CN112348078A - Gate machine controller with sub-health pre-diagnosis and fault type clustering functions - Google Patents
Gate machine controller with sub-health pre-diagnosis and fault type clustering functions Download PDFInfo
- Publication number
- CN112348078A CN112348078A CN202011222663.XA CN202011222663A CN112348078A CN 112348078 A CN112348078 A CN 112348078A CN 202011222663 A CN202011222663 A CN 202011222663A CN 112348078 A CN112348078 A CN 112348078A
- Authority
- CN
- China
- Prior art keywords
- fault
- sub
- running
- diagnosis
- state
- 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
- 230000006870 function Effects 0.000 title claims abstract description 26
- 238000003745 diagnosis Methods 0.000 title claims description 46
- 239000013598 vector Substances 0.000 claims abstract description 112
- 238000000034 method Methods 0.000 claims abstract description 51
- 230000002490 cerebral effect Effects 0.000 claims description 21
- 238000013528 artificial neural network Methods 0.000 claims description 20
- 238000006073 displacement reaction Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 10
- 238000013507 mapping Methods 0.000 claims description 10
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 210000004258 portal system Anatomy 0.000 claims 6
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000006243 chemical reaction Methods 0.000 description 6
- 230000001133 acceleration Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- VIEYMVWPECAOCY-UHFFFAOYSA-N 7-amino-4-(chloromethyl)chromen-2-one Chemical compound ClCC1=CC(=O)OC2=CC(N)=CC=C21 VIEYMVWPECAOCY-UHFFFAOYSA-N 0.000 description 2
- 101100290429 Mus musculus Mcm3 gene Proteins 0.000 description 2
- 230000003213 activating effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000001174 ascending effect Effects 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P6/00—Arrangements for controlling synchronous motors or other dynamo-electric motors using electronic commutation dependent on the rotor position; Electronic commutators therefor
- H02P6/08—Arrangements for controlling the speed or torque of a single motor
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P6/00—Arrangements for controlling synchronous motors or other dynamo-electric motors using electronic commutation dependent on the rotor position; Electronic commutators therefor
- H02P6/14—Electronic commutators
- H02P6/16—Circuit arrangements for detecting position
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Power Engineering (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a method for judging the sub-health state of a door machine system of a railway vehicle and clustering fault types and a corresponding door machine controller. The characteristic vectors corresponding to the normal running state and the main typical fault state are both stored in the gantry crane controller, and whether the gantry crane system is in a sub-health state can be judged by calculating the Minkowski distance between the current running state characteristic vector and the normal running state Terry vector. Further, fault type clustering can be realized by calculating Min distance between the current running characteristic vector and each typical state characteristic vector, and the invention also provides a software and hardware structure of the gate machine controller with the functions. The invention has the beneficial effects that: the characteristic vectors of the door opening and closing curves automatically extracted by the door controller are used for sub-health judgment and fault type clustering analysis, and the running safety of the rail vehicle can be improved.
Description
Technical Field
The invention relates to the technical field of rail vehicle door fault detection, in particular to a door machine controller with sub-health pre-diagnosis and fault type clustering functions.
Background
The rail vehicle door works in a complex and severe real environment, the safety of the rail vehicle door is the most important part of the overall safety of the vehicle, and the rail vehicle door is directly related to public safety. At present, a common rail vehicle door motor controller does not have the functions of online real-time sub-health pre-diagnosis and fault type clustering of a rail vehicle door motor system. At present with developed internet of things, online real-time evaluation of the health condition of a door machine system is an important means for improving the safety of a rail vehicle door system, a rail vehicle door machine controller comprises a door machine part and a door controller, the door machine runs under the control and the drive of the door controller, and the sub-health pre-diagnosis and fault type clustering functions of the door machine system can be embodied by characteristic vectors of door opening and closing curves automatically extracted by the door controller.
Disclosure of Invention
The present invention is directed to a gate machine controller with sub-health pre-diagnosis and fault type clustering functions, so as to solve the problems in the background art.
In order to achieve the purpose, the invention provides the following technical scheme:
a method with sub-health pre-diagnosis and fault type clustering functions is characterized in that: utilizing a cerebellar neural network to realize sub-health pre-diagnosis and fault type clustering, segmenting the switching process of a gate body controlled by a gate machine system to form an input sample of the cerebellar neural network, wherein the cerebellar neural network comprises a virtual memory and a stray code, taking the stray code value as a feature vector, and realizing the sub-health pre-diagnosis and the fault type clustering according to the Minkowski distance of each feature vector in a vector space, and the process comprises the following steps:
the method comprises the following steps: dividing the operation parameters of the rail vehicle door in the current door opening and closing process into 4 sections: the method comprises the following steps of (1) extracting time and displacement in the 4 sections of processes of an ascending section, a constant speed section, a descending section and a pressure velocity section to form input samples of the cerebellar neural network, wherein the input samples comprise: the method comprises the steps of rail vehicle door opening and closing acceleration time, rail vehicle door opening and closing acceleration displacement, rail vehicle door opening and closing constant speed time, rail vehicle door opening and closing constant speed displacement, rail vehicle door opening and closing deceleration time, rail vehicle door opening and closing deceleration displacement, rail vehicle door opening and closing speed pressing time and rail vehicle door opening and closing speed pressing displacement;
step two: obtaining a characteristic vector of the opening and closing state of the rail vehicle door after conceptual mapping and stray transformation of a cerebellar neural network according to the input sample obtained in the step one, wherein the specific process is as follows:
the input sample S after the door is opened and closed(k)Inputting the data into a door machine controller, activating C storage units in a concept memory AC through concept mapping, finding weights of an actual memory for the activated C storage units through stray coding, reading the weights, and outputting a vector WiVector WiThe characteristic vector of the door opening and closing state at each time is obtained;
step three: calculating the Minkowski distance between the current running feature vector and the typical feature vector obtained in advance, and performing sub-health pre-diagnosis and fault type clustering:
s1: calculating the Minkowski distance in the vector space between the running eigenvector for the no-fault condition and the running eigenvector for the typical fault condition:
taking the minimum of the above minkowski distance:
l0＝min{D(W1，W2)，D(W1，W3)，…，D(W1，W9)}；
s2: calculating the Minkowski distance between the running characteristic vector of the rail vehicle door in the door opening and closing state and the running characteristic vector of the rail vehicle door in the fault-free state in a vector space:
s3: the shortest minkowski distance l between the operating eigenvector for the no fault condition obtained by comparing S1 and the operating eigenvector for the typical fault condition in the vector space0And the Minkowski distance l of the running eigenvector of the rail vehicle door at the on-off state and the running eigenvector of the fault-free state obtained at S2 is used for performing sub-health pre-diagnosis on the on-off state:
when in use
Meanwhile, the running state of the gantry crane system is healthy;
when in use
When the door machine system is in a sub-healthy running state, the door machine controller gives an alarm and switches to a normal control state;
when in use
When the system of the gantry crane is in failure in operation state, fault type clustering is carried out, and a gantry crane controller gives an alarm and transfers to an emergency processing program;
when the rail vehicle door is in the door opening and closing fault at this time, the door machine controller carries out fault type clustering, and clustering conditions are as follows:
Output feature vector W of the operating stateiAnd WjThe method comprises the following steps: faultless output vector W1Output vector W of abnormal V-type fault2Output vector W of motor assembly loosening fault3And output vector W for small fault of centering size change4Output vector W for fault with large change of centering size5Output vector W of outward-moving fault of upper sliding way6Output vector W of lateral interference fault of lower stop pin7Output vector W of longitudinal interference fault of lower stop pin8Output vector W of pressing wheel overvoltage fault9。
And C of the cerebellar neural network is the generalization width and is taken as 6.
The spur encoding is a compression mapping, i.e. the space of hash values is much smaller than the space of inputs, different inputs may hash to the same output. The method is simply a function for compressing a message with an arbitrary length into a message digest with a fixed length, and here, the hash coding function adopts a division-leave remainder method. Taking the remainder of the key divided by a number p which is not more than the length m of the hash table as the hash address. I.e., H (key) key MOD p, p ≦ m. Where the divisor p is referred to as the modulus. The key to this approach is the choice of the modulo p. The probability that each keyword in the data element set is mapped to any address of the memory unit through the hash function is equal, and therefore the probability of hash collision is reduced as far as possible. One experience with the remainder of division is that if the hash table is m long, p is typically the smallest prime number less than or equal to the table length (preferably close to m) or a composite number that does not contain less than 20 prime factors.
The representative feature vector includes: the method comprises the following steps of storing characteristic vectors of 9 typical running states in a door machine controller of the railway vehicle in advance, wherein the characteristic vectors of the running states of a no-fault state, a running characteristic vector of a V-shaped abnormal fault state, a running characteristic vector of a motor assembly loosening fault state, a running characteristic vector of a fault state with small change in centering size, a running characteristic vector of a fault state with large change in centering size, a running characteristic vector of a fault state with outward movement of an upper slideway, a running characteristic vector of a fault state with transverse interference of a lower stop pin, a running characteristic vector of a fault state with longitudinal interference of a lower stop pin and a running characteristic vector of a fault state with overpressure of a pinch roller.
The minkowski distance calculation formula is:
The invention provides a control method with sub-health pre-diagnosis and fault type clustering functions and a gantry crane controller. A control method with sub-health pre-diagnosis and fault type clustering functions comprises the following steps:
the method comprises the following steps: h for rotor position sensor output in control system with sub-health pre-diagnosis and fault type clustering functionsa、Hb、HcDetecting to form a speed feedback signal omega*；
Step two: output current i of control system with sub-health pre-diagnosis and fault type clustering functionsA、iBDetecting and comparing with absoluteTorque direction signal DIR output by value circuit and H output by rotor position sensora、Hb、HcSynthesizing a current feedback signal i together by a current feedback signalf；
Step three: subtracting the aforementioned speed feedback signal omega from the speed given signal omega*Obtaining a speed error, and obtaining a signal u through proportional-integral regulation operation of integral saturation resistance on the speed errorgi；
Step four: signal ugiOutputting a torque control signal MCMD and the current feedback signal i through an absolute value circuitfComparing to obtain a current error, and outputting PWM (pulse width modulation) by proportional integral operation of integral saturation resistance of the current error;
step five: PWM of output and H of output of the aforementioned rotor position sensora、Hb、HcAnd the three-phase inverter is controlled together through phase change logic to drive the brushless direct current motor.
A door machine controller with sub-health pre-diagnosis and fault type clustering functions adopts a brushless direct current motor as a driving motor of a door controller, adopts a power MOS three-phase bridge circuit to feed the brushless direct current motor, adopts an STM32F429 chip as a built-in controller, and realizes the control of the position, the speed and the current of the brushless direct current motor. And the current is detected by two LEM Hall current sensors, the current value of the third phase can be calculated according to the kirchhoff current rule, and the sampling and conversion of the motor circuit signals are completed by utilizing a high-speed A/D channel of STM32F 429. The control software of the door machine controller further comprises a position control module, a speed control module, a D-axis current control module, a Q-axis current control module, a rotation conversion module, an inverse rotation conversion module, a speed position calculation software module, a speed position storage module, a torque calculation software module, a torque storage module, a CMAC sample input module, a stray code mapping module, a characteristic vector extraction module and a sub-health pre-diagnosis and alarm module, the automatic door controller further comprises a hardware module, the hardware module adopts a Hall current detector to form a two-phase current detection module, and adopts an intelligent power module IPM to form an inverter to drive a motor.
Compared with the prior art, the invention has the beneficial effects that:
1. the method comprises the steps of detecting and recording running parameters (time and displacement) of a rail vehicle door on-off state, inputting the running parameters (time and displacement) serving as input samples into a cerebellar neural network to obtain a running characteristic vector of the door on-off state at this time, calculating Minkowski distances of the running characteristic vector of the door on-off state and 9 typical running characteristic vectors stored in a rail vehicle door controller in advance in a vector space, and performing sub-health pre-diagnosis and fault type clustering of the door on-off state at this time according to a mathematical relationship among the Minkowski distances;
2. the automatic door control system can be applied to automatic door control systems of various important occasions and automatic door systems of railway vehicles, and can effectively improve the safety and reliability of door control, thereby improving the safety of the whole train operation.
Drawings
Fig. 1 is a software and hardware structure diagram of a door machine controller in the embodiment of the present invention.
Fig. 2 is a control schematic block diagram of a door machine controller in the embodiment of the present invention.
Fig. 3 is a schematic diagram of a gate controller with an automatic gate opening and closing curve feature vector extraction function according to an embodiment of the present invention.
Fig. 4 is a time velocity curve of the closing time of the rail vehicle according to the embodiment of the present invention.
Fig. 5 is a schematic diagram of a cerebellar neural network software system algorithm according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention are described below in detail with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without any inventive step, are within the scope of the present invention.
Referring to fig. 1-5, the present invention provides a technical solution:
a method with sub-health pre-diagnosis and fault type clustering functions is characterized in that: utilizing a cerebellar neural network to realize sub-health pre-diagnosis and fault type clustering, segmenting the switching process of a gate body controlled by a gate machine system to form an input sample of the cerebellar neural network, wherein the cerebellar neural network comprises a virtual memory and a stray code, taking the stray code value as a feature vector, and realizing the sub-health pre-diagnosis and the fault type clustering according to the Minkowski distance of each feature vector in a vector space, and the process comprises the following steps:
the method comprises the following steps: dividing the operation parameters of the rail vehicle door in the current door opening and closing process into 4 sections: the method comprises the following steps of (1) extracting time and displacement in the 4 sections of processes of an ascending section, a constant speed section, a descending section and a pressure velocity section to form input samples of the cerebellar neural network, wherein the input samples comprise: the method comprises the steps of rail vehicle door opening and closing acceleration time, rail vehicle door opening and closing acceleration displacement, rail vehicle door opening and closing constant speed time, rail vehicle door opening and closing constant speed displacement, rail vehicle door opening and closing deceleration time, rail vehicle door opening and closing deceleration displacement, rail vehicle door opening and closing speed pressing time and rail vehicle door opening and closing speed pressing displacement;
step two: obtaining a characteristic vector of the opening and closing state of the rail vehicle door after conceptual mapping and stray transformation of a cerebellar neural network according to the input sample obtained in the step one, wherein the specific process is as follows:
the input sample S after the door is opened and closed(k)Inputting the data into a door machine controller, activating C storage units in a concept memory AC through concept mapping, finding weights of an actual memory for the activated C storage units through stray coding, reading the weights, and outputting a vector WiVector WiThe characteristic vector of the door opening and closing state at each time is obtained;
step three: calculating the Minkowski distance between the current running feature vector and the typical feature vector obtained in advance, and performing sub-health pre-diagnosis and fault type clustering:
s1: calculating the Minkowski distance in the vector space between the running eigenvector for the no-fault condition and the running eigenvector for the typical fault condition:
taking the minimum of the above minkowski distance:
l0＝min{D(W1，W2)，D(W1，W3)，…，D(W1，W9)}；
s2: calculating the Minkowski distance between the running characteristic vector of the rail vehicle door in the door opening and closing state and the running characteristic vector of the rail vehicle door in the fault-free state in a vector space:
s3: the shortest minkowski distance l between the operating eigenvector for the no fault condition obtained by comparing S1 and the operating eigenvector for the typical fault condition in the vector space0And the Minkowski distance l of the running eigenvector of the rail vehicle door at the on-off state and the running eigenvector of the fault-free state obtained at S2 is used for performing sub-health pre-diagnosis on the on-off state:
when in use
Meanwhile, the running state of the gantry crane system is healthy;
when in use
When the door machine system is in a sub-healthy running state, the door machine controller gives an alarm and switches to a normal control state;
when in use
When the system of the gantry crane is in failure in operation state, fault type clustering is carried out, and a gantry crane controller gives an alarm and transfers to an emergency processing program;
when the rail vehicle door is in the door opening and closing fault at this time, the door machine controller carries out fault type clustering, and clustering conditions are as follows:
Output feature vector W of the operating stateiAnd WiThe method comprises the following steps: faultless output vector W1Output vector W of abnormal V-type fault2Output vector W of motor assembly loosening fault3Centering dimensional changeMinimizing the output vector W of a fault4Output vector W for fault with large change of centering size5Output vector W of outward-moving fault of upper sliding way6Output vector W of lateral interference fault of lower stop pin7Output vector W of longitudinal interference fault of lower stop pin8Output vector W of pressing wheel overvoltage fault9。
And C of the cerebellar neural network is the generalization width and is taken as 6.
The spur encoding is a compression mapping, i.e. the space of hash values is much smaller than the space of inputs, different inputs may hash to the same output. The method is simply a function for compressing a message with an arbitrary length into a message digest with a fixed length, and here, the hash coding function adopts a division-leave remainder method. Taking the remainder of the key divided by a number p which is not more than the length m of the hash table as the hash address. I.e., H (key) key MOD p, p ≦ m. Where the divisor p is referred to as the modulus. The key to this approach is the choice of the modulo p. The probability that each keyword in the data element set is mapped to any address of the memory unit through the hash function is equal, and therefore the probability of hash collision is reduced as far as possible. One experience with the remainder of division is that if the hash table is m long, p is typically the smallest prime number less than or equal to the table length (preferably close to m) or a composite number that does not contain less than 20 prime factors.
The representative feature vector includes: the method comprises the following steps of storing characteristic vectors of 9 typical running states in a door machine controller of the railway vehicle in advance, wherein the characteristic vectors of the running states of a no-fault state, a running characteristic vector of a V-shaped abnormal fault state, a running characteristic vector of a motor assembly loosening fault state, a running characteristic vector of a fault state with small change in centering size, a running characteristic vector of a fault state with large change in centering size, a running characteristic vector of a fault state with outward movement of an upper slideway, a running characteristic vector of a fault state with transverse interference of a lower stop pin, a running characteristic vector of a fault state with longitudinal interference of a lower stop pin and a running characteristic vector of a fault state with overpressure of a pinch roller.
The minkowski distance calculation formula is:
The invention provides a control method with sub-health pre-diagnosis and fault type clustering functions and a gantry crane controller. A control method with sub-health pre-diagnosis and fault type clustering functions comprises the following steps:
the method comprises the following steps: h for rotor position sensor output in control system with sub-health pre-diagnosis and fault type clustering functionsa、Hb、HcDetecting to form a speed feedback signal omega*；
Step two: output current i of control system with sub-health pre-diagnosis and fault type clustering functionsA、iBDetecting the torque direction signal DIR output by the absolute value circuit and the H output by the rotor position sensora、Hb、HcSynthesizing a current feedback signal i together by a current feedback signalf；
Step three: subtracting the aforementioned speed feedback signal omega from the speed given signal omega*Obtaining a speed error, and obtaining a signal u through proportional-integral regulation operation of integral saturation resistance on the speed errorgi；
Step four: signal ugiOutputting a torque control signal MCMD and the current feedback signal i through an absolute value circuitfComparing to obtain a current error, and outputting PWM (pulse width modulation) by proportional integral operation of integral saturation resistance of the current error;
step five: PWM of output and H of output of the aforementioned rotor position sensora、Hb、HcAnd the three-phase inverter is controlled together through phase change logic to drive the brushless direct current motor.
A door machine controller with sub-health pre-diagnosis and fault type clustering functions adopts a brushless direct current motor as a driving motor of a door controller, adopts a power MOS three-phase bridge circuit to feed the brushless direct current motor, adopts an STM32F429 chip as a built-in controller, and realizes the control of the position, the speed and the current of the brushless direct current motor. And the current is detected by two LEM Hall current sensors, the current value of the third phase can be calculated according to the kirchhoff current rule, and the sampling and conversion of the motor circuit signals are completed by utilizing a high-speed A/D channel of STM32F 429. The control software of the door machine controller further comprises a position control module, a speed control module, a D-axis current control module, a Q-axis current control module, a rotation conversion module, an inverse rotation conversion module, a speed position calculation software module, a speed position storage module, a torque calculation software module, a torque storage module, a CMAC sample input module, a stray code mapping module, a characteristic vector extraction module and a sub-health pre-diagnosis and alarm module, the automatic door controller further comprises a hardware module, the hardware module adopts a Hall current detector to form a two-phase current detection module, and adopts an intelligent power module IPM to form an inverter to drive a motor.
The invention relates to a door controller with sub-health pre-diagnosis and fault type clustering functions, which comprises the steps of detecting the door opening and closing actions of a rail vehicle door on line, automatically providing characteristic values of a door opening and closing curve for a monitoring system, pre-diagnosing whether the rail vehicle door is healthy or not in the door closing process through the calculation of an output characteristic vector of an operating state, and carrying out fault type clustering analysis on the operating state of the failed rail vehicle door opening and closing.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, and these embodiments are within the scope of the invention.
Claims (9)
1. A sub-health pre-diagnosis and fault type clustering method for a gantry crane system is characterized by comprising the following steps: utilizing a cerebellar neural network to realize sub-health pre-diagnosis and fault type clustering, segmenting the switching process of a gate body controlled by a gate machine system to form an input sample of the cerebellar neural network, wherein the cerebellar neural network comprises a virtual memory and a stray code, taking the stray code value as a feature vector, and realizing the sub-health pre-diagnosis and the fault type clustering according to the Minkowski distance of each feature vector in a vector space, and the process comprises the following steps:
the method comprises the following steps: the door machine controller divides the operation parameters in the current door opening and closing process into 4 sections: extracting time and displacement in the 4 stages of processes to form an input sample of the cerebellar neural network;
step two: obtaining a characteristic vector of the opening and closing state of the rail vehicle door through concept mapping and stray transformation of a cerebellar neural network according to the input sample obtained in the step one;
step three: and calculating the Minkowski distance between the current running characteristic vector and the typical characteristic vector obtained in advance, and performing sub-health pre-diagnosis and fault type clustering.
2. The method for sub-health pre-diagnosis and fault type clustering of a portal system according to claim 1, wherein: in the first step, the door opening and closing process is divided into the following steps: the method comprises the steps of rail vehicle door opening and closing speed increasing time, rail vehicle door opening and closing speed increasing displacement, rail vehicle door opening and closing speed constant time, rail vehicle door opening and closing speed constant displacement, rail vehicle door opening and closing speed reducing time, rail vehicle door opening and closing speed reducing displacement, rail vehicle door opening and closing speed pressing time and rail vehicle door opening and closing speed pressing displacement.
3. The method for sub-health pre-diagnosis and fault type clustering of a portal system according to claim 1, wherein: the generalization width C in the cerebellar neural network is 6. The spur coding function uses a division-residue method.
4. The method for sub-health pre-diagnosis and fault type clustering of a portal system according to claim 1, wherein: the representative feature vector includes: the method comprises the following steps of storing characteristic vectors of 9 typical running states in a door machine controller of the railway vehicle in advance, wherein the characteristic vectors of the running states of a no-fault state, a running characteristic vector of a V-shaped abnormal fault state, a running characteristic vector of a motor assembly loosening fault state, a running characteristic vector of a fault state with small change in centering size, a running characteristic vector of a fault state with large change in centering size, a running characteristic vector of a fault state with outward movement of an upper slideway, a running characteristic vector of a fault state with transverse interference of a lower stop pin, a running characteristic vector of a fault state with longitudinal interference of a lower stop pin and a running characteristic vector of a fault state with overpressure of a pinch roller.
5. The method for sub-health pre-diagnosis and fault type clustering of a portal system according to claim 1, wherein: said minkowski distance comprising:
minkowski distance in the vector space between the running eigenvector for a no-fault condition and the running eigenvector for a typical fault condition:
the shortest minkowski distance in the vector space between the running eigenvector for a no-fault condition and the running eigenvector for a typical fault condition:
l0＝min{D(W1，W2)，D(W1，W3)，...，D(W1，W9)]；
minkowski distance in vector space of the running eigenvector for this on-off state versus the running eigenvector for the no-fault state:
6. the method for sub-health pre-diagnosis and fault type clustering of a portal system according to claim 1, wherein: the basis for sub-health pre-diagnosis based on minkowski distance is:
when in use
Meanwhile, the running state of the gantry crane system is healthy;
when in use
Meanwhile, the running state of the gantry crane system is sub-healthy;
when in use
And meanwhile, the running state of the gantry crane system fails.
7. The method for sub-health pre-diagnosis and fault type clustering of a portal system according to claim 1, wherein: the basis for fault type clustering according to minkowski distance is:
8. A door machine controller with sub-health pre-diagnosis and fault type clustering functions is characterized in that: a brushless direct current motor is used as a driving motor of the door controller, a power MOS three-phase bridge circuit is used for feeding the brushless direct current motor, and STM32F429 is used as a control chip.
9. The door operator controller with sub-health pre-diagnosis and fault type clustering functions as claimed in claim 8, wherein the control software comprises a position control module, a speed control module, a D-axis current control module, a Q-axis current control module, a rotation transformation module, an inverse rotation transformation module, a speed position calculation software module, a speed position storage module, a torque calculation software module, a torque storage module, a cerebellar neural network sample input module, a stray code mapping module, a feature vector extraction module, and a sub-health pre-diagnosis and alarm module, and the door operator controller further comprises a hardware module, the hardware module adopts a hall current detector to form a two-phase current detection module, and adopts an intelligent power module IPM to form an inverter to drive the motor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011222663.XA CN112348078A (en) | 2020-11-09 | 2020-11-09 | Gate machine controller with sub-health pre-diagnosis and fault type clustering functions |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011222663.XA CN112348078A (en) | 2020-11-09 | 2020-11-09 | Gate machine controller with sub-health pre-diagnosis and fault type clustering functions |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112348078A true CN112348078A (en) | 2021-02-09 |
Family
ID=74429553
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011222663.XA Pending CN112348078A (en) | 2020-11-09 | 2020-11-09 | Gate machine controller with sub-health pre-diagnosis and fault type clustering functions |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112348078A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113392874A (en) * | 2021-05-13 | 2021-09-14 | 中车唐山机车车辆有限公司 | Rail vehicle abnormal state diagnosis method and device and terminal equipment |
| CN113591376A (en) * | 2021-07-23 | 2021-11-02 | 广州新科佳都科技有限公司 | Platform door abnormity detection method and device based on curve correlation segmentation mechanism |
| CN114488797A (en) * | 2022-01-07 | 2022-05-13 | 南京工程学院 | Automatic door opening and closing motion curve planning method, controller and control method |
| CN114488797B (en) * | 2022-01-07 | 2024-07-02 | 南京工程学院 | Automatic door opening and closing motion curve planning method, controller and control method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20000056839A (en) * | 1999-02-26 | 2000-09-15 | 윤종용 | hydraulic system using neural network and control method thereof |
| US6529816B1 (en) * | 1998-08-07 | 2003-03-04 | Yamaha Hatsudoki Kabushiki Kaisha | Evolutionary controlling system for motor |
| CN1472671A (en) * | 2003-06-19 | 2004-02-04 | 上海交通大学 | Method for apparatus status monitoring and performance degenerate fore casting based on network |
| CN101162141A (en) * | 2007-11-12 | 2008-04-16 | 浙江大学 | Agriculture working vehicle attitude estimating device based on nerval net |
| CN102830341A (en) * | 2012-08-28 | 2012-12-19 | 南京航空航天大学 | Online intelligent fault prediction method for power electronic circuit based on RS-CMAC (rough sets and cerebellar model articulation controller) |
| CN106406295A (en) * | 2016-12-02 | 2017-02-15 | 南京康尼机电股份有限公司 | Rail transit vehicle door system fault diagnosis and early warning method based on multiple conditions |
| CN107153841A (en) * | 2017-04-24 | 2017-09-12 | 南京康尼机电股份有限公司 | A kind of inferior health Forecasting Methodology of urban rail transit vehicles door system |
| CN109116150A (en) * | 2018-08-03 | 2019-01-01 | 福州大学 | A kind of converters method for diagnosing faults based on Cerebellar Model Articulation Controller |
-
2020
- 2020-11-09 CN CN202011222663.XA patent/CN112348078A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6529816B1 (en) * | 1998-08-07 | 2003-03-04 | Yamaha Hatsudoki Kabushiki Kaisha | Evolutionary controlling system for motor |
| KR20000056839A (en) * | 1999-02-26 | 2000-09-15 | 윤종용 | hydraulic system using neural network and control method thereof |
| CN1472671A (en) * | 2003-06-19 | 2004-02-04 | 上海交通大学 | Method for apparatus status monitoring and performance degenerate fore casting based on network |
| CN101162141A (en) * | 2007-11-12 | 2008-04-16 | 浙江大学 | Agriculture working vehicle attitude estimating device based on nerval net |
| CN102830341A (en) * | 2012-08-28 | 2012-12-19 | 南京航空航天大学 | Online intelligent fault prediction method for power electronic circuit based on RS-CMAC (rough sets and cerebellar model articulation controller) |
| CN106406295A (en) * | 2016-12-02 | 2017-02-15 | 南京康尼机电股份有限公司 | Rail transit vehicle door system fault diagnosis and early warning method based on multiple conditions |
| CN107153841A (en) * | 2017-04-24 | 2017-09-12 | 南京康尼机电股份有限公司 | A kind of inferior health Forecasting Methodology of urban rail transit vehicles door system |
| CN109116150A (en) * | 2018-08-03 | 2019-01-01 | 福州大学 | A kind of converters method for diagnosing faults based on Cerebellar Model Articulation Controller |
Non-Patent Citations (4)
| Title |
|---|
| KUO-HSIANG CHENG: "CMAC-based neuro-fuzzy approach for complex system modeling", 《NEUROCOMPUTING》, vol. 72, pages 1763 - 1774, XP025976348, DOI: 10.1016/j.neucom.2008.08.011 * |
| PI-YUN CHEN 等: "A Motor Fault Diagnosis System Based on Cerebellar Model Articulation Controller", 《IEEE ACCESS》, pages 120326 - 120336 * |
| 刘乾 等: "无人水下机器人推进系统故障诊断与容错控制", 《系统仿真学报》, vol. 22, no. 1, pages 96 - 101 * |
| 程学龙 等: "深海载人潜水器推进器系统故障诊断的新型主元分析算法", 《控制理论与应用》, vol. 35, no. 12, pages 1796 - 1804 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113392874A (en) * | 2021-05-13 | 2021-09-14 | 中车唐山机车车辆有限公司 | Rail vehicle abnormal state diagnosis method and device and terminal equipment |
| CN113591376A (en) * | 2021-07-23 | 2021-11-02 | 广州新科佳都科技有限公司 | Platform door abnormity detection method and device based on curve correlation segmentation mechanism |
| CN113591376B (en) * | 2021-07-23 | 2023-07-14 | 广州新科佳都科技有限公司 | Platform door anomaly detection method and device based on curve association segmentation mechanism |
| CN114488797A (en) * | 2022-01-07 | 2022-05-13 | 南京工程学院 | Automatic door opening and closing motion curve planning method, controller and control method |
| CN114488797B (en) * | 2022-01-07 | 2024-07-02 | 南京工程学院 | Automatic door opening and closing motion curve planning method, controller and control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | A review of fault detection and diagnosis for the traction system in high-speed trains | |
| CN112348078A (en) | Gate machine controller with sub-health pre-diagnosis and fault type clustering functions | |
| US7385365B2 (en) | Method for the detection of abnormalities of electric motors | |
| WO2008053554A1 (en) | Electric motor car control apparatus | |
| US10886865B2 (en) | Motor controller and electric power steering device | |
| CN104122479A (en) | Online detection method for open-circuit faults of power tubes of induction-motor vector control system | |
| JP3801906B2 (en) | Electric vehicle control device and control method | |
| CN111717031B (en) | Pure electric vehicle torque function safety monitoring method and system and vehicle | |
| US11267504B2 (en) | Apparatus and method for controlling steer-by-wire system | |
| CN105871283B (en) | The fault-tolerant drive system of automobile permanent magnet synchronous motor and method under sensor fault | |
| CN112511061A (en) | Rotor position sensor fault detection and fault-tolerant control method and system | |
| CN107787549A (en) | The drive device of three-phase synchronous motor | |
| JP2014045535A (en) | Motor drive system | |
| US7154404B2 (en) | Electric power steering system having failure detection apparatus | |
| Zhou et al. | UKF-based sensor fault diagnosis of PMSM drives in electric vehicles | |
| CN114441958A (en) | IGBT open-circuit fault diagnosis method for permanent magnet synchronous motor driver | |
| US10972032B2 (en) | Method and apparatus for detecting failure of current sensor of motor | |
| US9421998B2 (en) | Method and apparatus for detecting motor error of motor driven power steering | |
| WO2020124936A1 (en) | Fault detection method for servo dynamic brake circuit | |
| CN111190102A (en) | Method and device for detecting motor flux linkage | |
| JP5707761B2 (en) | Phase loss diagnosis apparatus and phase loss diagnosis method | |
| Zidani et al. | Diagnosis of speed sensor failure in induction motor drive | |
| Gaeid et al. | Wavelet fault diagnosis of induction motor | |
| JPWO2022224485A5 (en) | ||
| JP7111011B2 (en) | electric vehicle controller |
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 |