculates the normalized delta R, calculates the predefined membership grade of the motor grade, and takes the grade with the maximum membership grade as the performance grade of the vehicle door driving motor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111288095.8A CN114021452A (en) | 2021-11-02 | 2021-11-02 | Urban rail vehicle door motor performance detection and evaluation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111288095.8A CN114021452A (en) | 2021-11-02 | 2021-11-02 | Urban rail vehicle door motor performance detection and evaluation method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114021452A true CN114021452A (en) | 2022-02-08 |
Family
ID=80059752
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111288095.8A Pending CN114021452A (en) | 2021-11-02 | 2021-11-02 | Urban rail vehicle door motor performance detection and evaluation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114021452A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114800036A (en) * | 2022-06-24 | 2022-07-29 | 成都飞机工业(集团)有限责任公司 | Equipment health state assessment method |
-
2021
- 2021-11-02 CN CN202111288095.8A patent/CN114021452A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114800036A (en) * | 2022-06-24 | 2022-07-29 | 成都飞机工业(集团)有限责任公司 | Equipment health state assessment method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111351664B (en) | Bearing temperature prediction and alarm diagnosis method based on LSTM model | |
CN106080655B (en) | A kind of detection method, device and the train of train axle temperature exception | |
CN111240300A (en) | Vehicle health state evaluation model construction method based on big data | |
CN114742157B (en) | Vehicle health state determining method, device, equipment and storage medium | |
CN105976578A (en) | High-speed train axle temperature dynamic alarm threshold setting method based on monitoring data | |
CN110633729A (en) | Driving risk hierarchical clustering method for intelligent networking vehicle group test | |
CN112417766B (en) | Fault diagnosis method based on fault-free data | |
CN112987687B (en) | Cloud-end integrated intelligent line control chassis health monitoring system and method | |
CN111879534B (en) | Performance detection method, system and equipment for urban rail vehicle braking system | |
CN106650157A (en) | Method, device and system for vehicle part fault probability estimation | |
CN109615126A (en) | A kind of bearing residual life prediction technique | |
CN114021452A (en) | Urban rail vehicle door motor performance detection and evaluation method | |
CN108709744A (en) | Motor bearing fault diagnosis method under variable load working condition | |
CN116664015A (en) | Intelligent charging pile management system and method thereof | |
WO2020211846A1 (en) | Safety tree model-based electric vehicle operation and maintenance optimization method | |
CN111652393A (en) | Power equipment abnormity early warning method based on big data technology | |
CN116432397A (en) | Rail transit fault early warning method based on data model | |
CN115545101A (en) | High-speed train bogie fault diagnosis method based on residual error neural network | |
CN115520741A (en) | Elevator operation monitoring and early warning method and system based on neural network and storage medium | |
Jaramillo et al. | Vehicle online monitoring system based on fuzzy classifier | |
Shi et al. | The turnout abnormality diagnosis based on semi-supervised learning method | |
CN114088416A (en) | Fault diagnosis method and device for silencer of hydrodynamic retarder and computer equipment | |
CN113865879A (en) | Guide system for automobile detection | |
CN114757455B (en) | Trench trolley abnormity prediction method and system based on servo motor time sequence data | |
DE102013007007A1 (en) | Pattern and significance detection in databases with genetic algorithms |
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 |