CN116186610A - Motor fault diagnosis method of Lorenz stochastic resonance system based on particle swarm - Google Patents

Motor fault diagnosis method of Lorenz stochastic resonance system based on particle swarm Download PDF

Info

Publication number
CN116186610A
CN116186610A CN202310268032.9A CN202310268032A CN116186610A CN 116186610 A CN116186610 A CN 116186610A CN 202310268032 A CN202310268032 A CN 202310268032A CN 116186610 A CN116186610 A CN 116186610A
Authority
CN
China
Prior art keywords
lorenz
stochastic resonance
resonance system
data
signal
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
CN202310268032.9A
Other languages
Chinese (zh)
Inventor
于文新
刘美婷
王俊年
李目
周躜波
肖求美
钟广林
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.)
Hunan University of Science and Technology
Original Assignee
Hunan University of Science and Technology
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 Hunan University of Science and Technology filed Critical Hunan University of Science and Technology
Priority to CN202310268032.9A priority Critical patent/CN116186610A/en
Publication of CN116186610A publication Critical patent/CN116186610A/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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • G06N3/04—Architecture, e.g. interconnection topology
    • 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
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
    • Y04S10/52—Outage or fault management, e.g. fault detection or location

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Control Of Electric Motors In General (AREA)

Abstract

The invention provides a motor fault diagnosis method of a Lorenz stochastic resonance system based on particle swarms, which comprises the steps of collecting stator current data of a motor under different fault types by using a current sensor as an original signal; inputting the original signal data into a Lorenz-like stochastic resonance system based on a particle swarm, adaptively adjusting parameters of the Lorenz-like stochastic resonance system by taking a signal-to-noise ratio as an objective function, and outputting data with optimal signal-to-noise ratio under optimal parameters; and inputting the data into a learned extreme learning machine for fault diagnosis. The Lorenz-like stochastic resonance system provided by the invention has the advantages of multiple adjustable parameters, strong plasticity and the like, and provides a larger adjustment space for the system to adaptively adjust parameters of fault signals of different types. The optimal signal-to-noise ratio data output by the system can highlight signal characteristics, so that fault diagnosis can be accurately carried out.

Description

Motor fault diagnosis method of Lorenz stochastic resonance system based on particle swarm
Technical Field
The invention belongs to the field of intelligent motor fault diagnosis, and particularly provides a motor fault diagnosis method of a Lorenz stochastic resonance system based on particle swarms.
Background
The motor is one of the most important achievements in the modern energy conversion industry and is widely applied to the fields of medical treatment and sanitation, transportation, information communication and the like. The motor provides rapid service, comfort and powerful guarantee for human beings, and due to the fact that the service life of the motor is limited, performance is degraded, manufacturing defects and pollution are caused, or the motor is influenced by human factors and external environments in operation, the final performance is degraded and even is invalid, and the life of people can be endangered, and serious economic losses are generated, such as medical science, military operations or transportation. The fault of the motor cannot be completely avoided, but if the fault type can be detected and diagnosed in early stage, and the response can be accurately, quickly and timely made, serious accidents can be effectively avoided, and the stable operation of equipment and a system can be maintained.
The early failure of the motor usually reflects certain characteristic changes in different forms such as current signal changes, and parameters such as current and temperature which can reflect the running state of the motor are collected through the sensor, but the data collected by the sensor not only reflects useful information of the running state of the motor, but also contains error information such as measurement errors, transmission errors and environmental noise, and the analysis of the data through a signal processing technology is particularly important.
Along with the rapid progress of nonlinear dynamics theory, a weak signal detection and processing method based on stochastic resonance is provided. Stochastic resonance is a phenomenon in which weak signals are enhanced with noise in a nonlinear system. The stochastic resonance method has the following remarkable characteristics compared with the current method: the weak signal detection mechanism is different, can detect signals with lower signal to noise ratio, and is fast and applied in real time.
Disclosure of Invention
Based on the background problem, the invention provides a motor fault diagnosis method of a Lorenz-like stochastic resonance system based on a particle swarm, which carries out parameter optimization on the Lorenz-like stochastic resonance system with strong plasticity and multiple adjustable parameters through a particle swarm algorithm, so as to ensure that input fault signals with different characteristics can output optimal signal-to-noise ratio signals under different optimal parameters. And filling the output signal data with a fault diagnosis tag, and then inputting the fault diagnosis tag into the extreme learning machine for fault diagnosis.
The specific technical scheme of the invention is as follows:
a motor fault diagnosis method of a Lorenz stochastic resonance system based on particle swarms, which is characterized by comprising the following steps:
s1, collecting stator current data of a motor under different fault types by using a current sensor as an input signal;
s2, inputting the data of the input signals into a Lorenz-like stochastic resonance system based on particle swarms, and taking the signal-to-noise ratio as an objective function in the parameter range of the Lorenz-like stochastic resonance system
Figure SMS_1
Searching the optimal parameters, and outputting a signal with optimal signal-to-noise ratio under the optimal parameters;
s3, combining the data of the signals with different fault types output in the step S2, and filling different diagnostic labels for different faults to form a data set;
s4, randomly dividing the data set into a training set and a test set, inputting the data of the training set into the extreme learning machine for fault diagnosis and learning, and inputting the data of the test set into the learned extreme learning machine for fault diagnosis;
s5, integrating data processing of the Lorenz stochastic resonance system based on the particle swarm and algorithm diagnosis of an extreme learning machine, and inputting the newly acquired data into the integrated system to obtain a fault type corresponding to the input data.
A motor fault diagnosis method of Lorenz-like stochastic resonance system based on particle swarm is characterized in that the Lorenz-like stochastic resonance system in step S2 is as follows:
Figure SMS_2
（1）
wherein the method comprises the steps of
Figure SMS_3
For system parameters->
Figure SMS_4
Is a state variable +.>
Figure SMS_5
For inputting signals, composed of signals and noise
Figure SMS_6
The composition is formed.
A motor fault diagnosis method of Lorenz-like stochastic resonance system based on particle swarm is characterized in that the parameter range of Lorenz-like stochastic resonance system in step S2
Figure SMS_7
The determining method of (1) comprises the following steps:
when the Lorenz-like stochastic resonance system is at a balance point, there is always
Figure SMS_8
Thus, the system balance curve equation can be obtained as
Figure SMS_9
（2）
The pole of the equilibrium curve is
Figure SMS_10
The method comprises the steps of carrying out a first treatment on the surface of the When the system is at the equilibrium point, there is always +.>
Figure SMS_11
Hold, curve->
Figure SMS_12
And->
Figure SMS_13
The value of the intersection point of the system is the balance point of the system; the occurrence of stochastic resonance manifests itself as transitions of the input signal drive system between steady states; the amplitude of the pole of the balance curve affected by the parameter should therefore be smaller than the amplitude of the input signal, i.e. satisfy +.>
Figure SMS_14
System parameters>
Figure SMS_15
Form a collection
Figure SMS_16
；
The Lorenz-like stochastic resonance system is developed from the Lorenz chaotic system, and only when the Lorenz chaotic system is in a fixed point or periodic state under the proper initial value and parameter, the Lorenz-like stochastic resonance system generates stochastic resonance; therefore, lyapunov index spectrums of chaotic systems under different parameters
Figure SMS_17
After sorting from big to small, the maximum Lyapunov index ++>
Figure SMS_18
Parameter of->
Figure SMS_19
Is +.>
Figure SMS_20
；
Lorenz-like stochastic resonance system parameter range
Figure SMS_21
。
A motor fault diagnosis method of Lorenz-like stochastic resonance system based on particle swarm is characterized in that the Lorenz-like stochastic resonance system based on the particle swarm in step S2 is in a system parameter range
Figure SMS_22
The steps for searching the optimal parameters are as follows: />
S21, initializing system parameters, iteration step length, system initial value, population scale and learning rate, iteration times and maximum iteration times and inertia weight;
s22, calculating the signal-to-noise ratio of the output signal of the Lorenz-like stochastic resonance system, and searching an individual optimal value and a group optimal value;
s23, in parameter range
Figure SMS_23
Updating the particle speed and position in the range of the inner and iteration step sizes;
s24, calculating the signal-to-noise ratio of the output signal of the Lorenz-like stochastic resonance system, and updating the individual optimal value and the group optimal value;
and S25, when the termination condition is not met, repeating the steps S23 and S24, and when the termination condition is met, outputting signal data with the optimal signal-to-noise ratio under the optimal parameters.
Compared with the prior art, the invention has the beneficial effects that:
the Lorenz stochastic resonance system of the invention has the advantages of multiple adjustable parameters and strong plasticity, and provides an adjustment space for adaptively searching optimal parameters for different types of input signals.
Based on the particle swarm, the optimal parameters of the Lorenz stochastic resonance system are adaptively found for the data of motor current signals with different fault types, the output signal data with optimal signal-to-noise ratio can highlight different fault characteristics, and the diagnosis effect is good when the processed data are used for fault diagnosis.
Drawings
FIG. 1 is a flow chart of a motor fault diagnosis method of a Lorenz stochastic resonance system based on particle swarms;
FIG. 2 is a time domain plot of 1 set of data in different types of motor fault data collected by a current sensor;
FIG. 3 is a time domain diagram of fault signal data output under optimal parameters of a Lorenz-like stochastic resonance system based on particle swarms;
FIG. 4 is a graph of the diagnostic results of the extreme learning machine after the fault signal is processed by the Lorenz stochastic resonance system based on particle swarms;
FIG. 5 is a graph of the results of an extreme learning machine diagnosis with fault signals not processed by the particle swarm-based Lorenz stochastic resonance system;
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more clear, the technical solutions of the present invention are further described below with reference to the accompanying drawings and practical experiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Fig. 1 is a flow chart of a motor fault diagnosis method of a Lorenz stochastic resonance system based on particle swarm, which comprises the following steps:
s1, collecting stator current data of a motor under different fault types by using a current sensor as an input signal;
s2, inputting the data of the input signals into a Lorenz-like stochastic resonance system based on particle swarms, and taking the signal-to-noise ratio as an objective function in the parameter range of the Lorenz-like stochastic resonance system
Figure SMS_24
Searching the optimal parameters, and outputting a signal with optimal signal-to-noise ratio under the optimal parameters;
s3, combining the data of the signals with different fault types output in the step S2, and filling different diagnostic labels for different faults to form a data set;
s4, randomly dividing the data set into a training set and a test set, inputting the data of the training set into the extreme learning machine for fault diagnosis and learning, and inputting the data of the test set into the learned extreme learning machine for fault diagnosis;
s5, integrating data processing of the Lorenz stochastic resonance system based on the particle swarm and algorithm diagnosis of an extreme learning machine, and inputting the newly acquired data into the integrated system to obtain a fault type corresponding to the input data.
Further, in step S1, data is collected through an asynchronous motor experimental platform, and a specific experimental motor is a YSP90L-4 type asynchronous motor, and main parameters thereof are as follows: rated workThe rate is 1.5kW, the rated rotation speed is 1400r/min, the rated voltage is 380V, the polar logarithm is 2, the rated frequency is 50Hz, the wiring mode Y, the power factor is 0.79, and the ambient temperature is 40 o C。
Further, in step S1, different fault types are simulated by continuously replacing the motor to simulate different fault motors, so as to complete the current collection experiment under various faults. The fault types collected by the experiment include: stator winding faults (Stator winding fault, SWF), air gap eccentricity (Air gap eccentricity, AGE), no faults (NORMAL), bearing Faults (BF), shaft Bending (SB), rotor bar (RBB); the collected current data is extracted for 500 groups of 2000 continuous sampling points under each type of faults.
Further, the time domain diagram of 1 group of 2000 continuous sampling point data in the motor fault data of different types in step S1 is shown in fig. 2, fig. 2 (a) is that of 1 group of 2000 continuous sampling point data in the stator winding fault, fig. 2 (b) is that of 1 group of 2000 continuous sampling point data in the air gap eccentric fault, fig. 2 (c) is that of 1 group of 2000 continuous sampling point data in the no fault, fig. 2 (d) is that of 1 group of 2000 continuous sampling point data in the bearing fault, fig. 2 (e) is that of 1 group of 2000 continuous sampling point data in the rotating shaft bending fault, and fig. 2 (f) is that of 1 group of 2000 continuous sampling point data in the rotor breaking fault.
Further, in the Lorenz-like stochastic resonance system based on the particle swarm in step S2, the Lorenz-like stochastic resonance system is:
Figure SMS_25
（3）
wherein the method comprises the steps of
Figure SMS_26
For system parameters->
Figure SMS_27
Is a state variable +.>
Figure SMS_28
For inputting signals, by signalsAnd noise
Figure SMS_29
The composition is formed.
Further, in step S2, the objective function is determined by the system output signal-to-noise ratio, specifically:
Figure SMS_30
（4）
wherein, the liquid crystal display device comprises a liquid crystal display device,
Figure SMS_31
for the data length of the input signal, < > for>
Figure SMS_32
Is the single-side spectral amplitude of the input signal corresponding to the characteristic frequency.
Further, in step S2, lorenz-like stochastic resonance system parameter range
Figure SMS_33
The determining method of (1) comprises the following steps:
when the Lorenz-like stochastic resonance system is at a balance point, there is always
Figure SMS_34
Thus, the system balance curve equation can be obtained as
Figure SMS_35
（3）/>
The pole of the equilibrium curve is
Figure SMS_36
The method comprises the steps of carrying out a first treatment on the surface of the When the system is at the equilibrium point, there is always +.>
Figure SMS_37
Hold, curve->
Figure SMS_38
And->
Figure SMS_39
The value of the intersection point of the system is the balance point of the system; generation of stochastic resonanceThe input signal driving system is shown as transitioning between steady states; the amplitude of the pole of the balance curve affected by the parameter should therefore be smaller than the amplitude of the input signal, i.e. satisfy +.>
Figure SMS_40
System parameters>
Figure SMS_41
Form a collection
Figure SMS_42
；
The Lorenz-like stochastic resonance system is developed from the Lorenz chaotic system, and only when the Lorenz chaotic system is in a fixed point or periodic state under the proper initial value and parameter, the Lorenz-like stochastic resonance system generates stochastic resonance; therefore, lyapunov index spectrums of chaotic systems under different parameters
Figure SMS_43
After sorting from big to small, the maximum Lyapunov index ++>
Figure SMS_44
Parameter of->
Figure SMS_45
Is +.>
Figure SMS_46
；
Lorenz-like stochastic resonance system parameter range
Figure SMS_47
。
Further, the Lorenz stochastic resonance system parameter settings based on the particle swarm in step S2 are shown in the following table:
Figure SMS_48
further, in step S2, the particle swarm-based Lorenz-like stochastic resonance system is within the system parameter range
Figure SMS_49
The steps for searching the optimal parameters are as follows:
s21, initializing system parameters, iteration step length, system initial value, population scale and learning rate, iteration times and maximum iteration times and inertia weight;
s22, calculating the signal-to-noise ratio of the output signal of the Lorenz-like stochastic resonance system, and searching an individual optimal value and a group optimal value;
s23, in parameter range
Figure SMS_50
Updating the particle speed and position in the range of the inner and iteration step sizes;
s24, calculating the signal-to-noise ratio of the output signal of the Lorenz-like stochastic resonance system, and updating the individual optimal value and the group optimal value;
and S25, when the termination condition is not met, repeating the steps S23 and S24, and when the termination condition is met, outputting signal data with the optimal signal-to-noise ratio under the optimal parameters.
Further, after the stator current data collected by the current sensor in step S1 is processed in step S2, 1 group of different fault type data with optimal signal-to-noise ratio is output, a time domain diagram of 2000 continuous sampling point data is shown in fig. 3, fig. 3 (a) is a diagram of 2000 continuous sampling point data in 1 group of processed stator winding fault data, fig. 3 (b) is a diagram of 2000 continuous sampling point data in 1 group of air gap eccentric fault data, fig. 3 (c) is a diagram of 2000 continuous sampling point data in 1 group of processed non-fault data, fig. 3 (d) is a diagram of 2000 continuous sampling point data in 1 group of processed bearing fault data, fig. 3 (e) is a diagram of 2000 continuous sampling point data in 1 group of processed rotating shaft bending fault data, and fig. 3 (f) is a diagram of 2000 continuous sampling point data in 1 group of processed rotor bar breaking fault data.
Further, in step S3, the data of the signals with different fault types output in step S2 are combined, and different diagnostic tags are filled in the different faults, and the specific tags are shown in the following table:
Figure SMS_51
further, in step S4, the dividing manner of randomly dividing the data set into the training set and the test set is as follows:
80% of samples in the data set are randomly selected as a training set of the algorithm, and 20% of samples are selected as a test set.
Further, in step S4, the diagnosis result diagram of the extreme learning machine after the fault signal is processed by the particle swarm-based Lorenz stochastic resonance system is shown in fig. 4; fig. 4 (a) is a diagram comparing the diagnosis fault type of the 1 st limit learning machine diagnosis experiment with the real fault type in 20 repeated experiments, and fig. 4 (b) is a diagram of the diagnosis situation of each fault type; as can be seen from the graph, the average accuracy of the fault diagnosis of the extreme learning machine after the treatment of the Lorenz stochastic resonance system based on the particle swarm is 100%.
Further, comparing the fault signals in step S4 with the extreme learning machine diagnostic result chart of which the fault signals are not processed by the particle swarm-based Lorenz-like stochastic resonance system is shown in FIG. 5; fig. 5 (a) is a diagram comparing the diagnosis fault type of the 1 st limit learning machine diagnosis experiment with the real fault type in 20 repeated experiments, and fig. 5 (b) is a diagram of the diagnosis situation of each fault type; from the graph, the average accuracy of the extreme learning machine diagnosis without treatment by the Lorenz stochastic resonance system based on the particle swarm is 97.316%.
Further, as can be seen from comparing fig. 4 and fig. 5, the diagnosis result of the extreme learning machine processed by the Lorenz stochastic resonance system based on the particle swarm is better than the diagnosis result of the extreme learning machine processed by the Lorenz stochastic resonance system not based on the particle swarm, so that the fault diagnosis accuracy is improved, and the motor fault diagnosis method of the Lorenz stochastic system based on the particle swarm has good diagnosis effect.

Claims (4)

1. A motor fault diagnosis method of a Lorenz stochastic resonance system based on particle swarms, which is characterized by comprising the following steps:
s1, collecting stator current data of a motor under different fault types by using a current sensor as an input signal;
s2, inputting the data of the input signals into a Lorenz-like stochastic resonance system based on particle swarms, and taking the signal-to-noise ratio as an objective function in the parameter range of the Lorenz-like stochastic resonance system
Figure QLYQS_1
Searching the optimal parameters, and outputting a signal with optimal signal-to-noise ratio under the optimal parameters;
s3, combining the data of the signals with different fault types output in the step S2, and filling different diagnostic labels for different faults to form a data set;
s4, randomly dividing the data set into a training set and a test set, inputting the data of the training set into the extreme learning machine for fault diagnosis and learning, and inputting the data of the test set into the learned extreme learning machine for fault diagnosis;
s5, integrating data processing of the Lorenz stochastic resonance system based on the particle swarm and algorithm diagnosis of an extreme learning machine, and inputting the newly acquired data into the integrated system to obtain a fault type corresponding to the input data.
2. The method for diagnosing motor faults of a particle swarm-based Lorenz-like stochastic resonance system according to claim 1, wherein the Lorenz-like stochastic resonance system in step S2 is:
Figure QLYQS_2
,
wherein the method comprises the steps of
Figure QLYQS_3
For system parameters->
Figure QLYQS_4
Is a state variable +.>
Figure QLYQS_5
Is input signal, composed of signal and noise->
Figure QLYQS_6
The composition is formed.
3. The method for diagnosing motor faults of Lorenz-like stochastic resonance system based on particle swarms according to claim 1, wherein the range of parameters of the Lorenz-like stochastic resonance system in the step S2
Figure QLYQS_7
The determining method of (1) comprises the following steps:
when the Lorenz-like stochastic resonance system is at a balance point, there is always
Figure QLYQS_9
Therefore, the system balance curve equation is +.>
Figure QLYQS_11
The pole of the equilibrium curve is
Figure QLYQS_12
The method comprises the steps of carrying out a first treatment on the surface of the When the system is at the equilibrium point, there is always +.>
Figure QLYQS_13
Hold, curve->
Figure QLYQS_14
And->
Figure QLYQS_15
The value of the intersection point of the system is the balance point of the system; the occurrence of stochastic resonance manifests itself as transitions of the input signal drive system between steady states; the amplitude of the pole of the balance curve affected by the parameter should therefore be smaller than the amplitude of the input signal, i.e. satisfy +.>
Figure QLYQS_16
System parameters>
Figure QLYQS_8
Form a set->
Figure QLYQS_10
；
The Lorenz-like stochastic resonance system is developed from the Lorenz chaotic system, and only when the Lorenz chaotic system is in a fixed point or periodic state under the proper initial value and parameter, the Lorenz-like stochastic resonance system generates stochastic resonance; therefore, lyapunov index spectrums of chaotic systems under different parameters
Figure QLYQS_17
After sorting from big to small, the maximum Lyapunov index ++>
Figure QLYQS_18
Parameter of->
Figure QLYQS_19
Is +.>
Figure QLYQS_20
；
Lorenz-like stochastic resonance system parameter range
Figure QLYQS_21
。
4. The method for diagnosing motor failure in a particle swarm-based Lorenz-like stochastic resonance system according to claim 1, wherein the particle swarm-based Lorenz-like stochastic resonance system in step S2 is within a system parameter range
Figure QLYQS_22
The steps for searching the optimal parameters are as follows:
s21, initializing system parameters, iteration step length, system initial value, population scale and learning rate, iteration times and maximum iteration times and inertia weight;
s22, calculating the signal-to-noise ratio of the output signal of the Lorenz-like stochastic resonance system, and searching an individual optimal value and a group optimal value;
s23, atParameter range
Figure QLYQS_23
Updating the particle speed and position in the range of the inner and iteration step sizes;
s24, calculating the signal-to-noise ratio of the output signal of the Lorenz-like stochastic resonance system, and updating the individual optimal value and the group optimal value;
and S25, when the termination condition is not met, repeating the steps S23 and S24, and when the termination condition is met, outputting signal data with the optimal signal-to-noise ratio under the optimal parameters.
CN202310268032.9A 2023-03-20 2023-03-20 Motor fault diagnosis method of Lorenz stochastic resonance system based on particle swarm Pending CN116186610A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310268032.9A CN116186610A (en) 2023-03-20 2023-03-20 Motor fault diagnosis method of Lorenz stochastic resonance system based on particle swarm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310268032.9A CN116186610A (en) 2023-03-20 2023-03-20 Motor fault diagnosis method of Lorenz stochastic resonance system based on particle swarm

Publications (1)

Publication Number Publication Date
CN116186610A true CN116186610A (en) 2023-05-30

Family

ID=86450652

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310268032.9A Pending CN116186610A (en) 2023-03-20 2023-03-20 Motor fault diagnosis method of Lorenz stochastic resonance system based on particle swarm

Country Status (1)

Country Link
CN (1) CN116186610A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117406048A (en) * 2023-12-15 2024-01-16 国网山西省电力公司太原供电公司 Transformer discharge fault diagnosis method and device

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117406048A (en) * 2023-12-15 2024-01-16 国网山西省电力公司太原供电公司 Transformer discharge fault diagnosis method and device
CN117406048B (en) * 2023-12-15 2024-02-27 国网山西省电力公司太原供电公司 Transformer discharge fault diagnosis method and device

Similar Documents

Publication Publication Date Title
Ni et al. Physics-Informed Residual Network (PIResNet) for rolling element bearing fault diagnostics
Palácios et al. A comprehensive evaluation of intelligent classifiers for fault identification in three-phase induction motors
Ballal et al. Adaptive neural fuzzy inference system for the detection of inter-turn insulation and bearing wear faults in induction motor
CN116186610A (en) Motor fault diagnosis method of Lorenz stochastic resonance system based on particle swarm
CN110766175A (en) Pitch system fault detection method and device based on optimal interval distribution machine
CN113486868B (en) Motor fault diagnosis method and system
CN112200038B (en) CNN-based quick identification method for oscillation type of power system
Medina et al. Deep learning-based gear pitting severity assessment using acoustic emission, vibration and currents signals
US20230163713A1 (en) Motor driving system converter fault diagnosis method based on adaptive sparse filtering
Husari et al. Early stator fault detection and condition identification in induction motor using novel deep network
CN113325314A (en) Motor fault diagnosis method
Hassan et al. Evaluation of electric dynamic performance of an electric vehicle system using different control techniques
Han et al. A remaining useful life prediction approach based on low-frequency current data for bearings in spacecraft
CN108280424A (en) A kind of rolling bearing method for predicting residual useful life based on sparse coding
Wagner et al. Bearing fault detection using deep neural network and weighted ensemble learning for multiple motor phase current sources
Guolian et al. Research on fault diagnosis of wind turbine control system based on artificial neural network
CN112651426A (en) Fault diagnosis method for rolling bearing of wind turbine generator
Wen et al. A novel machine learning model for safety risk analysis in flywheel-battery hybrid energy storage system
Liu et al. Feasibility study of the GST-SVD in extracting the fault feature of rolling bearing under variable conditions
Akbar et al. State-of-the-art techniques for fault diagnosis in electrical machines: advancements and future directions
Han et al. Deep subclass alignment transfer network based on time–frequency features for intelligent fault diagnosis of planetary gearboxes under time-varying speeds
Shu et al. Enhanced feature extraction method for motor fault diagnosis using low-quality vibration data from wireless sensor networks
Ebrahimi et al. Detection of partial demagnetization fault in wind turbine permanent magnet generator using a data-driven method
CN116738338A (en) Small sample fault diagnosis method based on multi-scale integrated LightGBM
Zhao et al. Fault diagnosis of asynchronous induction motor based on BP neural network

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