CN114323706B - Train ATO control fault detection method, device, equipment and medium - Google Patents
Train ATO control fault detection method, device, equipment and medium Download PDFInfo
- Publication number
- CN114323706B CN114323706B CN202111383702.9A CN202111383702A CN114323706B CN 114323706 B CN114323706 B CN 114323706B CN 202111383702 A CN202111383702 A CN 202111383702A CN 114323706 B CN114323706 B CN 114323706B
- Authority
- CN
- China
- Prior art keywords
- train
- virtual resistance
- ato
- model
- control
- 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 18
- 238000000034 method Methods 0.000 claims abstract description 73
- 230000008569 process Effects 0.000 claims abstract description 40
- 230000002159 abnormal effect Effects 0.000 claims abstract description 16
- 238000005516 engineering process Methods 0.000 claims abstract description 11
- 238000001914 filtration Methods 0.000 claims abstract description 9
- 230000001133 acceleration Effects 0.000 claims description 40
- 239000011159 matrix material Substances 0.000 claims description 30
- 238000005259 measurement Methods 0.000 claims description 22
- 238000004364 calculation method Methods 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 8
- 230000004044 response Effects 0.000 claims description 8
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 230000007704 transition Effects 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 230000005856 abnormality Effects 0.000 claims description 2
- 230000009471 action Effects 0.000 claims description 2
- 238000012821 model calculation Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000003137 locomotive effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000011056 performance test Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Train Traffic Observation, Control, And Security (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention relates to a train ATO control fault detection method, device, equipment and medium, wherein the method comprises the following steps: s1, establishing an ATO control vehicle internal prediction model; s2, establishing a virtual resistance model during train operation; s3, estimating virtual resistance by using Kalman filtering; s4, counting virtual resistance distribution when no fault exists based on a histogram technology; s5, carrying out moving average on the virtual resistance estimated value during running of the train; and S6, giving an ATO control vehicle early warning when the virtual resistance average value is abnormal. Compared with the prior art, the invention has the advantages of early warning of faults of the traction (braking) unit in the running process of the train from the ATO angle in real time, quantitatively giving out fault grades and the like.
Description
Technical Field
The invention relates to the field of urban rail transit, in particular to a train ATO control vehicle fault detection method, device, equipment and medium for urban rail transit.
Background
Urban rail transit is the backbone of urban public transportation and carries the transport function of passenger traffic of a considerable scale. Especially during peak hours, the availability of the on-board ATO system has a great impact on the overall operating efficiency of the signaling system. According to the on-site operation feedback, various faults are inevitably encountered when ATO is in control, such as restarting of part of traction units in the process of train traction, if abnormal phenomena in the process of train control cannot be timely early-warned and corresponding measures are taken, train emergency braking and other consequences can be caused, and the operation efficiency of a line is seriously affected.
The vehicle-mounted ATP aims to strictly prevent dangers and ensure safety, while the vehicle-mounted ATO meets the requirements of safety requirements, availability and operation efficiency, and needs to run along a safety curve as much as possible, so that the train can reliably run at maximum capacity, and passengers can be safely and efficiently sent to a destination, which is the traffic control principle of the ATO. Therefore, when the vehicle traction or braking unit is partially out of order, unless the ATP applies for emergency braking, the ATO should not stop at the section, but the train is allowed to run to the platform as much as possible for subsequent processing.
Therefore, how to realize that the vehicle-mounted ATO system can detect faults affecting the ATO control of the vehicle in real time, not only can early warn the faults of the control of the vehicle, but also can detect the fault level, provide basis for the adjustment of the follow-up ATO control of the vehicle, and enhance the comfort and usability of the control of the vehicle, so that the vehicle-mounted ATO system is a technical problem to be solved.
Disclosure of Invention
In order to solve the practical application problem, the invention provides a train ATO control fault detection method, device, equipment and medium.
The aim of the invention can be achieved by the following technical scheme:
according to a first aspect of the present invention, there is provided a train ATO control failure detection method comprising the steps of:
s1, establishing an ATO control vehicle internal prediction model;
s2, establishing a virtual resistance model during train operation;
s3, estimating virtual resistance by using Kalman filtering;
s4, counting virtual resistance distribution when no fault exists based on a histogram technology;
s5, carrying out moving average on the virtual resistance estimated value during running of the train;
and S6, giving an ATO control vehicle early warning when the virtual resistance average value is abnormal.
As a preferable technical scheme, the step S1 specifically includes:
the ATO control vehicle internal prediction model is described by a first-order time delay model, and a parameter formula of a continuous system model is shown as (1):
wherein M(s) is a Lawster transformation expression of an ATO output control level mu (T), F(s) is a Lawster transformation expression of an ATO control vehicle internal prediction model acceleration F (T), K is a steady-state gain, T is a system response time, and τ is a system pure delay;
introducing a zero-order retainer to perform Z-transformation on the continuous system time delay model, wherein the control period of the vehicle-mounted ATO is T A An internal prediction model used by the ATO software is obtained as shown in formula (2):
f k ＝alpha*f k-1 +K*(1-alpha)*μ k-n-1 (2)
wherein f k Is a discrete expression of the acceleration f (t) of the predictive model in the ATO control vehicle, mu k-n-1 Is a discrete expression of the output control level μ (t), alpha is a discrete time constant, and the calculation formula is as follows:
alpha＝exp(-T A /T) (3)
n is the integer multiple pure delay of the control period, and the calculation formula is as follows:
n＝floor(τ/T) (4)
wherein the floor () operation means rounding down, and subscripts k, k-1, and k-n-1 represent ATO control moments, omitting the control period T A 。
As an optimal technical scheme, the steady-state gain K is a static parameter and is used for describing steady-state mapping relations between different traction braking levels and train acceleration, and T and tau in a train model are dynamic process parameters.
As a preferable technical scheme, the step S2 specifically includes:
establishing a train dynamics equation, wherein the calculation formula is as follows;
f(t)+g(t)+r(t)＝a(t) (5)
where f (t) is the predicted model acceleration inside the ATO controlled vehicle, g (t) is the equivalent gradient acceleration of the train body range due to gravity, r (t) is the virtual resistance acceleration during the running of the train, and a (t) is the train acceleration.
To simplify the calculation, the virtual resistance value at each moment is considered constant, the speed v (t) and the virtual resistance r (t) are taken as state variables, so that there is the following formula,
wherein the method comprises the steps ofIndicating the acceleration a (t) of the train->Is the rate of change of resistance;
discretizing according to the ATO control period to obtain an equation expressed in the following state space form,
wherein u is k Representing the sum of the predicted model acceleration and the gradient acceleration, T A Is the control period of the vehicle-mounted ATO, v k Representing estimated train speed, r k Representing the estimated virtual resistance.
As a preferable technical solution, the virtual resistance in the step S2 is a balance amount for compensating the difference between the actual response and the model of the train and the influence of the measurement noise.
As a preferable technical scheme, the virtual resistance in the step S2 reflects the ATO control state to a certain extent, and the numerical abnormality of the virtual resistance indicates that an uncontrollable factor exists in the ATO control.
As a preferable technical scheme, the step S3 specifically includes:
firstly, a dynamic process model and a measurement process model are established, the dynamic process model is shown as a formula (8),
x k ＝Ax k-1 +Bu k-1 +w k (8)
wherein,
x k is a state vector of the state of the object,v k representing estimated train speed, r k Representing the estimated virtual resistance force(s),
a is a state transition matrix of which,T A for the control period of the on-board ATO,
b is the control input matrix and,
u k-1 representing the sum of the predicted model acceleration and the slope acceleration at time k-1,
w k is the process noise vector which is the vector of the process noise,ω 1 represents estimated speed noise, ω 2 Representing estimated virtual resistance noise.
The measurement process model is shown in formula (9),
z k ＝Hx k +υ k (9)
wherein,
z k indicating the measured speed of the train,
h represents the measurement matrix, h= [ 10 ],
υ k representing measurement noise.
The estimation process based on the kalman filter algorithm is shown in formula (10),
wherein,
x k the state vector is represented as a function of the state vector,
representing the estimated state vector of the object,
p represents an error covariance matrix of the state vector, the initial value is a two-dimensional identity matrix,
an error covariance matrix representing the estimated state vector,
q represents the process noise covariance matrix, where Q takes on value
R represents the measurement noise variance, where R takes a value of 0.2,
k represents the kalman gain and,
I 2×2 representing a two-dimensional identity matrix,
a ', H' represent the transpose of the state transition matrix a and the measurement matrix H, respectively.
By the calculation of the formula (10), the relatively stable virtual resistance which can reflect the abnormal state of train control is obtained from the calculation result of the dynamic model which contains noise and has modeling uncertainty factors.
As a preferable technical scheme, the step S4 specifically includes:
firstly, counting fault-free running records of a train, obtaining corresponding frequency distribution of a virtual resistance estimated value by using a histogram technology, obtaining a probability density function of the virtual resistance based on a kernel density estimation method, and finally obtaining a virtual resistance threshold value when the train normally runs.
As a preferred embodiment, the moving average in step S5 is calculated as follows:
wherein the method comprises the steps ofRepresenting a sliding average of estimated virtual resistances, m is the sliding window period, r k-j Representing the virtual resistance estimate at time k-j.
As a preferable technical scheme, the step S6 specifically includes:
and judging the virtual resistance estimated value after the moving average, and when the estimated virtual resistance result exceeds the set multiple of the normal threshold value, indicating that the virtual resistance average value is abnormal, wherein the ATO control state does not reach the expected state, and the ATO control has faults.
According to a second aspect of the present invention, there is provided a train ATO control car fault detection device comprising:
the internal prediction model building module is used for building an ATO control vehicle internal prediction model;
the virtual resistance model construction module is used for establishing a virtual resistance model when the train runs;
a virtual resistance estimation module for estimating a virtual resistance using a kalman filter;
the virtual resistance distribution statistics module is used for counting virtual resistance distribution when no fault exists based on a histogram technology;
the running average calculation module is used for carrying out running average on the virtual resistance estimated value when the train runs;
and the ATO car control early warning module is used for giving an ATO car control early warning when the virtual resistance average value is abnormal.
According to a third aspect of the present invention there is provided an electronic device comprising a memory and a processor, the memory having stored thereon a computer program, the processor implementing the method when executing the program.
According to a fourth aspect of the present invention there is provided a computer readable storage medium having stored thereon a computer program which when executed by a processor implements the method.
Compared with the prior art, the invention has the following advantages:
1. the invention can early warn faults of the traction (braking) unit in the running process of the train from the ATO angle in real time, and can quantitatively give out fault grades at the same time;
2. the invention adopts the means of combining offline learning and online operation detection to detect the fault of the ATO control vehicle, and the method can be operated in the vehicle-mounted embedded ATO software rapidly and in real time.
3. In the off-line learning stage, the invention calculates the virtual resistance distribution without faults based on the histogram technology, and obtains a very accurate virtual resistance threshold.
4. The invention can provide the early warning information of the ATO control vehicle through the folding calculation of the virtual resistance for various expected and unexpected faults in the ATO control vehicle process, provides a basis for the subsequent ATO adjustment of the train, and enhances the reliability of the ATO control vehicle.
Drawings
FIG. 1 is a schematic diagram of the ATO control system of the urban rail transit train of the present invention;
FIG. 2 is a flow chart of the ATO control failure detection method of the train of the present invention;
FIG. 3 is a schematic diagram of the train ATO control failure detection device of the present invention;
FIG. 4 is a schematic diagram of the stress during the ATO control operation of the urban rail transit train of the present invention;
FIG. 5 is a schematic diagram of virtual resistance probability distribution based on histogram technique after ATO of the urban rail transit train operates without faults;
fig. 6 is a schematic diagram of the detection result of the train fault during the on-line running of the ATO of the urban rail transit train.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
The train automatic operation function is completed by the cooperation of the vehicle-mounted signal system and the vehicle system, the vehicle-mounted signal system can acquire the real-time condition of the line from the trackside equipment, calculate the train control command of the macroscopic level, and the vehicle system is positioned on the relative microscopic level and is used for coordinating the control of the traction and braking units of each carriage. FIG. 1 is a schematic diagram of the structure of the main equipment of an ATO control system of a train, wherein a vehicle-mounted signal system mainly comprises an ATP, an ATO and a VIOM subsystem, the ATP is responsible for the safety protection function of the train, the ATO is responsible for the automatic driving function of the train, and the VIOM is responsible for the input and output functions of the signal system; the vehicle system mainly comprises a central control unit CCU, a traction unit TCU, a brake unit BCU and mechanical executing components, such as traction motors, air brakes, etc.
Based on the above analysis, various unexpected faults may occur during the ATO control of the train, and possible reasons for such faults are vehicle traction, brake unit faults, vehicle response delay drift, or VIOM output faults, etc. In order to avoid the emergency stop phenomenon caused by various unexpected faults in the train running process, the running state in the ATO control process needs to be monitored, the ATO control early warning is given, a basis is provided for the subsequent ATO control adjustment, and the comfort and usability of the control are enhanced. The ATO control fault detection method for the train is realized through the following technical means, and as shown in fig. 2, the method comprises the following steps:
s1, establishing an ATO control vehicle internal prediction model;
s2, establishing a virtual resistance model during train operation;
s3, estimating virtual resistance by using Kalman filtering;
s4, counting virtual resistance distribution when no fault exists based on a histogram technology;
s5, carrying out moving average on the virtual resistance estimated value during running of the train;
and S6, giving an ATO control vehicle early warning when the virtual resistance average value is abnormal.
According to the invention, a prediction model of a train control command is established, the train acceleration and line information in a vehicle body range are combined, the virtual resistance of the train is calculated based on a Kalman filtering algorithm, the virtual resistance distribution is counted based on a histogram technology, the virtual resistance estimated value is subjected to sliding average when the train runs, and when the virtual resistance average value is detected to be abnormal, an ATO (automatic train control) early warning is given.
1. Establishing an ATO control vehicle internal prediction model
Because the train is a large inertial time delay system, the ATO control command can be responded by the vehicle after a plurality of times, and in order to detect the ATO control fault in real time, an ATO control internal prediction model is necessary to be established. Based on a traction (braking) performance test curve of a train, building related models, including a train traction model, a train electric braking model and a train mechanical braking model, wherein a kinematic model of the train electric braking model can be described by a first-order time delay model, and a parameter formula of a continuous system model is shown as (1):
wherein M(s) is Lawster transformation expression of ATO output control level mu (T), F(s) is Lawster transformation expression of ATO control vehicle internal prediction model acceleration F (T), K is steady-state gain, T is system response time, and τ is system pure delay. The steady-state gain K in the train model is a static parameter used for describing the steady-state mapping relation between different traction braking levels and the acceleration of the train, and T and tau in the train model are dynamic process parameters.
Introducing a zero-order retainer to perform Z-transformation on the continuous system time delay model, wherein the control period of the vehicle-mounted ATO software is T A An internal prediction model used by the ATO software is obtained as shown in formula (2):
f k ＝alpha*f k-1 +K*(1-alpha)*μ k-n-1 (2)
wherein f k Is a discrete expression of the acceleration f (t) of the predictive model in the ATO control vehicle, mu k-n-1 Is a discrete expression of the output control level μ (t), alpha is a discrete time constant, and the calculation formula is as follows:
alpha＝exp(-T A /T) (3)
n is the integer multiple pure delay of the control period, and the calculation formula is as follows:
n＝floor(τ/T) (4)
wherein the meaning of floor () operation is rounding down, where the subscripts k, k-1 and k-n-1 denote ATO control moments, omitting the control period T A 。
2. Establishing a virtual resistance model during train operation
The train running on the track is affected by locomotive traction/braking force, gradient and resistance, as shown in fig. 4, because the train is a delay response system, the traction/braking command output at the current moment needs to be actually acted on the train after a certain time delay, and therefore the unit mass traction currently acted on the train is expressed by adopting the ATO control train internal prediction model acceleration. The train can obtain information such as positioning information, speed information, line gradient and the like through the coded odometer, the transponder and the stored electronic line map, a train dynamics equation is shown in a formula (5),
f(t)+g(t)+r(t)＝a(t) (5)
where f (t) is the predicted model acceleration inside the ATO controlled vehicle, g (t) is the equivalent gradient acceleration of the train body range due to gravity, r (t) is the virtual resistance acceleration during the running of the train, and a (t) is the train acceleration. In general, real drag refers to track adhesion drag, curve drag, and speed-dependent wind drag, and virtual drag is used herein to mean that modeling inaccuracy factors, such as positioning bias, slope determination error, and the like, are included in addition to real drag. To balance equation (5), uncertainty factors such as model errors and the like are uniformly summarized to a variable of virtual resistance. Such as drift in the train response delay parameters, the virtual resistance that is ultimately calculated is abnormal.
To simplify the calculation, the virtual resistance value at each moment is considered to be constant, the speed v (t) and the virtual resistance r (t) are taken as state variables, so there is the following formula (6),
wherein the method comprises the steps ofIndicating the acceleration a (t) of the train->Is the rate of change of resistance.
Discretizing according to the ATO control period to obtain an equation (7) expressed in the following state space form,
wherein u is k Representing the sum of the predicted model acceleration and the gradient acceleration, T A Is the control period of the vehicle-mounted ATO, v k Representing estimated train speed, r k Representing the estimated virtual resistance.
3. Estimating virtual resistance using Kalman filtering (Kalman)
In order to overcome the influence of modeling inaccuracy factors and measurement noise on virtual resistance estimation, a Kalman filtering algorithm is adopted. The Kalman filtering algorithm is an optimal autoregressive estimation algorithm, and although process noise and measurement noise exist, based on priori knowledge of the whole dynamic process, the accurate virtual resistance can be estimated through proper configuration of noise parameters, so that traction/braking unit faults or model parameter drift existing in the process of controlling the vehicle can be accurately reflected. In order to estimate the reference velocity and the reference acceleration using the kalman filter algorithm, a dynamic process model and a measurement process model are first established, the dynamic process model is shown as formula (8),
x k ＝Ax k-1 +Bu k-1 +w k (8)
wherein,
x k is a state vector of the state of the object,v k representing estimated train speed, r k Representing the estimated virtual resistance force(s),
a is a state transition matrix of which,
b is the control input matrix and,
u k-1 representing the sum of the predicted model acceleration and the slope acceleration at time k-1,
w k is the process noise vector which is the vector of the process noise,ω 1 represents estimated speed noise, ω 2 Representing estimated virtual resistance noise.
The measurement process model is shown in formula (9),
z k ＝Hx k +υ k (9)
wherein,
z k indicating the measured speed of the train,
h represents the measurement matrix, h= [ 10 ],
υ k representing measurement noise.
The estimation process based on the kalman filter algorithm is shown in formula (10),
wherein,
x k the state vector is represented as a function of the state vector,
representing the estimated state vector of the object,
p represents an error covariance matrix of the state vector, the initial value is a two-dimensional identity matrix,
an error covariance matrix representing the estimated state vector,
q represents the process noise covariance matrix, where Q takes on value
R represents the measurement noise variance, where R takes a value of 0.2,
k represents the kalman gain and,
I 2×2 representing a two-dimensional identity matrix,
a ', H' represent the transpose of the state transition matrix a and the measurement matrix H, respectively.
By the calculation of the formula (10), the relatively stable virtual resistance which can reflect the abnormal state of train control can be obtained from the calculation result of the dynamic model which contains noise and has modeling uncertainty factors.
4. Histogram-based technique for statistics of virtual resistance distribution without faults
In order to determine the train control state by applying the virtual resistance estimation value in real time, a virtual resistance probability distribution model of the train during fault-free operation needs to be counted and learned first. As shown in FIG. 5, a large number of fault-free running records of the train are counted, corresponding frequency distribution is obtained for the estimated value of the virtual resistance by using a histogram technology, then a probability density function of the virtual resistance is obtained based on a kernel density estimation method, and finally a virtual resistance threshold value in normal running of the train is obtained. It should be clear that the estimated virtual resistance actually reflects the difference between the expected state of the control car and the actual state of the train, and is influenced by factors such as uncertainty in modeling, and the like, and the value of the estimated virtual resistance is positive or negative, and generally takes a negative value as a majority, and also accords with the actual situation.
5. ATO control fault detection for train on-line running
ATO control vehicle fault detection is carried out based on the virtual resistance threshold statistical result obtained in the learning stage, in order to obtain relatively reliable and stable control vehicle state estimation, false alarm phenomenon of ATO control vehicle fault detection caused by uncertain factors such as inaccurate modeling, instantaneous interference and the like is eliminated, and when a train runs, a real-time estimated virtual resistance value is subjected to sliding average, as shown in a formula (11).
Wherein the method comprises the steps ofRepresenting a sliding average of estimated virtual resistances, m is the sliding window period, r k-j Representing the virtual resistance estimate at time k-j.
And judging the virtual resistance estimated value after the moving average, and when the estimated virtual resistance result exceeds the set multiple of the normal threshold value, indicating that the virtual resistance average value is abnormal, wherein the ATO control state does not reach the expected state, and the ATO control has faults. As shown in fig. 6, during the running process of the train in the section, the traction unit of one half of the train is reset and restarted in a certain time period due to unknown reasons, and the traction force of the whole train is lost by one half, so that the actual acceleration of the train is only half of the expected value, and at the moment, the train control early warning can be given according to the detected ATO train control fault signal, so that the basis is provided for the subsequent ATO train control adjustment.
The above description of the method embodiments further describes the solution of the present invention by means of device embodiments.
As shown in fig. 3, the train ATO control fault detection device of the present invention comprises:
the internal prediction model construction module 100 is used for building an ATO control vehicle internal prediction model;
the virtual resistance model construction module 200 is used for establishing a virtual resistance model during train operation;
a virtual resistance estimation module 300 for estimating a virtual resistance using kalman filtering;
a virtual resistance distribution statistics module 400 for counting virtual resistance distribution without faults based on histogram technique;
the moving average calculating module 500 is configured to perform moving average on the estimated virtual resistance value during running of the train;
the ATO control vehicle early warning module 600 is configured to give an ATO control vehicle early warning when the virtual resistance average value is abnormal.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the described modules may refer to corresponding procedures in the foregoing method embodiments, which are not described herein again.
The electronic device of the present invention includes a Central Processing Unit (CPU) that can perform various appropriate actions and processes according to computer program instructions stored in a Read Only Memory (ROM) or computer program instructions loaded from a storage unit into a Random Access Memory (RAM). In the RAM, various programs and data required for the operation of the device can also be stored. The CPU, ROM and RAM are connected to each other by a bus. An input/output (I/O) interface is also connected to the bus.
A plurality of components in a device are connected to an I/O interface, comprising: an input unit such as a keyboard, a mouse, etc.; an output unit such as various types of displays, speakers, and the like; a storage unit such as a magnetic disk, an optical disk, or the like; and communication units such as network cards, modems, wireless communication transceivers, and the like. The communication unit allows the device to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processing unit performs the respective methods and processes described above, for example, the methods S1 to S6. For example, in some embodiments, methods S1-S6 may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device via the ROM and/or the communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of the methods S1 to S6 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to perform methods S1-S6 in any other suitable manner (e.g., by means of firmware).
The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a load programmable logic device (CPLD), etc.
Program code for carrying out methods of the present invention may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (8)
1. The train ATO control fault detection method is characterized by comprising the following steps of:
s1, establishing an ATO control vehicle internal prediction model;
s2, establishing a virtual resistance model during train operation;
s3, estimating virtual resistance by using Kalman filtering;
s4, counting virtual resistance distribution when no fault exists based on a histogram technology;
s5, carrying out moving average on the virtual resistance estimated value during running of the train;
s6, giving an ATO control vehicle early warning when the virtual resistance average value is abnormal;
the step S1 specifically comprises the following steps:
the ATO control vehicle internal prediction model is described by a first-order time delay model, and a parameter formula of a continuous system model is shown as (1):
wherein, m(s) is Law's transformation expression of ATO output control level mu (T), F(s) is Law's transformation expression of ATO control vehicle internal prediction model acceleration F (T), K is steady-state gain, T is system response time, and τ is system pure delay;
introducing a zero-order retainer to perform Z-transformation on the continuous system time delay model, wherein the vehicle-mounted ATO control period is T A An internal prediction model used by the ATO software is obtained as shown in formula (2):
f k ＝alpha*f k-1 +K*(1-alpha)*μ k-n-1 (2)
wherein f k Is a discrete expression of the acceleration f (t) of the predictive model in the ATO control vehicle, mu k-n-1 Is a discrete expression of the output control level μ (t), alpha is a discrete time constant, and the calculation formula is as follows:
alpha＝exp(-T A /T) (3)
n is the integer multiple pure delay of the control period, and the calculation formula is as follows:
n＝floor(τ/T) (4)
wherein the floor () operation means rounding down, and subscripts k, k-1, and k-n-1 represent ATO control moments, omitting the control period T A ；
The step S2 specifically comprises the following steps:
establishing a train dynamics equation, wherein the calculation formula is as follows;
f(t)+g(t)+r(t)＝a(t) (5)
wherein f (t) is the predicted model acceleration in the ATO control vehicle, g (t) is the equivalent gradient acceleration of the train body range caused by gravity action, r (t) is the virtual resistance acceleration in the running process of the train, and a (t) is the train acceleration;
to simplify the calculation, the virtual resistance value at each moment is considered constant, the speed v (t) and the virtual resistance r (t) are taken as state variables, so that there is the following formula,
wherein the method comprises the steps ofThe train acceleration a (t) is indicated,/>is the rate of change of resistance;
discretizing according to the ATO control period to obtain an equation expressed in the following state space form,
wherein u is k Representing the sum of the predicted model acceleration and the gradient acceleration, T A Is the control period of the vehicle-mounted ATO, v k Representing estimated train speed, r k Representing the estimated virtual resistance;
the step S3 specifically comprises the following steps:
firstly, a dynamic process model and a measurement process model are established, the dynamic process model is shown as a formula (8),
x k ＝Ax k-1 +Bu k-1 +w k (8)
wherein,
x k is a state vector of the state of the object,v k representing estimated train speed, r k Representing the estimated virtual resistance force(s),
a is a state transition matrix of which,T A for the control period of the on-board ATO,
b is the control input matrix and,
u k-1 representing the sum of the predicted model acceleration and the slope acceleration at time k-1,
w k is the process noise vector which is the vector of the process noise,ω 1 represents estimated speed noise, ω 2 Representing estimated virtual resistance noise;
the measurement process model is shown in formula (9),
z k ＝Hx k +υ k (9)
wherein,
z k indicating the measured speed of the train,
h represents the measurement matrix, h= [ 10 ],
υ k representing measurement noise;
the estimation process based on the kalman filter algorithm is shown in formula (10),
wherein,
x k the state vector is represented as a function of the state vector,
representing the estimated state vector of the object,
p represents an error covariance matrix of the state vector, the initial value is a two-dimensional identity matrix,
an error covariance matrix representing the estimated state vector,
q represents the process noise covariance matrix, where Q takes on value
R represents the measurement noise variance, where R takes a value of 0.2,
k represents the kalman gain and,
I 2×2 representing a two-dimensional identity matrix,
a ', H' represent transpose matrices of the state transition matrix a and the measurement matrix H, respectively;
obtaining relatively stable virtual resistance capable of reflecting abnormal train control states from a dynamic model calculation result containing noise and having modeling uncertainty factors through calculation of a formula (10);
the step S4 specifically comprises the following steps:
firstly, counting fault-free running records of a train, obtaining corresponding frequency distribution of a virtual resistance estimated value by using a histogram technology, obtaining a probability density function of the virtual resistance based on a kernel density estimation method, and finally obtaining a virtual resistance threshold value when the train normally runs;
the moving average in the step S5 is calculated as follows:
wherein the method comprises the steps ofRepresenting a sliding average of estimated virtual resistances, m is the sliding window period, r k-j Representing the virtual resistance estimate at time k-j.
2. The method for detecting the fault of the ATO control of the train according to claim 1, wherein the steady-state gain K is a static parameter for describing a steady-state mapping relationship between different traction braking levels and acceleration of the train, and T and τ in a train model are dynamic process parameters.
3. The method for detecting the fault of the ATO control of the train according to claim 1, wherein the virtual resistance in the step S2 is a balance quantity for compensating the difference between the actual response and the model of the train and the influence of the measurement noise.
4. The method for detecting the fault of the ATO control of the train according to claim 1, wherein the virtual resistance in the step S2 reflects the ATO control state to a certain extent, and the numerical abnormality indicates that the ATO control has an uncontrollable factor.
5. The method for detecting the fault of the ATO control of the train according to claim 1, wherein the step S6 is specifically:
and judging the virtual resistance estimated value after the moving average, and when the estimated virtual resistance result exceeds the set multiple of the normal threshold value, indicating that the virtual resistance average value is abnormal, wherein the ATO control state does not reach the expected state, and the ATO control has faults.
6. An apparatus for use in the train ATO control car failure detection method of claim 1, characterized in that the apparatus comprises:
the internal prediction model building module is used for building an ATO control vehicle internal prediction model;
the virtual resistance model construction module is used for establishing a virtual resistance model when the train runs;
a virtual resistance estimation module for estimating a virtual resistance using a kalman filter;
the virtual resistance distribution statistics module is used for counting virtual resistance distribution when no fault exists based on a histogram technology;
the running average calculation module is used for carrying out running average on the virtual resistance estimated value when the train runs;
and the ATO car control early warning module is used for giving an ATO car control early warning when the virtual resistance average value is abnormal.
7. An electronic device comprising a memory and a processor, the memory having stored thereon a computer program, characterized in that the processor, when executing the program, implements the method according to any of claims 1-5.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111383702.9A CN114323706B (en) | 2021-11-22 | 2021-11-22 | Train ATO control fault detection method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111383702.9A CN114323706B (en) | 2021-11-22 | 2021-11-22 | Train ATO control fault detection method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114323706A CN114323706A (en) | 2022-04-12 |
| CN114323706B true CN114323706B (en) | 2024-04-12 |
Family
ID=81045860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111383702.9A Active CN114323706B (en) | 2021-11-22 | 2021-11-22 | Train ATO control fault detection method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114323706B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004080838A (en) * | 2002-08-09 | 2004-03-11 | Toshiba Corp | Automatic train operating apparatus |
| CN102981408A (en) * | 2012-12-10 | 2013-03-20 | 华东交通大学 | Running process modeling and adaptive control method for motor train unit |
| CN108932356A (en) * | 2017-05-27 | 2018-12-04 | 南京理工大学 | A kind of ATO speed command energy conservation optimizing method considering Train delay factor |
| JP2019018755A (en) * | 2017-07-19 | 2019-02-07 | 株式会社東芝 | Abnormality detection device, abnormality detection method, and computer program |
| CN111102978A (en) * | 2019-12-05 | 2020-05-05 | 深兰科技（上海）有限公司 | Method and device for determining vehicle motion state and electronic equipment |
| CN112550359A (en) * | 2020-11-22 | 2021-03-26 | 卡斯柯信号有限公司 | Train smooth tracking control method based on stepped target speed curve |
| CN113341171A (en) * | 2021-06-01 | 2021-09-03 | 北京全路通信信号研究设计院集团有限公司 | Train speed measurement noise reduction filtering method and device with low delay characteristic |
| CN113479218A (en) * | 2021-08-09 | 2021-10-08 | 哈尔滨工业大学 | Roadbed automatic driving auxiliary detection system and control method thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101256315B1 (en) * | 2011-10-18 | 2013-04-18 | 엘에스산전 주식회사 | Apparatus and method for controlling train speed |
| KR101583878B1 (en) * | 2013-11-15 | 2016-01-08 | 엘에스산전 주식회사 | Apparatus for controlling speed in railway vehicles |
-
2021
- 2021-11-22 CN CN202111383702.9A patent/CN114323706B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004080838A (en) * | 2002-08-09 | 2004-03-11 | Toshiba Corp | Automatic train operating apparatus |
| CN102981408A (en) * | 2012-12-10 | 2013-03-20 | 华东交通大学 | Running process modeling and adaptive control method for motor train unit |
| CN108932356A (en) * | 2017-05-27 | 2018-12-04 | 南京理工大学 | A kind of ATO speed command energy conservation optimizing method considering Train delay factor |
| JP2019018755A (en) * | 2017-07-19 | 2019-02-07 | 株式会社東芝 | Abnormality detection device, abnormality detection method, and computer program |
| CN111102978A (en) * | 2019-12-05 | 2020-05-05 | 深兰科技（上海）有限公司 | Method and device for determining vehicle motion state and electronic equipment |
| CN112550359A (en) * | 2020-11-22 | 2021-03-26 | 卡斯柯信号有限公司 | Train smooth tracking control method based on stepped target speed curve |
| CN113341171A (en) * | 2021-06-01 | 2021-09-03 | 北京全路通信信号研究设计院集团有限公司 | Train speed measurement noise reduction filtering method and device with low delay characteristic |
| CN113479218A (en) * | 2021-08-09 | 2021-10-08 | 哈尔滨工业大学 | Roadbed automatic driving auxiliary detection system and control method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 动车组自动驾驶制动过程双自适应优化控制;李中奇；杨振村;《计算机仿真》;20160615;第33卷(第06期);第121-126页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114323706A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109689470B (en) | Abnormality diagnosis device, abnormality diagnosis method, and computer program | |
| US9517780B2 (en) | Apparatus for controlling speed in railway vehicles | |
| US20130096741A1 (en) | Apparatus and method for controlling train speed | |
| US9150209B2 (en) | System and method for monitoring braking effort | |
| WO2023071442A1 (en) | Data processing method and apparatus | |
| CN113247047B (en) | Method and device for protecting consistency of train speed sensor | |
| CN112298189A (en) | Method for training a neural network, method and device for estimating a friction coefficient, and machine-readable storage medium | |
| KR20200058132A (en) | Railway vehicle major component and system diagnosis apparatus | |
| Cunillera et al. | Real-time train motion parameter estimation using an unscented Kalman filter | |
| CN115447554A (en) | Method and device for detecting braking performance of unmanned mining vehicle, electronic equipment and storage medium | |
| JP6816379B2 (en) | Collision avoidance system, information processing device, collision avoidance program and information processing program | |
| CN117584996A (en) | New energy automobile control method | |
| CN114323706B (en) | Train ATO control fault detection method, device, equipment and medium | |
| CN109883598B (en) | Method for detecting traction force of rail locomotive | |
| CN116558469A (en) | Distance error evaluation method, device, electronic equipment and storage medium | |
| CN112158237B (en) | Deep fusion system integrating TCMS and ATO functions and train | |
| CN116186877A (en) | Method and device for detecting characteristics of a vehicle | |
| US11987276B2 (en) | Monitoring system for axles of a vehicle | |
| KR102261155B1 (en) | Method and apparatus for controlling a vehicle using two virtual sensors | |
| CN107219774A (en) | A kind of ATO Train Parameters based on captured at jobsite determine method | |
| CN113415319B (en) | State detection method and system of speed sensor | |
| WO2023074319A1 (en) | Brake device health determination device and brake device health determination method | |
| KR20180065628A (en) | Apparatus for determining failings of railroad train, and determining method for failings of railroad train | |
| CN114670896B (en) | Train speed sharing system and method | |
| CN116506309B (en) | Vehicle-mounted ATP communication signal comprehensive monitoring system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |