CN111965540A - Fault detection method for shaft belt motor in shaft belt propelling mechanism - Google Patents

Fault detection method for shaft belt motor in shaft belt propelling mechanism Download PDF

Info

Publication number
CN111965540A
CN111965540A CN202010810218.9A CN202010810218A CN111965540A CN 111965540 A CN111965540 A CN 111965540A CN 202010810218 A CN202010810218 A CN 202010810218A CN 111965540 A CN111965540 A CN 111965540A
Authority
CN
China
Prior art keywords
fault
neural network
expert system
convolutional neural
shaft
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
Application number
CN202010810218.9A
Other languages
Chinese (zh)
Inventor
岳凡
邵诗逸
武治江
王晓梅
高双建
赵红品
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuxi Silent Electric System Ses Technology Co ltd
Original Assignee
Wuxi Silent Electric System Ses Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Wuxi Silent Electric System Ses Technology Co ltd filed Critical Wuxi Silent Electric System Ses Technology Co ltd
Priority to CN202010810218.9A priority Critical patent/CN111965540A/en
Publication of CN111965540A publication Critical patent/CN111965540A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/34—Testing dynamo-electric machines
    • G01R31/343—Testing dynamo-electric machines in operation

Abstract

The invention discloses a fault detection method for a shaft motor in a shaft belt propelling mechanism, which analyzes and processes various detection data of the shaft motor through a convolutional neural network expert system, and compares and analyzes an analysis result with a fault database to obtain the fault type of the shaft motor so as to guide the maintenance operation of related personnel. The convolutional neural network expert system has the functions of continuous self-learning and updating, perfects the database to improve the detection accuracy, and can reduce the fault removal difficulty of maintenance personnel, thereby saving a large amount of time.

Description

Fault detection method for shaft belt motor in shaft belt propelling mechanism
Technical Field
The invention relates to a fault detection method for a shaft belt motor in a shaft belt propelling mechanism.
Background
When the shaft belt motor has a fault, an operator needs to inquire a related fault list according to a series of fault codes fed back by the fault monitoring system, further confirm the fault type and then carry out fault maintenance, so that the operation is complicated, and the fault maintenance time is prolonged. The neural network expert system-based micro-inverter fault detection method disclosed in patent ZL103293415B describes a fault detection method for a micro-inverter by a neural network expert system, but the method is not suitable for fault detection of an axle belt motor with complex working conditions, and the method lacks a detailed construction process of the neural network expert system.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: a fault detection method of a shaft belt motor in a shaft belt propulsion mechanism based on a convolutional neural network is provided.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: the fault detection method of the shaft belt motor in the shaft belt propelling mechanism comprises the shaft belt motor, wherein the shaft belt motor is connected with a direct current bus through a power utilization circuit, and a circuit breaker, a frequency conversion module and a fuse are sequentially arranged on the power utilization circuit from the shaft belt motor; the output shaft of the shaft motor is connected with the first input end of the three-terminal gear box; the output end of the three-terminal gear box is connected with the propeller, the second input end of the three-terminal gear box is connected with the output shaft of the diesel propulsion host through the clutch, the frequency conversion module is provided with a first temperature sensor, a first alternating voltage sensor is arranged on a line between the frequency conversion module and the circuit breaker, a first alternating current sensor is arranged on a line between the circuit breaker and the shaft motor, and a second temperature sensor, a third temperature sensor, a fourth temperature sensor and a rotating speed sensor are arranged at the shaft motor; the first temperature sensor, the first alternating current voltage sensor, the first alternating current sensor, the second temperature sensor, the third temperature sensor, the fourth temperature sensor and the rotating speed sensor are electrically connected with the shaft belt propulsion mechanism control module;
the detection process is as follows:
step a, constructing a database
Constructing a fault database according to the measured values and fault types of the shaft motor in the working process, wherein the database comprises known fault types which are possible to generate and expression forms of corresponding faults;
step b, constructing a convolutional neural network expert system
Step b.1, presetting the number N of convolution components of the convolutional neural network expert system, and initializing N to be 1;
step b.2, constructing M measured value input matrixes X and corresponding fault matrixes Y of the convolutional neural network expert system based on the measured values and the fault types in the database;
step b.3, performing convolution calculation on the input matrix X, wherein the formula is as follows:
Figure BDA0002630694460000021
in the formula, XiIs the ith input map (1. ltoreq. i. ltoreq.M),
Figure BDA0002630694460000022
is the L-th convolution kernel, LiIs the total number of cores in the ith convolutional layer,
Figure BDA0002630694460000023
is the deviation of the measured value,
Figure BDA0002630694460000024
is the ith convolution output map, r represents the local region of shared weight;
step b.4, increase convolution output mapping by following formula
Figure BDA0002630694460000025
Non-linear property of
Figure BDA0002630694460000026
Step b.5, through the calculation of the pooling layer, the translation invariance of the data is increased and overfitting is prevented, and the formula is as follows:
Figure BDA0002630694460000031
in the formula, each neuron
Figure BDA0002630694460000032
Part is converged at
Figure BDA0002630694460000033
2 × 2 area in (a);
step b.6, calculating an output matrix of the convolutional neural network expert system:
Figure BDA0002630694460000034
b.7 calculating the neurons according to the following formula
Figure BDA0002630694460000035
Probability distribution of corresponding fault
Figure BDA00026306944600000312
Figure BDA0002630694460000036
In the formula, exp (θ)i) Is the ith neuron
Figure BDA0002630694460000037
The probability distribution of (a) is determined,
Figure BDA0002630694460000038
Figure BDA0002630694460000039
is a neuron
Figure BDA00026306944600000310
The weight of (c); s is the number of elements contained in the output matrix;
step b.8, determining a final output matrix W according to the probability distribution of each fault;
step b.9, the convolutional neural network expert system compares and analyzes the output matrix W and the known fault matrix Y:
if the fault type in the output matrix W calculated by the convolutional neural network expert system is not identical to the fault type in the known fault matrix Y, executing the steps b.10 to b.11;
if the fault type in the output matrix W calculated by the convolutional neural network expert system is the same as the fault type in the known fault matrix Y, executing the step b.12;
and b.10, calculating the convolution component as follows: n is N + 1;
step b.11, output size phiiConverting into an input matrix X, and executing the steps b.3 to b.9;
step b.12, determining the value of the number N of the convolution components and each scale factor
Figure BDA00026306944600000311
Numerical value of (A) and each neuron
Figure BDA0002630694460000041
Weight of (2)
Figure BDA0002630694460000042
And constructing a convolutional neural network expert system;
step c, data acquisition
Through setting up the sensor real-time acquisition required data in axle area motor and frequency conversion module department to transmit to based on convolution neural network expert system, include: the first temperature sensor collects the temperature t of the frequency conversion module1(ii) a Three-phase voltage U of shaft motor acquired by first alternating voltage sensoru、Uv、Uw(ii) a The first AC current sensor collects three-phase current I of the shaft motoru、Iv、Iw(ii) a The second temperature sensor collects the temperature t of the bearing at the driving end of the shaft motor2(ii) a The third temperature sensor acquires the temperature t of the three-phase winding of the shaft motoru、tv、tw(ii) a The fourth temperature sensor acquires the temperature t of the bearing at the non-driving end of the shaft-driven motor3(ii) a The rotation speed sensor collects the output rotation speed n of the shaft motor,the shaft belt propulsion mechanism control module calculates the frequency f of the shaft belt motor according to a formula f, wherein p is the pole pair number of the shaft belt motor and the like;
d, analyzing and processing the data collected in the step c through a convolutional neural network expert system:
step d.1, initializing the execution times j of the convolutional neural network expert system to be 0;
d.2, constructing an input matrix X of the convolutional neural network expert system: x ═ t1 Uu Uv Uw … Iw n]；
D.3, analyzing the input matrix X by adopting the convolutional neural network expert system constructed in the step b;
step d.4, calculating the convolution component as follows: j-1;
d.5, judging whether the j is less than or equal to N by the convolutional neural network expert system; if yes, the output matrix phi is setiConverting into an input matrix X, and executing the steps d.3 to d.5; if not, the output matrix phi of the convolutional neural network expert systemi；
Step e, failure inquiry
The control module of the shaft belt propulsion mechanism outputs a result phi according to the convolutional neural network expert systemiAccessing a database; and if the corresponding fault is inquired in the database, the fault is sent to the fault display module for displaying, and the fault display module is used for guiding maintenance workers to maintain.
As a preferred scheme, in the step e, if no corresponding fault is inquired in the database, the expert judges the expression forms of the fault and the corresponding fault, and if the expression forms of the fault and the corresponding fault are matched, the expression forms of the fault and the corresponding fault are brought into the database and sent to the fault display module for displaying, so as to guide maintenance workers to maintain and update the convolutional neural network expert system; and if the expression forms of the faults are not matched with the expression forms of the corresponding faults, obtaining the faults corresponding to the corresponding expression forms according to the experience of expert personnel, bringing the faults into the database, and updating the convolutional neural network expert system.
The invention has the beneficial effects that:
in the aspect of shaft motor fault detection, the convolutional neural network expert system analyzes and processes various detection data of the shaft motor, compares the analysis result with a fault database, and acquires the fault type of the shaft motor so as to guide maintenance operation of related personnel.
The convolutional neural network expert system has the functions of continuous self-learning and updating, perfects the database to improve the detection accuracy, and can reduce the trouble shooting difficulty of maintenance personnel so as to save a large amount of time.
Drawings
FIG. 1 is a schematic representation of the construction of the present invention shaft belt propulsion system.
In fig. 1: the system comprises a 1-fuse, a 2-frequency conversion module, a 3-circuit breaker, a 4-shaft motor, a 5-three-terminal gear box, a 6-propeller, a 7-clutch, an 8-diesel propulsion host, a 9-shaft propulsion mechanism control module, a 10-first temperature sensor, a 11-first alternating voltage sensor, a 12-first alternating current sensor, a 13-second temperature sensor, a 14-third temperature sensor, a 15-fourth temperature sensor and a 16-rotating speed sensor.
Detailed Description
Specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
As shown in fig. 1, a method for detecting a fault of a shaft belt motor in a shaft belt propulsion mechanism, wherein the shaft belt propulsion mechanism comprises a shaft belt motor 4, the shaft belt motor 4 is connected with a direct current bus through an electric line, and a circuit breaker 3, a frequency conversion module 2 and a fuse 1 are sequentially arranged on the electric line from the shaft belt motor 4; the output shaft of the shaft motor 4 is connected with the first input end of the three-terminal gear box 5; the output end of the three-terminal gear box 5 is connected with the propeller 6, the second input end of the three-terminal gear box 5 is connected with the output shaft of the diesel propulsion host 8 through the clutch 7, the frequency conversion module 2 is provided with a first temperature sensor 10, a first alternating current voltage sensor 11 is arranged on a line between the frequency conversion module 2 and the circuit breaker 3, a first alternating current sensor 12 is arranged on a line between the circuit breaker 3 and the shaft motor 4, and a second temperature sensor 13, a third temperature sensor 14, a fourth temperature sensor 15 and a rotating speed sensor 16 are arranged at the shaft motor 4; the first temperature sensor 10, the first alternating voltage sensor 11, the first alternating current sensor 12, the second temperature sensor 13, the third temperature sensor 14, the fourth temperature sensor 15 and the rotating speed sensor 16 are electrically connected with the shaft belt propulsion mechanism control module 9;
the detection process is as follows:
step a, constructing a database
Constructing a fault database according to the measured values and fault types of the shaft motor 4 in the working process, wherein the database comprises known fault types which are possible to generate and expression forms of corresponding faults;
step b, constructing a convolutional neural network expert system
Step b.1, presetting the number N of convolution components of the convolutional neural network expert system, and initializing N to be 1;
step b.2, constructing M measured value input matrixes X and corresponding fault matrixes Y of the convolutional neural network expert system based on the measured values and the fault types in the database: x ═ tu tv tw … Iu Iv Iw]；
Y ═ motor winding insulation damage … motor coil interphase short circuit ];
step b.3, performing convolution calculation on the input matrix X, wherein the formula is as follows:
Figure BDA0002630694460000071
in the formula, XiIs the ith input map (1. ltoreq. i. ltoreq.M),
Figure BDA0002630694460000072
is the L-th convolution kernel, LiIs the total number of cores in the ith convolutional layer,
Figure BDA0002630694460000073
is the deviation of the measured value,
Figure BDA0002630694460000074
is the ith convolution output map, r represents the local region of shared weight;
step b.4, increase convolution output mapping by following formula
Figure BDA0002630694460000075
The non-linear property of (2):
Figure BDA0002630694460000076
step b.5, through the calculation of the pooling layer, the translation invariance of the data is increased and overfitting is prevented, and the formula is as follows:
Figure BDA0002630694460000077
in the formula, each neuron
Figure BDA0002630694460000078
Part is converged at
Figure BDA0002630694460000079
2 × 2 area in (a);
step b.6, calculating an output matrix of the convolutional neural network expert system:
Figure BDA0002630694460000081
b.7 calculating the neurons according to the following formula
Figure BDA0002630694460000082
Probability distribution of corresponding fault
Figure BDA0002630694460000083
Figure BDA0002630694460000084
In the formula, exp (θ)i) Is the ith neuron
Figure BDA0002630694460000085
The probability distribution of (a) is determined,
Figure BDA0002630694460000086
Figure BDA0002630694460000087
is a neuron
Figure BDA0002630694460000088
The weight of (c); s is the number of elements contained in the output matrix;
step b.8, determining a final output matrix W according to the probability distribution of each fault;
step b.9, the convolutional neural network expert system compares and analyzes the output matrix W and the known fault matrix Y:
if the fault type in the output matrix W calculated by the convolutional neural network expert system is not identical to the fault type in the known fault matrix Y, executing the steps b.10 to b.11;
if the fault type in the output matrix W calculated by the convolutional neural network expert system is the same as the fault type in the known fault matrix Y, executing the step b.12;
and b.10, calculating the convolution component as follows: n is N + 1;
step b.11, output size phiiConverting into an input matrix X, and executing the steps b.3 to b.9;
step b.12, determining the value of the number N of the convolution components and each scale factor
Figure BDA0002630694460000089
Numerical value of (A) and each neuron
Figure BDA00026306944600000810
Weight of (2)
Figure BDA00026306944600000811
And constructing a convolutional neural network expert system;
step c, data acquisition
Through setting up the sensor real-time acquisition required data in axle area motor 4 and frequency conversion module 2 department to transmit to based on convolution neural network expert system, include: the first temperature sensor 10 collects the temperature t of the frequency conversion module 21(ii) a The first alternating voltage sensor 11 collects the three-phase voltage U of the shaft motor 4u、Uv、Uw(ii) a The first alternating current sensor 12 collects the three-phase current I of the shaft motor 4u、Iv、Iw(ii) a The second temperature sensor 13 collects the temperature t of the bearing at the driving end of the shaft motor 42(ii) a The third temperature sensor 14 acquires the temperature t of the three-phase winding of the shaft motor 4u、tv、tw(ii) a The fourth temperature sensor 15 acquires the temperature t of the bearing at the non-driving end of the shaft motor 43(ii) a The rotating speed sensor 16 acquires the output rotating speed n of the shaft motor 4, and the shaft propulsion mechanism control module 9 calculates the frequency f of the shaft motor 4 according to a formula f, namely np/60, wherein p is the pole pair number of the shaft motor 4 and the like;
d, analyzing and processing the data collected in the step c through a convolutional neural network expert system:
step d.1, initializing the execution times j of the convolutional neural network expert system to be 0;
d.2, constructing an input matrix X of the convolutional neural network expert system: x ═ t1 Uu Uv Uw … Iw n]；
D.3, analyzing the input matrix X by adopting the convolutional neural network expert system constructed in the step b;
step d.4, calculating the convolution component as follows: j-1;
d.5, judging whether the j is less than or equal to N by the convolutional neural network expert system; if yes, the output matrix phi is setiConverting into an input matrix X, and executing the steps d.3 to d.5; if not, the output matrix phi of the convolutional neural network expert systemi；
Step e, failure inquiry
The shaft belt propulsion mechanism control module 9 is based on a convolutional neural networkOutput result phi of expert systemiAccessing a database; if the corresponding fault is inquired in the database, the fault is sent to a fault display module for displaying and is used for guiding maintenance workers to maintain;
if the corresponding fault is not inquired in the database, the expert judges the expression forms of the fault and the corresponding fault, if the expression forms of the fault and the corresponding fault are matched, the expression forms of the fault and the corresponding fault are brought into the database, and the fault is sent to a fault display module to be displayed for guiding maintenance workers to maintain, and meanwhile, the convolutional neural network expert system is updated; and if the expression forms of the faults are not matched with the expression forms of the corresponding faults, obtaining the faults corresponding to the corresponding expression forms according to the experience of expert personnel, bringing the faults into the database, and updating the convolutional neural network expert system.
The above-mentioned embodiments are merely illustrative of the principles and effects of the present invention, and some embodiments may be used, not restrictive; it should be noted that, for those skilled in the art, various changes and modifications can be made without departing from the inventive concept of the present invention, and these changes and modifications belong to the protection scope of the present invention.

Claims (2)

1. A fault detection method for a shaft belt motor in a shaft belt propelling mechanism is characterized in that the shaft belt propelling mechanism comprises a shaft belt motor, the shaft belt motor is connected with a direct current bus through a power utilization circuit, and a circuit breaker, a frequency conversion module and a fuse are sequentially arranged on the power utilization circuit from the shaft belt motor; the output shaft of the shaft motor is connected with the first input end of the three-terminal gear box; the output end of the three-terminal gear box is connected with the propeller, the second input end of the three-terminal gear box is connected with the output shaft of the diesel propulsion host through the clutch, the frequency conversion module is provided with a first temperature sensor, a first alternating voltage sensor is arranged on a line between the frequency conversion module and the circuit breaker, a first alternating current sensor is arranged on a line between the circuit breaker and the shaft motor, and a second temperature sensor, a third temperature sensor, a fourth temperature sensor and a rotating speed sensor are arranged at the shaft motor; the first temperature sensor, the first alternating current voltage sensor, the first alternating current sensor, the second temperature sensor, the third temperature sensor, the fourth temperature sensor and the rotating speed sensor are electrically connected with the shaft belt propulsion mechanism control module;
the detection process is as follows:
step a, constructing a database
Constructing a fault database according to the measured values and fault types of the shaft motor in the working process, wherein the database comprises known fault types which are possible to generate and expression forms of corresponding faults;
step b, constructing a convolutional neural network expert system
Step b.1, presetting the number N of convolution components of the convolutional neural network expert system, and initializing N to be 1;
step b.2, constructing M measured value input matrixes X and corresponding fault matrixes Y of the convolutional neural network expert system based on the measured values and the fault types in the database;
step b.3, performing convolution calculation on the input matrix X, wherein the formula is as follows:
Figure FDA0002630694450000011
Figure FDA0002630694450000021
in the formula, XiIs the ith input map (1. ltoreq. i. ltoreq.M),
Figure FDA0002630694450000022
is the L-th convolution kernel, LiIs the total number of cores in the ith convolutional layer,
Figure FDA0002630694450000023
is the deviation of the measured value,
Figure FDA0002630694450000024
is the ith convolution output mapAnd r represents a local region sharing a weight;
step b.4, increase convolution output mapping by following formula
Figure FDA0002630694450000025
The non-linear property of (2):
Figure FDA0002630694450000026
step b.5, through the calculation of the pooling layer, the translation invariance of the data is increased and overfitting is prevented, and the formula is as follows:
Figure FDA0002630694450000027
in the formula, each neuron
Figure FDA0002630694450000028
All are converged at
Figure FDA0002630694450000029
2 × 2 area in (a);
step b.6, calculating an output matrix of the convolutional neural network expert system:
Figure FDA00026306944500000210
b.7 calculating the neurons according to the following formula
Figure FDA00026306944500000211
Probability distribution of corresponding fault
Figure FDA00026306944500000212
Figure FDA00026306944500000213
In the formula, exp (θ)i) Is the ith neuron
Figure FDA00026306944500000214
The probability distribution of (a) is determined,
Figure FDA00026306944500000215
Figure FDA00026306944500000216
is a neuron
Figure FDA00026306944500000217
The weight of (c); s is the number of elements contained in the output matrix;
step b.8, determining a final output matrix W according to the probability distribution of each fault;
step b.9, the convolutional neural network expert system compares and analyzes the output matrix W and the known fault matrix Y:
if the fault type in the output matrix W calculated by the convolutional neural network expert system is not identical to the fault type in the known fault matrix Y, executing the steps b.10 to b.11;
if the fault type in the output matrix W calculated by the convolutional neural network expert system is the same as the fault type in the known fault matrix Y, executing the step b.12;
and b.10, calculating the convolution component as follows: n is N + 1;
step b.11, output size phiiConverting into an input matrix X, and executing the steps b.3 to b.9;
step b.12, determining the value of the number N of the convolution components and each scale factor
Figure FDA0002630694450000031
Numerical value of (A) and each neuron
Figure FDA0002630694450000032
Weight of (2)
Figure FDA0002630694450000033
And constructing a convolutional neural network expert system;
step c, data acquisition
Through setting up the sensor real-time acquisition required data in axle area motor and frequency conversion module department to transmit to based on convolution neural network expert system, include: the first temperature sensor collects the temperature t of the frequency conversion module1(ii) a Three-phase voltage U of shaft motor acquired by first alternating voltage sensoru、Uv、Uw(ii) a The first AC current sensor collects three-phase current I of the shaft motoru、Iv、Iw(ii) a The second temperature sensor collects the temperature t of the bearing at the driving end of the shaft motor2(ii) a The third temperature sensor acquires the temperature t of the three-phase winding of the shaft motoru、tv、tw(ii) a The fourth temperature sensor acquires the temperature t of the bearing at the non-driving end of the shaft-driven motor3(ii) a The method comprises the steps that a rotating speed sensor collects the output rotating speed n of a shaft motor, and a shaft propulsion mechanism control module calculates the frequency f of the shaft motor according to the formula f, wherein p is the number of pole pairs of the shaft motor and the like;
d, analyzing and processing the data collected in the step c through a convolutional neural network expert system:
step d.1, initializing the execution times j of the convolutional neural network expert system to be 0;
d.2, constructing an input matrix X of the convolutional neural network expert system: x ═ t1 Uu Uv Uw…Iw n]；
D.3, analyzing the input matrix X by adopting the convolutional neural network expert system constructed in the step b;
step d.4, calculating the convolution component as follows: j-1;
d.5, judging whether the j is less than or equal to N by the convolutional neural network expert system; if yes, the output matrix phi is setiConverting into an input matrix X, and executing the steps d.3 to d.5; if not, the output matrix phi of the convolutional neural network expert systemi；
Step e, failure inquiry
Shaft belt propulsion mechanism control module rootOutput result phi according to convolutional neural network expert systemiAccessing a database; and if the corresponding fault is inquired in the database, the fault is sent to the fault display module for displaying, and the fault display module is used for guiding maintenance workers to maintain.
2. The method for detecting the fault of the shaft belt motor in the shaft belt propelling mechanism as claimed in claim 1, wherein: in the step e, if the corresponding fault is not inquired in the database, the expert judges the expression forms of the fault and the corresponding fault, if the expression forms of the fault and the corresponding fault are matched, the expression forms of the fault and the corresponding fault are brought into the database, and the fault is sent to a fault display module to be displayed, so that a maintenance worker can be guided to maintain the fault, and meanwhile, the convolutional neural network expert system is updated; and if the expression forms of the faults are not matched with the expression forms of the corresponding faults, obtaining the faults corresponding to the corresponding expression forms according to the experience of expert personnel, bringing the faults into the database, and updating the convolutional neural network expert system.
CN202010810218.9A 2020-08-13 2020-08-13 Fault detection method for shaft belt motor in shaft belt propelling mechanism Pending CN111965540A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010810218.9A CN111965540A (en) 2020-08-13 2020-08-13 Fault detection method for shaft belt motor in shaft belt propelling mechanism

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010810218.9A CN111965540A (en) 2020-08-13 2020-08-13 Fault detection method for shaft belt motor in shaft belt propelling mechanism

Publications (1)

Publication Number Publication Date
CN111965540A true CN111965540A (en) 2020-11-20

Family

ID=73365329

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010810218.9A Pending CN111965540A (en) 2020-08-13 2020-08-13 Fault detection method for shaft belt motor in shaft belt propelling mechanism

Country Status (1)

Country Link
CN (1) CN111965540A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114035120A (en) * 2021-11-04 2022-02-11 合肥工业大学 Three-level inverter open-circuit fault diagnosis method and system based on improved CNN
CN116882968A (en) * 2023-06-29 2023-10-13 三峡科技有限责任公司 Design and implementation method for fault defect overall process treatment
CN116882968B (en) * 2023-06-29 2024-04-26 三峡科技有限责任公司 Design and implementation method for fault defect overall process treatment

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114035120A (en) * 2021-11-04 2022-02-11 合肥工业大学 Three-level inverter open-circuit fault diagnosis method and system based on improved CNN
CN116882968A (en) * 2023-06-29 2023-10-13 三峡科技有限责任公司 Design and implementation method for fault defect overall process treatment
CN116882968B (en) * 2023-06-29 2024-04-26 三峡科技有限责任公司 Design and implementation method for fault defect overall process treatment

Similar Documents

Publication Publication Date Title
Gao et al. Soft computing methods in motor fault diagnosis
Siddique et al. Applications of artificial intelligence techniques for induction machine stator fault diagnostics
CN106777606A (en) A kind of gearbox of wind turbine failure predication diagnosis algorithm
CN111907679B (en) Control method of direct-current networking ship hybrid power propulsion system
CN107247230A (en) A kind of electric rotating machine state monitoring method based on SVMs and data-driven
CN105512483B (en) Aero-engine predicting residual useful life based on damage propagation model and data analysis
CN110735769A (en) method, device and system for predicting fan faults
CN114813105A (en) Gear box fault early warning method and system based on working condition similarity evaluation
CN112987687B (en) Cloud-end integrated intelligent line control chassis health monitoring system and method
CN112041819B (en) Method for monitoring and identifying sensor faults in an electric drive system
CN111965540A (en) Fault detection method for shaft belt motor in shaft belt propelling mechanism
CN114154731A (en) Wind power ultra-short-term power prediction method and system based on deep learning
CN109709483A (en) A kind of variable-pitch system of wind turbine generator method for diagnosing faults
CN114491828A (en) Compound fault diagnosis method based on multi-gating hybrid expert multi-task learning
CN113887606A (en) Electronic equipment control system fault diagnosis method based on fault tree establishment
CN116067432A (en) Escalator variable working condition fault diagnosis method
CN115614292A (en) Vibration monitoring device and method for vertical water pump unit
Vladov et al. Neural network method for helicopters turboshaft engines working process parameters identification at flight modes
CN106872894B (en) Fault detection method and device for three-phase motor
CN105652120A (en) Power transformer fault detection method and detection system
CN116068303A (en) Private capacity-increasing on-line monitoring method for special transformer based on data driving
Pashaee et al. Fault Diagnosing Of An Induction Motor Based On Signal Fusion Using One-Dimensional Convolutional Neural Network
CN112613613A (en) Three-phase induction motor fault analysis method based on pulse neurolemma system
Qiang et al. State-of-the-art in soft computing-based motor fault diagnosis
CN114046228B (en) Wind turbine generator abnormality diagnosis method and system

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