CN116956215A - Fault diagnosis method and system for transmission system - Google Patents
Fault diagnosis method and system for transmission system Download PDFInfo
- Publication number
- CN116956215A CN116956215A CN202310842038.2A CN202310842038A CN116956215A CN 116956215 A CN116956215 A CN 116956215A CN 202310842038 A CN202310842038 A CN 202310842038A CN 116956215 A CN116956215 A CN 116956215A
- Authority
- CN
- China
- Prior art keywords
- data
- dimensional
- fault diagnosis
- transmission system
- sequence
- 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
- 238000003745 diagnosis Methods 0.000 title claims abstract description 71
- 238000000034 method Methods 0.000 title claims abstract description 56
- 230000005540 biological transmission Effects 0.000 title claims abstract description 45
- 230000004927 fusion Effects 0.000 claims abstract description 38
- 230000001133 acceleration Effects 0.000 claims abstract description 17
- 230000008569 process Effects 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 18
- 238000010606 normalization Methods 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 13
- 238000011176 pooling Methods 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 9
- 238000009826 distribution Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 4
- 230000017105 transposition Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 abstract description 5
- 238000005516 engineering process Methods 0.000 description 9
- 238000005096 rolling process Methods 0.000 description 5
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000036541 health Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 238000005299 abrasion Methods 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/02—Gearings; Transmission mechanisms
- G01M13/028—Acoustic or vibration analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
- G06F18/15—Statistical pre-processing, e.g. techniques for normalisation or restoring missing data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2131—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on a transform domain processing, e.g. wavelet transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06N3/047—Probabilistic or stochastic networks
-
- 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
- G06N3/048—Activation functions
-
- 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
- G06N3/0985—Hyperparameter optimisation; Meta-learning; Learning-to-learn
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- 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
Abstract
The invention discloses a fault diagnosis method and a fault diagnosis system of a transmission system, wherein the fault diagnosis method comprises the steps of acquiring vibration acceleration signal data and current signal data acquired by a sensor in the transmission system, and preprocessing the signal data to obtain a one-dimensional time sequence data sequence; performing short-time Fourier transform on the one-dimensional time sequence data sequence to generate two-dimensional data; and inputting the two-dimensional data into a GAP strategy multi-source data fusion convolutional network model, and diagnosing and identifying faults to finish fault diagnosis in a transmission system. The method fuses the data of the multiple sensors, processes the data from the multiple sensors in multiple levels, multiple aspects and multiple layers, eliminates errors caused by failure of the single type of sensor, increases the data characteristics, and simultaneously effectively reflects the fault characteristics of signals so as to improve the reliability of analysis.
Description
Technical Field
The invention belongs to the technical field of intelligent fault diagnosis, and relates to a fault diagnosis method and system for a transmission system.
Background
Along with the rapid progress of information technology and industrial generation technology in China, modern industrial equipment gradually realizes enlargement, complexity, high speed, integration and intellectualization, and has higher requirements on equipment stability, safety and the like. In order to prevent the problem that the antenna cannot be effectively observed due to shutdown maintenance caused by failure reasons such as aging, mechanical faults and the like of mechanical equipment, electrical equipment and a transmission system, a proper fault diagnosis method needs to be researched to rapidly locate a fault occurrence place, and timely maintain the fault occurrence place, so that great economic loss is avoided. The fault diagnosis is one of key technologies for maintaining safe and reliable operation of equipment, and aims to quickly and accurately judge fault types and the like and reduce operation and maintenance cost of the equipment. The intelligent fault diagnosis technology is a modern mainstream technology, can automatically learn the fault mode from historical big data, and has the advantages of simple modeling, high accuracy, multiple application scenes and the like. The advanced learning technology in recent years has stronger feature extraction capability and big data processing capability, and compared with the classical fault diagnosis technology, the advanced learning technology can realize end-to-end intelligent fault diagnosis in more complex scenes. The related data show that after fault diagnosis is implemented in the production and processing process of the equipment, the accident rate is reduced by 75%, and the maintenance cost is reduced by 25% -50%.
The development of equipment health monitoring and fault diagnosis is an important topic of research in the industry and academia until now, and is a multi-disciplinary crossing technology, and a plurality of methods are proposed by related scholars. For example, yang et al dynamically model the dc motor, design a kalman filter to recursively estimate the motor speed, and implement fault diagnosis of the dc motor. The model-based diagnosis method has high accuracy requirement on model modeling, and when the established mathematical model estimation result and the actual model measurement result have larger errors, the fault diagnosis method can be misdiagnosed or missed diagnosis. The method is low in efficiency, has high requirements on technical experience of personnel, and is not beneficial to popularization of the diagnosis method. Zhang et al establish a network model of CNN variants, and can effectively improve the accuracy of the fault diagnosis result of the rolling bearing aiming at the noise environment and the fault form of the rolling bearing under nonlinear and non-steady loads. Such methods, while much more efficient than conventional methods, are single in data acquisition, low in information content, and less reliable due to large errors caused by possible sensor failures.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a transmission system fault diagnosis method and a transmission system fault diagnosis system, thereby solving the technical problems of single data acquisition, less information quantity and lower reliability possibly caused by large fault analysis errors due to sensor failure in the prior art.
The invention is realized by the following technical scheme:
a transmission system fault diagnosis method comprising the steps of:
vibration acceleration signal data and current signal data acquired by a sensor in a transmission system are acquired, and the signal data are preprocessed to obtain a one-dimensional time sequence data sequence; the preprocessing comprises normalization processing of data;
performing short-time Fourier transform on the one-dimensional time sequence data sequence to generate two-dimensional data;
and inputting the two-dimensional data into a GAP strategy multi-source data fusion convolutional network model, and diagnosing and identifying faults to finish fault diagnosis in the transmission system.
Preferably, the normalization process specifically includes:
s101: calculating a data mean value:
s102: calculating the data variance:
s103: normalizing the processed sample data values:
s104: sample sequence after normalization processing:
X' n ＝[x′ 1 ,x' 2 ,…,x' n ] T
wherein:mean value of sample data; n is the number of sample data; x is x i Is a sample value; s is the sample data variance; x's' i The normalized sample value; x'. n For normalizing the obtained one-dimensional time sequence data sequence, T is a transposition process.
Preferably, the short-time fourier transform process specifically includes:
wherein h (t) is a one-dimensional time sequence X' n Regarding one-dimensional time series data of time t, f (t) is a time window function; t is time; j is an imaginary unit, ω is frequency, and e is a natural constant.
Preferably, the time window function is specifically:
wherein: f (t) is a time window function; t is time and M is window width.
Preferably, the GAP policy multi-source data fusion convolutional network model comprises three rolling and pooling layers, three fusion layers, one pooling layer, one global average pooling layer and one softmax layer.
Preferably, the construction process of the GAP policy multi-source data fusion convolutional network model comprises the following steps:
s301: vibration acceleration signal data and current signal data acquired by different sensors in a transmission system and actual fault types corresponding to the different signal data are acquired;
s302: preprocessing the vibration acceleration signal data and the current signal data acquired by different sensors in the acquired transmission system, wherein the preprocessing comprises normalization processing and data dividing processing in sequence, and acquiring a one-dimensional time sequence training data set and a one-dimensional time sequence testing data set;
s303: performing short-time Fourier transform on the one-dimensional time sequence training data set and the one-dimensional time sequence testing data set to generate two-dimensional data, and obtaining a two-dimensional data training sample and a two-dimensional data testing sample;
s304: inputting a two-dimensional data training sample into the GAP strategy multi-source data fusion convolutional network model, obtaining a predicted fault diagnosis result, comparing the predicted fault diagnosis result with an actual fault type, outputting an error value, and if the error value is a constant value, re-verifying the model by using a two-dimensional data test sample to complete the construction of the GAP strategy multi-source data fusion convolutional network model;
s305: if the error value continues to attenuate, updating the weight parameter of the network model, and repeating S304 until the error value meets the requirement, and completing the construction of the GAP strategy multi-source data fusion convolution network model.
Preferably, the cross entropy loss function of the GAP policy multi-source data fusion convolutional network model is as follows:
wherein y is i Representing the true probability distribution of the i-th class of samples,is the distribution of the network outputs of the i-th class of samples, n being the total class number.
A driveline fault diagnostic system comprising:
and a data acquisition module: the data acquisition module is used for acquiring vibration acceleration signal data and current signal data acquired by a sensor in the transmission system, and preprocessing the signal data to obtain a one-dimensional time sequence data sequence;
and a data processing module: the data processing module is used for performing short-time Fourier transform on the one-dimensional time sequence data to generate two-dimensional data;
and a result output module: the result output module is used for inputting the two-dimensional data into a GAP strategy multi-source data fusion convolution network model, diagnosing and identifying faults, completing fault diagnosis in a transmission system and outputting diagnosis results.
A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the above method when the computer program is executed.
A computer readable storage medium storing a computer program which, when executed by a processor, performs the steps of the method described above.
Compared with the prior art, the invention has the following beneficial technical effects:
a fault diagnosis method for a transmission system includes the steps of collecting vibration acceleration signals and current signals collected by sensors in the transmission system, performing data normalization pretreatment, converting the vibration acceleration signals and the current signals into two-dimensional data images through STFT (short time Fourier transform), inputting the two-dimensional data images into a trained GAP strategy multi-source data fusion convolution network model, and finally outputting a diagnosis device state result to achieve end-to-end intelligent diagnosis. The model fuses the data of multiple sensors, processes the data from multiple sensors in multiple levels, multiple aspects and multiple layers, effectively eliminates errors caused by failure of a single type of sensor, and simultaneously preprocesses the data to obtain a one-dimensional time sequence data, and performs short-time Fourier transform on the one-dimensional time sequence data sequence to generate two-dimensional data, so that the data characteristics are increased, meanwhile, the fault characteristics of signals are effectively reflected, and the reliability of analysis is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a transmission system fault diagnosis method according to the present invention;
fig. 2 is a schematic structural diagram of a transmission system fault diagnosis system according to the present invention.
FIG. 3 is a flow chart of the training and fault diagnosis process for the model in the embodiment 2 of the present invention;
FIG. 4 is a STFT time-frequency conversion result;
FIG. 5 is a schematic diagram of a GAP policy multi-source data fusion convolutional network model;
fig. 6 is a schematic diagram of GAP operation.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
In the description of the embodiments of the present invention, it should be noted that, if the terms "upper," "lower," "horizontal," "inner," and the like indicate an azimuth or a positional relationship based on the azimuth or the positional relationship shown in the drawings, or the azimuth or the positional relationship in which the inventive product is conventionally put in use, it is merely for convenience of describing the present invention and simplifying the description, and does not indicate or imply that the apparatus or element to be referred to must have a specific azimuth, be configured and operated in a specific azimuth, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and the like, are used merely to distinguish between descriptions and should not be construed as indicating or implying relative importance.
Furthermore, the term "horizontal" if present does not mean that the component is required to be absolutely horizontal, but may be slightly inclined. As "horizontal" merely means that its direction is more horizontal than "vertical", and does not mean that the structure must be perfectly horizontal, but may be slightly inclined.
In the description of the embodiments of the present invention, it should also be noted that, unless explicitly specified and limited otherwise, the terms "disposed," "mounted," "connected," and "connected" should be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
The invention is described in further detail below with reference to the attached drawing figures:
example 1
As shown in fig. 1, a transmission system fault diagnosis method includes the steps of:
s1: vibration acceleration signal data and current signal data acquired by a sensor in a transmission system are acquired, and the signal data are preprocessed to obtain a one-dimensional time sequence data sequence;
the preprocessing comprises the step of sequentially carrying out normalization processing on the data to obtain the one-dimensional time sequence data sequence.
The normalization process comprises the following steps:
s101: calculating a data mean value:
s102: calculating the data variance:
s103: normalizing the processed sample data values:
s104: sample sequence after normalization processing:
X' n ＝[x′ 1 ,x' 2 ,…,x' n ] T
wherein:mean value of sample data; n is the number of sample data; x is x i Is a sample value; s is the sample data variance; x's' i The normalized sample value; x'. n For normalizing the obtained one-dimensional time sequence data sequence, T is a transposition process.
S2: performing short-time Fourier transform on the one-dimensional time sequence data sequence to generate two-dimensional data;
when short-time Fourier transform is performed, the method specifically comprises the following steps:
wherein h (t) is a one-dimensional time sequence X' n Regarding one-dimensional time series data of time t, f (t) is a time window function; t is time; j is an imaginary unit, ω is frequency, and e is a natural constant.
The time window function f (t) is specifically:
wherein: f (t) is a time window function; t is time, M is window width, and the window width is selected to be 2 times frequency.
S3: and inputting the two-dimensional data into a GAP strategy multi-source data fusion convolutional network model, and diagnosing and identifying faults to finish fault diagnosis in a transmission system.
The GAP strategy multi-source data fusion convolutional network model comprises three rolling and pooling layers, three fusion layers, one pooling layer, one global average pooling layer and one softmax layer.
The cross entropy loss function of the GAP strategy multi-source data fusion convolutional network model is as follows:
wherein y is i Representing the true probability distribution of the i-th class of samples,is the distribution of the network outputs of the i-th class of samples, n being the total class number.
The construction process of the GAP strategy multi-source data fusion convolutional network model comprises the following steps:
s301: vibration acceleration signal data and current signal data acquired by different sensors in a transmission system and actual fault types corresponding to the different signal data are acquired;
s302: preprocessing the vibration acceleration signal data and the current signal data acquired by different sensors in the acquired transmission system, wherein the preprocessing comprises normalization processing and data dividing processing in sequence, and a one-dimensional time sequence data sequence is acquired; the data dividing process is to divide the acquired data into a training data set and a test data set;
s303: performing short-time Fourier transform on the one-dimensional time sequence data to generate two-dimensional data, and obtaining a two-dimensional data training sample and a two-dimensional data testing sample;
s304: inputting a two-dimensional data training sample into the GAP strategy multi-source data fusion convolutional network model, obtaining a predicted fault diagnosis result, comparing the predicted fault diagnosis result with an actual fault type, outputting an error value, and if the error value is a constant value, re-verifying the model by using a two-dimensional data test sample to complete the construction of the GAP strategy multi-source data fusion convolutional network model;
s305: if the error value continues to attenuate, updating the weight parameter of the network model, and repeating S303-S304 until the error value meets the requirement, wherein updating the weight parameter of the network model mainly comprises increasing the iteration times, and completing the construction of the GAP strategy multi-source data fusion convolution network model.
The invention discloses a transmission system fault diagnosis method, which utilizes CNN deep learning theory to fuse each sensor information, associates high-level characteristics of a multisource sensor, and fully utilizes the multisensor information to obtain better diagnosis results; on one hand, the defect that a single sensor and a single measuring point are insufficient for representing the fault state of a transmission system and the fault recognition accuracy is low can be overcome; on the other hand, the defects that the measured data quantity of the sensor is increased in proportion, and the current fault diagnosis is carried out by using a fault tree and a fault dictionary diagnosis method, and the information of each sensor forms an island and is not associated with the island, so that the fault recognition efficiency is lower can be overcome. Meanwhile, GAP is adopted to carry out pooling dimension reduction collection on the multi-source data fusion characteristics to form a one-dimensional vector, then the one-dimensional vector is input into a classifier to classify, the number of network parameters is reduced, the problems of over-fitting, weak network generalization capability and the like of a network are prevented, the model is simple, the calculation speed is high, and good guarantee can be accurately provided for fault diagnosis. The invention is an accurate and rapid intelligent fault diagnosis method, can be applied to the variety of fault types, and can be widely applied to the fields of antennas, aerospace, mechanical equipment, electricity and the like.
Further, as shown in fig. 2, the present invention also discloses a system for diagnosing a transmission system fault, including:
and a data acquisition module: the data acquisition module is used for acquiring vibration acceleration signal data and current signal data acquired by a sensor in the transmission system, and preprocessing the signal data to obtain a one-dimensional time sequence data sequence;
and a data processing module: the data processing module is used for performing short-time Fourier transform on the one-dimensional time sequence data to generate two-dimensional data;
and a result output module: the result output module is used for inputting the two-dimensional data into a GAP strategy multi-source data fusion convolution network model, diagnosing and identifying faults, completing fault diagnosis in a transmission system and outputting diagnosis results.
Example 2
For further explanation of the invention, reference is made to fig. 3.
A transmission system fault diagnosis method comprises the following specific steps:
step 1, vibration acceleration signals of a bearing and a gear in a transmission system are acquired through acceleration data, motor current signals are acquired through a driving circuit current state sensor, data normalization processing is carried out on various signals to ensure the same distribution of data, and then the data are divided to increase training samples; table 1 shows the ratio and number of the total number of samples occupied by the various status data. The proportion of the training set and the test set is 80% and 20% respectively.
The corresponding pretreatment steps are as follows:
(1) Data normalized to zero mean, unit variance:
calculating a data mean value:
calculating the data variance:
normalizing the processed sample data values:
sample sequence after normalization processing:
X' n ＝[x′ 1 ,x' 2 ,…,x' n ] T
wherein:mean value of sample data; n is the number of sample data; x is x i Is a sample value; s is the sample data variance; x's' i The normalized sample value; x'. n For normalizing the obtained one-dimensional time sequence data sequence, T is a transposition process.
Table 1 total ratio of the data collected and the different data in the total data
Step 2, the divided data set is converted into two-dimensional data through short-time Fourier transform (STFT), fault characteristics of the signals cannot be completely expressed by directly using one-dimensional time domain signals for fault diagnosis, the characteristics extracted from the data by using a deep learning model are not enough outstanding, the fault characteristics obtained by time-frequency domain analysis are more obvious, and the analysis based on the time-frequency domain is more comprehensive and accurate, so that the one-dimensional data is converted into a time-frequency domain image by using the STFT, and the STFT conversion result is as shown in fig. 4.
Expression of STFT function:
wherein h (t) is a one-dimensional time sequence X' n Regarding one-dimensional time series data of time t, f (t) is a time window function; t is time; j is an imaginary unit, ω is frequency, and e is a natural constant.
STFT is the STFT transform of the signal within a very short window. It is necessary to select an appropriate window function and width to enhance the time-frequency domain characteristics of the data. Using a Hamming window function, the function is as follows
Wherein: f (t) is a time window function; t is time, M is window width, and the window width is selected to be 2 times frequency.
And 3, establishing a GAP strategy multi-source data fusion convolutional network model by using a python Tensorflow2.3 library, training the model by using training samples until the model converges and the performance meets the requirement, and obtaining the GAP strategy multi-source data fusion convolutional network model. Wherein the network model structure is shown in fig. 5, the network parameters are as follows in table 2:
table 2 network structure and parameters
The model structure is provided with a plurality of inputs, firstly, one-dimensional time sequence data is intercepted by window segments to be subjected to STFT conversion to generate two-dimensional data, each time-frequency image input is subjected to 3 layers of rolling and pooling operations to learn depth characteristics of each sensor, then, depth characteristic images of each data are spliced to obtain fused depth characteristic images, then, complementary association characteristics among the sensors are learned through 3 layers of convolution operation, and then, GAP (Global Average Pooling ) operations are performed, the GAP working principle is as shown in fig. 6, mapping relation is generated between the last layer of characteristic images of the fusion convolution force and categories, and the compressed characteristic images are directly input into softmax layers for classification, so that diagnosis and identification of different faults are realized.
And 4, inputting the divided test data into a trained network to realize diagnosis and identification of different faults.
From the experiment, the invention can be used for carrying out fault diagnosis on the multisource sensor information fusion neural network of the transmission system. The three input signals (two vibration and one current signal) are arbitrarily combined, and three layers of convolutional neural networks and GAP are directly adopted for fault classification for single sensor input. Training the depth network for 10 times by using the training set to obtain fault recognition rate, wherein k is as follows 1 、k 2 、k 3 Respectively representing a gear vibration signal, a bearing vibration signal and a current signal; f (F) 1 、F 2 、F 3 、F 4 、F 5 、F 6 、F 7 Respectively representing the health state, bearing inner ring failure, bearing outer ring failure, gear tooth breakage, tooth root crack, serious abrasion of the gear and abrupt load change. It can be seen from table 3 that the failure recognition rate of the three sensors fused together reaches 95.71%, and the recognition rate of the three sensors for 7 different health states is 92% or more. From the comparison result, the multi-sensor decision level data fusion depth fault diagnosis model established by the invention has high diagnosis precision and fusion performance, and can provide guidance significance for a multi-sensor fusion state monitoring system of a large-scale reflecting surface antenna transmission system.
Table 3 different input values of the model and corresponding output fault diagnosis type and recognition rate
In summary, the driving system fault diagnosis method of the invention provides a more accurate and rapid intelligent diagnosis method, inputs the image data after STFT conversion into a network model, rapidly outputs a diagnosis result, and provides an end-to-end fault diagnosis solution, which can be widely applied in the fields of machine manufacturing, antennas, aerospace, electric power and the like.
The embodiment of the invention provides a schematic diagram of terminal equipment. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. The steps of the various method embodiments described above are implemented when the processor executes the computer program. Alternatively, the processor may implement the functions of the modules/units in the above-described device embodiments when executing the computer program.
The computer program may be divided into one or more modules/units, which are stored in the memory and executed by the processor to accomplish the present invention.
The terminal equipment can be computing equipment such as a desktop computer, a notebook computer, a palm computer, a cloud server and the like. The terminal device may include, but is not limited to, a processor, a memory.
The processor may be a central processing unit (CentralProcessingUnit, CPU), but may also be other general purpose processors, digital signal processors (DigitalSignalProcessor, DSP), application specific integrated circuits (ApplicationSpecificIntegratedCircuit, ASIC), off-the-shelf programmable gate arrays (Field-ProgrammableGateArray, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like.
The memory may be used to store the computer program and/or module, and the processor may implement various functions of the terminal device by running or executing the computer program and/or module stored in the memory and invoking data stored in the memory.
The modules/units integrated in the terminal device may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the present invention may implement all or part of the flow in the above-described example methods, or may be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, where the computer program may implement the steps of the above-described respective embodiment methods when executed by a processor. Wherein the computer program comprises computer program code which may be in source code form, object code form, executable file or some intermediate form etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a Read-only memory (ROM), a random access memory (RAM, randomAccessMemory), an electrical carrier signal, a telecommunication signal, a software distribution medium, and so forth. It should be noted that the computer readable medium contains content that can be appropriately scaled according to the requirements of jurisdictions in which such content is subject to legislation and patent practice, such as in certain jurisdictions in which such content is subject to legislation and patent practice, the computer readable medium does not include electrical carrier signals and telecommunication signals.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A method of transmission fault diagnosis, comprising the steps of:
vibration acceleration signal data and current signal data acquired by a sensor in a transmission system are acquired, and the signal data are preprocessed to obtain a one-dimensional time sequence data sequence; the preprocessing comprises normalization processing of data;
performing short-time Fourier transform on the one-dimensional time sequence data sequence to generate two-dimensional data;
and inputting the two-dimensional data into a GAP strategy multi-source data fusion convolutional network model, and diagnosing and identifying faults to finish fault diagnosis in the transmission system.
2. The transmission system fault diagnosis method according to claim 1, wherein the normalization process specifically comprises:
s101: calculating a data mean value:
s102: calculating the data variance:
s103: normalizing the processed sample data values:
s104: sample sequence after normalization processing:
X′ n ＝[x′ 1 ,x′ 2 ,…,x′ n ] T
wherein:mean value of sample data; n is the number of sample data; x is x i Is a sample value; s is the sample data variance; x's' i The normalized sample value; x'. n For normalizing the obtained one-dimensional time sequence data sequence, T is a transposition process.
3. A transmission system failure diagnosis method according to claim 1, wherein the short-time fourier transform process is specifically:
wherein h (t) is a one-dimensional time sequence X' n Regarding one-dimensional time series data of time t, f (t) is a time window function; t is time; j is the unit of an imaginary number,ω is frequency and e is a natural constant.
4. A method of diagnosing a driveline fault as claimed in claim 3, wherein the time window function is embodied as:
wherein: f (t) is a time window function; t is time and M is window width.
5. The driveline fault diagnosis method of claim 1, wherein the GAP policy multi-source data fusion convolutional network model comprises three convolution and pooling layers, three fusion layers, one pooling layer, one global average pooling layer, and one softmax layer.
6. The transmission system fault diagnosis method according to claim 1, wherein the construction process of the GAP policy multi-source data fusion convolutional network model comprises the following steps:
s301: vibration acceleration signal data and current signal data acquired by different sensors in a transmission system and actual fault types corresponding to the different signal data are acquired;
s302: preprocessing vibration acceleration signal data and current signal data acquired by different sensors in an acquired transmission system, wherein the preprocessing comprises normalization processing and data dividing processing in sequence, and a one-dimensional time sequence training data set and a one-dimensional time sequence testing data set are acquired;
s303: performing short-time Fourier transform on the one-dimensional time sequence training data set and the one-dimensional time sequence testing data set to generate two-dimensional data, and obtaining a two-dimensional data training sample and a two-dimensional data testing sample;
s304: inputting a two-dimensional data training sample into the GAP strategy multi-source data fusion convolutional network model, obtaining a predicted fault diagnosis result, comparing the predicted fault diagnosis result with an actual fault type, outputting an error value, and if the error value is constant, using a two-dimensional data test sample to verify the model again to complete the construction of the GAP strategy multi-source data fusion convolutional network model;
s305: if the error value continues to attenuate, updating the weight parameter of the network model, and repeating S304 until the error value meets the requirement, and completing the construction of the GAP strategy multi-source data fusion convolution network model.
7. The driveline fault diagnosis method of claim 1, wherein the GAP policy multi-source data fusion convolutional network model cross entropy loss function is:
wherein y is i Representing the true probability distribution of the i-th class of samples,is the distribution of the network outputs of the i-th class of samples, n being the total class number.
8. A transmission system fault diagnosis system, comprising:
and a data acquisition module: the data acquisition module is used for acquiring vibration acceleration signal data and current signal data acquired by a sensor in the transmission system, and preprocessing the signal data to obtain a one-dimensional time sequence data sequence;
and a data processing module: the data processing module is used for performing short-time Fourier transform on the one-dimensional time sequence data to generate two-dimensional data;
and a result output module: the result output module is used for inputting the two-dimensional data into a GAP strategy multi-source data fusion convolution network model, diagnosing and identifying faults, completing fault diagnosis in a transmission system and outputting diagnosis results.
9. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1-7 when the computer program is executed.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310842038.2A CN116956215A (en) | 2023-07-10 | 2023-07-10 | Fault diagnosis method and system for transmission system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310842038.2A CN116956215A (en) | 2023-07-10 | 2023-07-10 | Fault diagnosis method and system for transmission system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116956215A true CN116956215A (en) | 2023-10-27 |
Family
ID=88461243
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310842038.2A Pending CN116956215A (en) | 2023-07-10 | 2023-07-10 | Fault diagnosis method and system for transmission system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116956215A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117515131A (en) * | 2024-01-04 | 2024-02-06 | 之江实验室 | Method, device, storage medium and equipment for monitoring abrasion of planetary reducer |
| CN117723782A (en) * | 2024-02-07 | 2024-03-19 | 山东大学 | Sensor fault identification positioning method and system for bridge structure health monitoring |
| CN117723782B (en) * | 2024-02-07 | 2024-05-03 | 山东大学 | Sensor fault identification positioning method and system for bridge structure health monitoring |
-
2023
- 2023-07-10 CN CN202310842038.2A patent/CN116956215A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117515131A (en) * | 2024-01-04 | 2024-02-06 | 之江实验室 | Method, device, storage medium and equipment for monitoring abrasion of planetary reducer |
| CN117723782A (en) * | 2024-02-07 | 2024-03-19 | 山东大学 | Sensor fault identification positioning method and system for bridge structure health monitoring |
| CN117723782B (en) * | 2024-02-07 | 2024-05-03 | 山东大学 | Sensor fault identification positioning method and system for bridge structure health monitoring |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Han et al. | An adaptive spatiotemporal feature learning approach for fault diagnosis in complex systems | |
| CN107941537B (en) | A kind of mechanical equipment health state evaluation method | |
| Zou et al. | Bearing fault diagnosis method based on EEMD and LSTM | |
| CN111830408A (en) | Motor fault diagnosis system and method based on edge calculation and deep learning | |
| CN111680875B (en) | Unmanned aerial vehicle state risk fuzzy comprehensive evaluation method based on probability baseline model | |
| CN113188794B (en) | Gearbox fault diagnosis method and device based on improved PSO-BP neural network | |
| CN116956215A (en) | Fault diagnosis method and system for transmission system | |
| CN114091504A (en) | Rotary machine small sample fault diagnosis method based on generation countermeasure network | |
| CN111273125A (en) | RST-CNN-based power cable channel fault diagnosis method | |
| CN111855202A (en) | Gear box fault diagnosis method and system | |
| Cao et al. | Fault diagnosis of rolling bearing based on multiscale one-dimensional hybrid binary pattern | |
| Du et al. | From anomaly detection to novel fault discrimination for wind turbine gearboxes with a sparse isolation encoding forest | |
| WO2019178930A1 (en) | Fault diagnosis method for mechanical device | |
| Zhi et al. | Anti-noise motor fault diagnosis method based on decision tree and the feature extraction methods in the time domain and frequency domain | |
| Tian et al. | Multi-scale deep coupling convolutional neural network with heterogeneous sensor data for intelligent fault diagnosis | |
| CN116224950A (en) | Intelligent fault diagnosis method and system for self-organizing reconstruction of unmanned production line | |
| Bing et al. | Machinery fault diagnosis method of HV circuit breaker based on EEMD and RBF neural network | |
| Kou et al. | A random forest and current fault texture feature–based method for current sensor fault diagnosis in three-phase pwm vsr | |
| Li et al. | Deep learning based end-to-end rolling bearing fault diagnosis | |
| CN113807431A (en) | Intelligent spindle state evaluation method and system based on multi-source information fusion | |
| Wang et al. | Fault Diagnosis Method of Wind Turbine Gearbox Based on Fusion Multispectrogram and Improved CNN Neural Network | |
| Chu et al. | A relaxed support vector data description algorithm based fault detection in distribution systems | |
| Zhang et al. | CNN4GCDD: A one-dimensional convolutional neural network-based model for gear crack depth diagnosis | |
| Zhang et al. | Fault pattern recognition of rolling bearing based on smoothness prior approach and dual-input depth spatial–temporal fusion | |
| CN115556099B (en) | Sustainable learning industrial robot fault diagnosis system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |