CN112541511A

CN112541511A - Multi-channel time series data fault diagnosis method based on convolutional neural network

Info

Publication number: CN112541511A
Application number: CN201910890283.4A
Authority: CN
Inventors: 宫文峰
Original assignee: Individual
Current assignee: Individual
Priority date: 2019-09-20
Filing date: 2019-09-20
Publication date: 2021-03-23

Abstract

The invention discloses a multichannel time series data fault diagnosis method based on a convolutional neural network, which comprises the following steps of: (1) collecting multi-channel one-dimensional time series fault data of a monitored object; (2) constructing a multichannel one-dimensional time sequence original fault data set, and carrying out normalization and data truncation pretreatment; (3) constructing a multi-channel two-dimensional feature map fault data set; (4) dividing the training set, the verification set and the test set; (5) constructing a multi-channel deep learning fault diagnosis model, which comprises an input layer, a feature extraction layer, a dimensionality reduction and parameter reduction layer, a softmax classification layer and a support vector machine output layer, wherein the dimensionality reduction and parameter reduction layer comprises a transition convolution layer with a convolution kernel of 1 multiplied by 1 and a global mean pooling layer; (6) training, verifying and testing the multi-channel deep learning fault diagnosis model, and finally automatically outputting the diagnosis result by a support vector machine, so that people can more intelligently and conveniently diagnose the faults of the electromechanical equipment.

Description

Multi-channel time series data fault diagnosis method based on convolutional neural network

Technical Field

The invention belongs to the technical field of fault diagnosis, and particularly relates to a multi-channel time series data fault diagnosis method based on a convolutional neural network.

Background

With the arrival of the intelligent era, more and more electromechanical equipment products develop towards the direction of intellectualization, automation, multifunctionalization and precision, the complexity of the application environment of the electromechanical equipment products is gradually increased nowadays, various faults are easy to occur in the long-time continuous operation of the electromechanical equipment under the complicated and changeable working environment, if the faults cannot be diagnosed and eliminated in time, once the fault hazard spreads, the major loss is likely to be brought, and therefore, an effective intelligent fault diagnosis method is extremely necessary to be provided for the electromechanical equipment; with the wide application of the internet plus, the internet of things and the advanced intelligent sensor technology in the electromechanical equipment, big data reflecting the running health state of the electromechanical equipment product is easily acquired and utilized; for example, in machines or components such as an aircraft engine, a motor, an engine, a rolling bearing, a gear and the like, one-dimensional time series monitoring data such as vibration acceleration, noise, rotating speed, temperature, voltage or current can be acquired and obtained, the monitoring data records the health state and change characteristics of the electromechanical equipment in the operation process, and the purpose of fault diagnosis of the electromechanical equipment can be achieved by performing characteristic extraction and statistical analysis on the one-dimensional time series data.

Before the method, the method for fault diagnosis and state monitoring of electromechanical equipment is mainly the traditional modes of 'after repair', 'planned repair' and 'timed maintenance', the method is often quite low in efficiency and has no intelligence, in addition, the traditional maintenance mode of regularly maintaining and regularly replacing parts according to experience and estimating the service life of the parts according to experience is easy to cause waste and misjudgment, and brings potential safety hazards, so that the requirements of technical personnel on intelligent fault diagnosis and online state monitoring cannot be met. Hinton et al propose a deep learning theory, which utilizes a deep Neural Network to greedy learn input sample data layer by layer and automatically extract and replace surface features, has strong feature extraction capability with the potential of identifying tiny fault features, overcomes the inherent defects in the traditional intelligent diagnosis method, and has a Convolutional Neural Network (CNN) which is one of important branches of deep learning and strong feature extraction capability. In recent years, some scholars apply CNN to the field of fault diagnosis, but still need to perform feature extraction pretreatment on original fault data by using a traditional feature extraction method, and cannot fully utilize the strong feature extraction capability of a convolutional neural network, so that the further improvement of the fault diagnosis effect is limited; the problem that the traditional convolutional neural network has too many parameters is solved, the traditional convolutional neural network comprises a 2-3 layers of full-connection layer network structure part which is usually positioned between the last pooling layer and the Softmax classified output layer, the training parameters generated due to the existence of the full-connection layer occupy 80% -90% of the total parameters of the CNN, the defect is that the CNN is greatly offset by the advantage of reducing the parameters through pooling dimension reduction, the structure of the full-connection layer occupies too much calculation resources, meanwhile, the CNN model is easy to train and overfit, particularly, the full-connection layer comprises a plurality of hidden layers, the parameter number of the CNN model increases exponentially along with the increase of the number of the full-connection layers, and therefore, the traditional CNN model is too large in parameter amount, too long in time for fault on-line diagnosis, and not beneficial to real-time rapid fault diagnosis. The Softmax classification function used in the traditional CNN is far less powerful than that of an SVM in multi-classification function, and is difficult to exert superior performance in the aspect of fault intelligent diagnosis.

Disclosure of Invention

Aiming at the defects or the needs to be improved in the prior art, the invention provides a multi-channel time series data fault diagnosis method based on a convolutional neural network, which can be used for intelligently diagnosing faults of electromechanical equipment under various monitoring signals and data acquired by a plurality of sensors, and carrying out more accurate fault diagnosis on a monitored object by effectively utilizing the data of a plurality of channels, so that the whole diagnosis process does not need any manual feature extraction operation, and the defect that the existing fault diagnosis method excessively depends on expert priori knowledge is overcome; firstly, the model structure of the traditional convolutional neural network is improved and designed, the combination of a 1 × 1 transition convolutional layer and a global mean pooling layer is adopted to replace the full-connection layer structure of the traditional CNN, the number of model training parameters and the calculation time are greatly reduced, and secondly, a Support Vector Machine (SVM) is designed in the model test stage to replace a Softmax classification function in the traditional CNN, so that the model diagnosis accuracy is further improved; the invention can automatically carry out fault intelligent diagnosis on the electromechanical equipment, automatically finishes the whole diagnosis process without manual intervention, has better operability and lower use threshold, and enables fault diagnosis technicians to carry out more intelligent, convenient and quick fault diagnosis on the electromechanical equipment.

In order to achieve the above object, the present invention provides a multi-channel time series data fault diagnosis method based on a convolutional neural network, which comprises the following steps:

(1) collecting multi-channel one-dimensional time series fault data of a monitored object; arranging sensors on a monitored object, and acquiring one-dimensional time sequence data monitoring signals generated when the monitored object runs in multiple health states by using the sensors, wherein the number of measuring points arranged on the monitored object by the sensors is set to be T (namely T channels, T is greater than or equal to 1) and used for acquiring fault data of T parts of the monitored object, each measuring point is provided with one sensor, the data acquired by each sensor is a continuous one-dimensional time sequence original data segment, and the sample length of the one-dimensional time sequence original data segment is L, namely the sample length comprises L data points; the health state of the monitoring object is set to be N health states, wherein the N health states comprise a normal state and N-1 fault states, and each health state comprises a one-dimensional time sequence original fault data segment of T channels;

(2) constructing a multi-channel one-dimensional time sequence original fault data set (phi)_{Original source}(ii) a Constructing a multichannel one-dimensional time sequence original fault data set { phi } for deep learning model training and testing by using the monitoring data of the N health states_{Original source}The multi-channel one-dimensional time series original fault data set { phi }_{Original source}Set to include N subsets: { phi }_{Original source}={φ₁,φ₂, ...,φ_i, ...,φ_N}_{Original source}For N health states, each of the subsets [ phi ]_i}_{Original source}Each containing T one-dimensional time-series original data segments, thereby forming an NxTxL multi-dimensional tensor data set { phi }_{Original source}Namely: the length of each one-dimensional time sequence original data segment is L data points, each health state type comprises T one-dimensional time sequence data segments, and a multi-channel original fault data set { phi }_{Original source}N health status types are included;

(3) preprocessing data; acquiring a multichannel one-dimensional time sequence original fault data set (phi)_{Original source}Of (N) x TPerforming data preprocessing operation on the original data segment of the one-dimensional time sequence, wherein the preprocessing comprises normalization and data truncation; firstly, carrying out normalization data processing on the data of the NxT one-dimensional time sequence original data segments one by one, and converting the magnitude of the values of all data points in each one-dimensional time sequence original data segment into 0-1, wherein the normalization method comprises the following steps:X={x _i}=(x _i-x _min)/(x _max-x _min) (ii) a Secondly, segmenting and equally dividing the original data segment of the one-dimensional time sequence of each channel in the T channels after normalization processing of each health state, equally dividing the original data segment of the one-dimensional time sequence of each channel containing L data points into h original small data segments of the one-dimensional time sequence with equal length (assuming that the length of each small data segment with equal length is k data points, k x h = L, and the value range of k is between 100 and 10000), obtaining T x h original small data segments of the one-dimensional time sequence by the T channels, and thus obtaining h original small data segment groups of the one-dimensional time sequence with T channels by each health state, forming a fault sample by each original small data segment group of the one-dimensional time sequence with T channels, namely each fault sample contains T channels, wherein each channel has k data points, the original fault data set { phi } of the multichannel one-dimensional time sequence after the equal division truncation operation is carried out_{Original source}Is marked as { phi }_{Cutting block}And therefore { φ }_{Cutting block}The method comprises the steps that N x h one-dimensional time sequence small data segment groups with T channels are contained, and each one-dimensional time sequence original small data segment group with T channels is marked as a fault sampleX _(k)}^(T)That is, each fault sample contains T channels, each channel containing k data points;

(4) constructing a multi-channel two-dimensional feature map fault data set; (phi) obtained after the equal division truncation operation_{Cutting block}Is last fault sampleX _(k)}^(T)The one-dimensional time sequence data of each channel is further reconstructed into a two-dimensional characteristic diagram in a data format, and the specific data reconstruction method comprises the following steps: first, a fault sample is mappedX _(k)}^(T)In each caseOne-dimensional time series small data segment component with track length kx _(k)Is reconstructed into a matrix form of a two-dimensional feature map in a one-dimensional vector format of (2)x]_m×n(i.e., the size of each feature map is m × n = k), the construction method is set as follows: dividing a data segment containing k data points into m parts, wherein each part contains n data points, and the arrangement sequence is as follows: placing the 1 st n data point on the 1 st row, the 2 nd n data point on the 2 nd row, the 3 rd n data point on the 3 rd row, and sequentially sequencing, … …, wherein the mth n data point is placed on the mth row, thereby obtaining an m multiplied by n two-dimensional characteristic diagram; according to the same method, each fault sampleX _(k)}^(T)T two-dimensional characteristic graphs with the size of m multiplied by n can be obtained through the middle T channels; secondly, the fault sample is mappedX _(k)}^(T)The T two-dimensional feature maps are sequentially superposed according to the sequence of a channel 1, a channel 2, a channel … … and a channel T, and then a fault sample tone containing the two-dimensional feature map of the T channel can be constructedX _[m×n]}^(T)The sample format is a readable sample form for the fault diagnosis of the convolutional neural network designed by the invention in the subsequent step; according to the same sample reconstruction method, the { phi } can be obtained_{Cutting block}N x h one-dimensional time series small data segment group fault samples with T channelX _(k)}^(T)Reconstructed into N x h two-dimensional characteristic graphs with T channels superposed and with size of m x NX _[m×n]}^(T)(ii) a And reconstructs the data format to { phi }_{Cutting block}Recording as a multi-channel two-dimensional feature map fault data set (phi)_2D；

(5) Dividing a data set; a multi-channel two-dimensional feature map fault data set (phi)_2DThe h samples in each health state type are divided into a training set, a verification set and a test set, and the division method comprises the following steps: firstly, a multichannel two-dimensional feature map fault data set { phi }_2DRandomly selecting 30% of samples as a test set from h fault samples in each health state type, and randomly taking out 80% of the rest 70% of samples to be classified into a training set and 20% of samples to be a verification set; and finally, a multi-channel two-dimensional feature map fault data set { phi }_2DAll in the N-class health states in (1) areThe set of fault samples in the training set is the total training set { D }_{Training device}Will { phi }_2DAll the failure sample group set classified as verification set in the N-type health states are set as a total verification set { D }_{Test (experiment)}Will { phi }_2DAll the fault sample groups classified as test sets of the N-type health states of (1) are set as a total test set { D }_Measuring；

(6) Constructing a multi-channel deep learning fault diagnosis model; the multi-channel deep learning fault diagnosis model comprises an input layer, a feature extraction layer, a dimension reduction and parameter reduction layer, a softmax classification layer and a support vector machine output layer, wherein the feature extraction layer comprises a first convolution layer, a first pooling layer, a second convolution layer and a second pooling layer, the dimension reduction and parameter reduction layer comprises a transition convolution layer with a convolution kernel of 1 x 1 and a global mean pooling layer, and the multi-channel deep learning fault diagnosis model is characterized in that: a dimension reduction parameter reduction layer is arranged between the feature extraction layer and the softmax classification layer and is used for replacing a fully-connected network layer part of a traditional convolutional neural network, and the input layer is used for receiving a multi-channel two-dimensional feature map fault data set (phi)_2DTotal training set of (D) }_{Training device}Total verification set { D }_{Test (experiment)}And the total test set { D }_MeasuringThe multi-channel two-dimensional feature map data are input into a first convolution layer in a feature extraction layer, a 1 x 1 transition convolution layer is used for receiving an output feature map of a second pooling layer of the feature extraction layer, the input layer, the first convolution layer, the first pooling layer, the second convolution layer, the second pooling layer, the 1 x 1 transition convolution layer and a global mean pooling layer are sequentially connected in series, a softmax classification layer and a support vector machine output layer are connected in parallel behind the global mean pooling layer, the multi-channel deep learning fault diagnosis model comprises two model combinations of a model training phase model combination and a model testing phase model combination, and the design is as follows: in the training stage of the multi-channel deep learning fault diagnosis model, the softmax classification layer is connected behind the global mean pooling layer and used for training model parameters of each layer of the convolutional neural network, and in the testing stage of the multi-channel deep learning fault diagnosis model, the support vector machine output layer is connected behind the global mean pooling layer and used for further improving and outputting a final diagnosis result.

(7) Training a multi-channel deep learning fault diagnosis model; firstly, adopting the model combination of the training phase of the multi-channel deep learning fault diagnosis model in the step (6), connecting a softmax classification layer behind a global mean pooling layer to execute CNN model parameter training, initializing model parameters, and collecting the total training set { D }_{Training device}Inputting the multichannel two-dimensional feature map data sample into the multichannel deep learning fault diagnosis model for training and learning the deep learning model parameters, repeatedly executing the iterative computation process of forward propagation and backward propagation, and training the model parameters of each layer of the first convolutional layer, the first pooling layer, the second convolutional layer, the second pooling layer, the 1 × 1 transition convolutional layer and the global mean pooling layer;

(8) verifying a multi-channel deep learning fault diagnosis model; while training a multi-channel deep learning fault diagnosis model, a total verification set { D }is obtained_{Test (experiment)}The multi-channel two-dimensional characteristic pattern data is used for verifying the diagnosis accuracy of the multi-channel deep learning fault diagnosis model in the training process in real time, verifying the accuracy of the model on a verification set and checking whether overfitting occurs, and the verification standard is designed as follows: when the accuracy rates of the training set and the verification set increase along with the increase of the number of the training rounds and the accuracy rate of the training set is higher than that of the verification set, the model parameters are normally trained, and the step (7) is continuously executed for training; when the accuracy on the training set and the accuracy on the verification set increase along with the increase of the number of training rounds and the accuracy on the training set begins to be lower than the accuracy on the verification set, which indicates that the model parameter training is overfitting, stopping the model training, skipping to the step (6), revising the hyper-parameters of the multi-channel deep learning fault diagnosis model, and repeating the steps until the multi-channel deep learning fault diagnosis model is not overfitting and the accuracy on the verification set reaches a set target value or the number of iteration rounds;

(9) when the accuracy of the verification set reaches a set target value or training iteration times, ending model training, and simultaneously storing the optimal parameter values of each layer of a first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer, a 1 × 1 transition convolution layer and a global mean value pooling layer in the multi-channel deep learning fault diagnosis model;

(10) testing a multi-channel deep learning fault diagnosis model; firstly, adopting the model combination of the testing stage of the multi-channel deep learning fault diagnosis model in the step (6), connecting a support vector machine output layer after a global mean pooling layer, and secondly, combining the total training set { D }_{Training device}The multichannel two-dimensional feature map data sample is input into a trained multichannel deep learning fault diagnosis model, a feature value is sequentially provided for the input sample by a trained first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer, a 1 x 1 transition convolution layer and a global mean pooling layer, then the output value of the global mean pooling layer is input into a support vector machine, the training of the model parameter of the support vector machine is completed, and the model parameter of the support vector machine is saved; finally, the total test set { D }_MeasuringThe method comprises the steps of inputting sample data into a trained model of a test stage model combination of the multi-channel deep learning fault diagnosis model, sequentially providing features for input samples by a trained first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer, a 1 x 1 transition convolution layer, a global mean value pooling layer and a support vector machine output layer, and finally outputting a final fault diagnosis result by the support vector machine output layer.

The invention designs that the sensor type is set to include one or more combinations of a vibration acceleration sensor, a noise sensor, a pressure sensor, a displacement sensor, a rotating speed sensor, a voltage sensor and a current sensor.

The invention designs that different types and numbers of sensors for monitoring object data acquisition are set to have the same sampling frequency and sampling time.

The invention designs that the monitored objects comprise conventional rotating electromechanical equipment commonly used in the technical field, such as an engine, a motor, a rolling bearing, a gear and the like.

Compared with the prior art, the multi-channel time series data fault diagnosis method based on the convolutional neural network improves the structure of a traditional convolutional neural network model, firstly adopts the combination of the 1 x 1 transitional convolutional layer and the global mean pooling layer to replace the full-connection layer structure of the traditional CNN, effectively reduces the training parameters of the CNN, improves the diagnosis speed of the model, and then adopts a support vector machine to replace a Softmax classifier in the test stage to further improve the diagnosis accuracy.

The invention does not need any manual feature extraction operation and does not need operators to master various complex advanced signal processing technologies, the original fault data can be directly input into the multi-channel deep learning fault diagnosis model designed by the invention, the diagnosis model can automatically carry out data preprocessing, two-dimensional feature map reconstruction, automatic feature extraction and automatic fault classification diagnosis on multi-channel one-dimensional time series fault data acquired by electromechanical equipment, the final diagnosis result is automatically output, the whole diagnosis process is automatically finished without manual intervention, the invention has better operability and lower use threshold, and the fault diagnosis of the electromechanical equipment by fault diagnosis technicians is more intelligent, convenient and rapid.

Drawings

FIG. 1 is a flow chart of a method of multi-channel time series data fault diagnosis based on a convolutional neural network.

Fig. 2 is a schematic structural diagram of the multi-channel deep learning fault diagnosis model of the present invention.

FIG. 3 is a schematic diagram of multi-channel one-dimensional time series fault data of the present invention.

FIG. 4 is a multi-channel one-dimensional time series raw fault data set { φ } for N health states of the present invention_{Original source}Schematic representation of (a).

FIG. 5 is a diagram of a multi-channel one-dimensional time series original data segment being segmented and equally divided according to the present invention.

FIG. 6 is a schematic diagram of a fault sample with a T-channel of the present invention.

FIG. 7 is a schematic diagram of the operation of reconstructing one-dimensional time series data into a two-dimensional feature map according to the present invention.

Fig. 8 is a schematic diagram of a process of reconstructing T one-dimensional time series data segments of a fault sample of a T channel to obtain T two-dimensional feature maps according to the present invention.

FIG. 9 is a schematic diagram of a multi-channel two-dimensional signature graph structure of a fault sample according to the present invention.

Fig. 10 is a schematic diagram of a rolling bearing fault data generation test bed according to a preferred embodiment of the invention.

Detailed Description

For a better understanding of the present invention, reference is made to the following description taken in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

Referring to fig. 1 to 10, a method for diagnosing a fault of a multi-channel time series data based on a convolutional neural network according to a preferred embodiment of the present invention is shown in fig. 1, and includes the following steps:

step (1), collecting multi-channel one-dimensional time series fault data of a monitored object; arranging sensors on a monitored object, and acquiring one-dimensional time sequence data monitoring signals generated when the monitored object runs in multiple health states by using the sensors, wherein the number of measuring points arranged on the monitored object by the sensors is set to be T (namely T channels, T is greater than or equal to 1) and used for acquiring fault data of T parts of the monitored object, each measuring point is provided with one sensor, the data acquired by each sensor is a continuous one-dimensional time sequence original data segment, and the sample length of the one-dimensional time sequence original data segment is L, namely the one-dimensional time sequence original data segment comprises L data points, which is shown in FIG. 3; the health state of the monitoring object is set to be N health states, wherein the N health states comprise a normal state and N-1 fault states, and each health state comprises a one-dimensional time sequence original fault data segment of T channels;

step (2), constructing a multichannel one-dimensional time sequence original fault data set (phi)_{Original source}(ii) a Constructing a multichannel one-dimensional time sequence original fault data set { phi } for deep learning model training and testing by using the monitoring data of the N health states_{Original source}As shown in FIG. 4, the multi-channel one-dimensional time series original fault data set { φ }_{Original source}Set to include N subsets: { phi }_{Original source}={φ₁,φ₂, ...,φ_i, ...,φ_N}_{Original source}For N health states, each of the subsets [ phi ]_i}_{Original source}Each containing T one-dimensional time-series original data segments, thereby forming an NxTxL multi-dimensional tensor data set { phi }_{Original source}Namely: the length of each one-dimensional time sequence original data segment is L data points, each health state type comprises T one-dimensional time sequence data segments, and a multi-channel original fault data set { phi }_{Original source}Contains N health status types (as shown in fig. 4);

step (3), preprocessing data; acquiring a multichannel one-dimensional time sequence original fault data set (phi)_{Original source}Carrying out data preprocessing operation on the N multiplied by T one-dimensional time sequence original data segments, wherein the preprocessing comprises normalization and data truncation; firstly, carrying out normalization data processing on the data of the NxT one-dimensional time sequence original data segments one by one, and converting the magnitude of the values of all data points in each one-dimensional time sequence original data segment into 0-1, wherein the normalization method comprises the following steps:X={x _i}=(x _i-x _min)/(x _max-x _min) (ii) a Secondly, segmenting and equally dividing the one-dimensional time sequence original data segment of each channel in the T channels after normalization processing of the health state, as shown in fig. 5, equally dividing each one-dimensional time sequence original data segment containing L data points into h one-dimensional time sequence original small data segments with equal length (assuming that the length of each small data segment with equal length is k data points, k × h = L, and the value range of k is between 100 and 10000), as shown in fig. 5, T channels can obtain T × h one-by-one data segmentsDimension time series original small data segments, so that each health state can obtain h one-dimensional time series original small data segment groups with T channels, each one-dimensional time series original small data segment group with T channels forms a fault sample, as shown in fig. 6, that is, each fault sample comprises T channels, wherein each channel has k data points, and a multichannel one-dimensional time series original fault data set { phi } subjected to an equally dividing and truncating operation is subjected to equally dividing and truncating operation_{Original source}Is marked as { phi }_{Cutting block}And therefore { φ }_{Cutting block}The method comprises the steps that N x h one-dimensional time sequence small data segment groups with T channels are contained, and each one-dimensional time sequence original small data segment group with T channels is marked as a fault sampleX _(k)}^(T)I.e., each fault sample contains T channels, each channel containing k data points (as shown in fig. 6);

step (4), constructing a multi-channel two-dimensional feature map fault data set; (phi) obtained after the equal division truncation operation_{Cutting block}Is last fault sampleX _(k)}^(T)The one-dimensional time series data of each channel in (1) is further reconstructed into a two-dimensional characteristic map in a data format, as shown in fig. 7, the specific data reconstruction method is as follows: first, a fault sample is mappedX _(k)}^(T)K length one-dimensional time series small data segment component of each channel in the channelx _(k)Is reconstructed into a matrix form of a two-dimensional feature map in a one-dimensional vector format of (2)x]_m×n(i.e., the size of each feature map is m × n = k), the construction method is set as follows: dividing a data segment containing k data points into m parts, wherein each part contains n data points, and the arrangement sequence is as follows: placing the 1 st n data point on the 1 st row, the 2 nd n data point on the 2 nd row, the 3 rd n data point on the 3 rd row, and sequentially sequencing, … …, placing the mth n data point on the mth row, thereby obtaining an m × n two-dimensional characteristic diagram (as shown in fig. 7); according to the same method, each fault sampleX _(k)}^(T)T two-dimensional characteristic graphs with the size of m multiplied by n can be obtained through the middle T channels, and the two-dimensional characteristic graphs are shown in FIG. 8; secondly, the fault sample is mappedX _(k)}^(T)According to the slave channel 1, the channel 2 and the … …And the channels T are sequentially overlapped, as shown in FIG. 9, a fault sample containing a two-dimensional characteristic diagram of the T channel can be constructedX _[m×n]}^(T)The sample format is a readable sample form for the fault diagnosis of the convolutional neural network designed by the invention in the subsequent step; according to the same sample reconstruction method, the { phi } can be obtained_{Cutting block}N x h one-dimensional time series small data segment group fault samples with T channelX _(k)}^(T)Reconstructed into N x h two-dimensional characteristic graphs with T channels superposed and with size of m x NX _[m×n]}^(T)(ii) a And reconstructs the data format to { phi }_{Cutting block}Recording as a multi-channel two-dimensional feature map fault data set (phi)_2D；

Step (5), dividing a data set; a multi-channel two-dimensional feature map fault data set (phi)_2DThe h samples in each health state type are divided into a training set, a verification set and a test set, and the division method comprises the following steps: firstly, a multichannel two-dimensional feature map fault data set { phi }_2DRandomly selecting 30% of samples as a test set from h fault samples in each health state type, and randomly taking out 80% of the rest 70% of samples to be classified into a training set and 20% of samples to be a verification set; and finally, a multi-channel two-dimensional feature map fault data set { phi }_2DAll the sets of fault samples classified as training sets in the N-type health states in (1) are the total training set { D }_{Training device}Will { phi }_2DAll the failure sample group set classified as verification set in the N-type health states are set as a total verification set { D }_{Test (experiment)}Will { phi }_2DAll the fault sample groups classified as test sets of the N-type health states of (1) are set as a total test set { D }_Measuring；

Step (6), constructing a multi-channel deep learning fault diagnosis model; as shown in fig. 2, the multi-channel deep learning fault diagnosis model includes an input layer 1, a feature extraction layer 2, a dimensionality reduction parameter reduction layer 3, a softmax classification layer 4 and a support vector machine output layer 5, the feature extraction layer 2 includes a first convolution layer 21, a first pooling layer 22, a second convolution layer 23 and a second pooling layer 24, the dimensionality reduction parameter reduction layer 3 includes a transition convolution layer 31 with a convolution kernel of 1 × 1 and a global mean pooling layer 32, characterized in that: a dimension reduction parameter reduction layer 3 is arranged between the feature extraction layer 2 and the softmax classification layer 4 and used for replacing a fully-connected network layer part of a traditional convolutional neural network, and the input layer 1 is used for receiving a multi-channel two-dimensional feature map fault data set (phi)_2DTotal training set of (D) }_{Training device}Total verification set { D }_{Test (experiment)}And the total test set { D }_MeasuringThe multi-channel two-dimensional feature map data is input into a first convolution layer 21 in a feature extraction layer 2, the 1 × 1 transition convolution layer 31 is used for receiving an output feature map of a second pooling layer 24 of the feature extraction layer 2, the input layer 1, the first convolution layer 21, the first pooling layer 22, the second convolution layer 23, the second pooling layer 24, the 1 × 1 transition convolution layer 31 and a global mean pooling layer 32 are sequentially connected in series, a softmax classification layer 4 and a support vector machine output layer 5 are connected in parallel behind the global mean pooling layer 32, the multi-channel deep learning fault diagnosis model comprises two model combinations, namely a model training stage model combination and a model testing stage model combination, and as shown in fig. 2, the model combination mode is designed as follows: in the training stage of the multi-channel deep learning fault diagnosis model, the softmax classification layer 4 is connected behind the global mean pooling layer 32 and is used for training model parameters of each layer of the convolutional neural network, and in the testing stage of the multi-channel deep learning fault diagnosis model, the support vector machine output layer 5 is connected behind the global mean pooling layer 32 and is used for further improving and outputting a final diagnosis result.

Step (7), training a multi-channel deep learning fault diagnosis model; firstly, adopting the model combination of the training phase of the multi-channel deep learning fault diagnosis model in the step (6), connecting the softmax classification layer 4 behind the global mean pooling layer 32 to execute CNN model parameter training, initializing model parameters, and integrating the total training set { D }_{Training device}Inputting the multi-channel two-dimensional feature map data samples into the multi-channel deep learning fault diagnosis model for training and learning the parameters of the deep learning model, repeatedly executing the iterative computation process of forward propagation and backward propagation, and performing the iterative computation process on the first convolution layer 21, the first pooling layer 22, the second convolution layer 23, the second pooling layer 24, the 1 x 1 transition convolution layer 31 and the global regionTraining the model parameters of each layer in the mean pooling layer 32;

step (8), verifying a multi-channel deep learning fault diagnosis model; while training a multi-channel deep learning fault diagnosis model, a total verification set { D }is obtained_{Test (experiment)}The multi-channel two-dimensional characteristic pattern data is used for verifying the diagnosis accuracy of the multi-channel deep learning fault diagnosis model in the training process in real time, verifying the accuracy of the model on a verification set and checking whether overfitting occurs, and the verification standard is designed as follows: when the accuracy rates of the training set and the verification set increase along with the increase of the number of the training rounds and the accuracy rate of the training set is higher than that of the verification set, the model parameters are normally trained, and the step (7) is continuously executed for training; when the accuracy on the training set and the accuracy on the verification set increase along with the increase of the number of training rounds and the accuracy on the training set begins to be lower than the accuracy on the verification set, which indicates that the model parameter training is overfitting, stopping the model training, skipping to the step (6), revising the hyper-parameters of the multi-channel deep learning fault diagnosis model, and repeating the steps until the multi-channel deep learning fault diagnosis model is not overfitting and the accuracy on the verification set reaches a set target value or the number of iteration rounds;

step (9), when the accuracy of the verification set reaches a set target value or training iteration times, the model training is finished, and the optimal model parameter values of each layer of the first convolution layer 21, the first pooling layer 22, the second convolution layer 23, the second pooling layer 24, the 1 × 1 transition convolution layer 31 and the global mean pooling layer 32 in the multi-channel deep learning fault diagnosis model are stored;

step (10), testing a multi-channel deep learning fault diagnosis model; firstly, adopting the model combination of the testing stage of the multi-channel deep learning fault diagnosis model in the step (6), connecting a support vector machine output layer 5 behind a global mean pooling layer 32, and secondly, combining the total training set { D }_{Training device}The multichannel two-dimensional feature map data sample is input into a trained multichannel deep learning fault diagnosis model and comprises a trained first convolution layer 21, a trained first pooling layer 22, a trained second convolution layer 23, a trained second pooling layer 24,The 1 × 1 transition convolution layer 31 and the global mean pooling layer 32 sequentially provide characteristic values for input samples, and then output values of the global mean pooling layer 32 are input into the support vector machine to complete training of model parameters of the support vector machine and store the model parameters of the support vector machine; finally, the total test set { D }_MeasuringThe sample data is input into a model of a test phase model combination of the trained multichannel deep learning fault diagnosis model, the trained first convolution layer 21, the trained first pooling layer 22, the trained second convolution layer 23, the trained second pooling layer 24, the trained 1 × 1 transition convolution layer 31, the trained global mean value pooling layer 32 and the trained support vector machine output layer 5 sequentially present features for the input sample, and finally the trained support vector machine output layer 5 outputs a final fault diagnosis result.

In the experimental implementation, the sensor types are set to include one or more combinations of a vibration acceleration sensor, a noise sensor, a pressure sensor, a displacement sensor, a rotating speed sensor, a voltage sensor and a current sensor.

In the experiment, different types and numbers of sensors for monitoring the data acquisition of the object are set to have the same sampling frequency and sampling time.

In the experimental implementation, the monitoring objects include conventional rotating electromechanical equipment commonly used in the technical field, such as an engine, a motor, a rolling bearing, a gear and the like.

In order to further illustrate the feasibility and effectiveness of the method for diagnosing faults based on the multi-channel time series data of the convolutional neural network provided by the invention in fault diagnosis for the multi-channel one-dimensional time series data of electromechanical equipment, in this embodiment, application example description and verification are further performed by using multi-channel data of a bearing laboratory (as shown in fig. 10) of an electrical engineering laboratory of the university of caesarean storage, usa.

The method provided by the invention comprises the following steps in sequence: firstly, the monitoring object of the experiment is a rolling bearing of the motor, two vibration acceleration sensors are respectively arranged right above a driving end and a fan end of the motor by magnetic seats and are used for acquiring vibration acceleration one-dimensional time series signals of two parts of the motor, and each sensor acquires data of one channel, so the experiment is one-dimensional time series data of 2 channels.

The bearing of the experiment selects a 6205-2RS JEM deep groove ball bearing produced by Swedish SKF company, the bearing is used for supporting a main shaft of a motor, the fault type of the bearing comprises 1 normal state and 9 fault states, and 10 health states are counted; the 9 fault states are pits with pit sizes of 0.18mm, 0.36mm and 0.53mm, which are respectively arranged on an inner ring, an outer ring and balls of the rolling bearing, and the number and the size of the faults are described in table 1; the sampling frequency of the experiment for 10 health states is 12KHz (10,000 data points per second), the sampling time is 10 seconds, namely 10 × 12,000=120,000 data points, and for the convenience of subsequent sample segmentation calculation, the experiment is rounded, and only the first 100,000 data points in the 120,000 data points are reserved, so each health state comprises two channels, and each channel obtains a one-dimensional time sequence data segment with the length of 100,000 data points.

Secondly, performing normalization preprocessing operation on the one-dimensional time sequence data section with the length of 100,000 data points of each channel in 10 health states, so that the data point value of each data section is converted into 0-1;

thirdly, equally dividing the data segment, equally dividing the one-dimensional time series data segment with the length of 100,000 data points in each of the 2 channels of each health state of the experiment into 200 parts according to the method step (3) provided by the invention and the methods shown in fig. 5 to fig. 6, and equally dividing the one-dimensional time series data segment with the length of 500 data points in each channel, so that 200 samples are obtained in each health state, each sample comprises 2 channels, and each channel comprises 500 data points, which is shown in table 1;

fourthly, reconstructing each 2-channel time series sample (500 data points per channel) into a multi-channel two-dimensional feature map sample according to the method described in step (4) of the method provided by the invention and the method described in fig. 7 to fig. 9, equally dividing 500 data points per channel in each 2-channel time series sample into 25 parts, each part containing 20 data points, thereby reconstructing a one-dimensional time series data segment containing 500 data points into a two-dimensional feature map with a size of [25,20] matrix form; thus, each sample is a 2-channel two-dimensional profile sample [25,20,2] comprising two [25,20] two-dimensional profiles superimposed on each other, where 2 represents 2 channels, and thus the experiment contains 10 types of health states in total, each health state containing 200 samples, each sample being a 2-channel two-dimensional profile in the format [25,20,2], as shown in table 1.

Table 1 fault data set for experimental bearings.

Fifthly, 30% of 200 samples of each health state type were taken as a test set (200 × 0.3= 60), 20% of the remaining 70% were taken as a verification set (200 × 0.7 × 0.2= 28), and the remaining 80% were taken as a training set (200 × 0.7 × 0.8= 112), and the experiment included 10 state types, and therefore, the total number of training set samples was 1120 (10 × 112 equals 1120), the total number of verification set samples was 280 (10 × 28 equals 280), and the total number of test set samples was 720 (10 × 60 equals 600).

Sixthly, in the embodiment, the constructed multi-channel deep learning fault diagnosis model is shown in fig. 2 and comprises a first convolution layer 21, a first pooling layer 22, a second convolution layer 23, a second pooling layer 24, a 1 × 1 transition convolution layer 31, a global mean pooling layer 32, a softmax classification output layer 4 and a support vector machine output layer 5 which are mutually sequential, the construction, training, verification and final test links of the multi-channel deep learning fault diagnosis model are completed according to the steps (6) to (10) of the method provided by the invention, and the detailed multi-channel deep learning fault diagnosis model hyper-parameter of the experiment is shown in table 2.

Table 2 multiparameter of the multi-channel deep learning fault diagnosis model.

In this embodiment, the data processing flow of the multi-channel deep learning fault diagnosis model designed in this experiment shown in table 2 is as follows: the feature map format of the input sample is [25,20,2], the first convolution layer 21 adopts 64 convolution kernels of 3 × 3 to perform the same convolution operation on the input feature map [25,20,2], respectively, so as to obtain output feature maps of 64 channels: [25,20,64 ]; then, performing maximum pooling operation on the output characteristic diagram [25,20,64] of the first convolution layer 21 by the first pooling layer 22, wherein the pooling core of the first pooling layer 22 is 2 × 2, the step length is [2,2], and obtaining a characteristic diagram [12,10,64] after the pooling operation; then, the second convolution layer 23 performs a second convolution operation on the output feature map [12,10,64] of the first pooling layer 22, and the second convolution layer 23 respectively performs a same convolution operation on the feature maps [12,10,64] by using 32 convolution kernels of 3 × 3 to obtain an output feature map of 32 channels: [12,10,32 ]; then, the second pooling layer 24 performs maximum pooling operation on the output characteristic diagram [12,10,32] of the second convolutional layer 23, the pooling kernel of the second pooling layer 24 is 2 × 2, the step length is [2,2], and a characteristic diagram [6,5,32] is obtained after the pooling operation; then, a third convolution operation is performed on the output feature maps [6,5,32] of the second pooling layer 24 by the 1 × 1 transition convolution layer 31, and the third convolution layer 31 respectively performs a same convolution operation on the feature maps [6,5,32] by using 10 1 × 1 convolution kernels to obtain 10-channel output feature maps: [6,5,10 ]; then, a global mean pooling layer 32 is arranged behind the third convolutional layer 31, wherein the global mean pooling layer 32 adopts 10 pooling cores of 6 × 5 to perform global mean pooling calculation on the output feature maps [6,5,10] of the third convolutional layer 31, and all values in the feature maps of each [6,5] are subjected to a global mean value [1,10 ]; in the training phase, the output feature vector of the global mean pooling layer 32 is continuously input to the Softmax classification layer 4 for result calculation, error calculation and error back propagation, model parameters of the CNN feature extraction layer 2 and the dimensionality reduction and parameter reduction layer 3 are trained, and in the testing phase, the output feature vector [1,10] of the global mean pooling layer 32 is directly input to the support vector machine output layer 5 for final diagnosis result output, as shown in table 2.

Compared with the conventional CNN with the same scale, the multichannel time series data fault diagnosis method based on the convolutional neural network provided by the invention has the advantages that the parameter quantity is obviously reduced, as shown in Table 3, the total number of model parameters of the conventional CNNs-full-connection network is 143,978, and the number of model parameters provided by the invention is only 20,120.

Table 3 CNN model training parameter number comparison table.

Compared with the conventional CNN with the same scale, the multichannel time series data fault diagnosis method based on the convolutional neural network has the advantages that the accuracy is improved, and the display of the test time and the training time of the fault is reduced, as shown in Table 4. As can be seen from the comparison of Table 4, the performance of the method provided by the invention is obviously improved compared with the traditional full-connection CNN method.

Table 4 fault diagnosis results comparison table.

Model name	Accuracy of test	Training time	Time of measurement
				The method provided by the invention	99.89%	340.12 seconds	0.194 second
Conventional CNN method	98.75%	358.96 seconds	0.259 second

As shown in table 4, the method designed by the present invention includes two parts, one is the accuracy obtained when the improved CNNs and Softmax are combined for performing the back propagation optimization training of the CNNs model, and the other is the accuracy obtained when the trained CNNs model is used as a feature extractor for performing feature extraction on new fault data and then input into the SVM for fault classification; as can be seen from a comparison of Table 4, the following are obtained in terms of accuracy: the accuracy rate of the traditional CNNs is 98.75 percent, but the accuracy rate of the method provided by the invention is improved to 99.89 percent; in terms of time, the method provided by the invention has the advantages that the quantity of model parameters is greatly reduced due to the elimination of the full-connection part, the training time and the testing time are obviously reduced, and the method is of great significance for the rapid diagnosis and online monitoring of the fault.

In order to further verify the effectiveness of the method provided by the present invention compared with the current mainstream intelligent diagnosis method, the present embodiment compares and verifies the diagnosis result of the present invention with the current mainstream Support Vector Machine (SVM), BP neural network (BPNN), K-nearest neighbor method (KNN), and deep BP neural network (DNN), and the result is shown in table 5.

As can be seen from Table 5, the diagnostic accuracy of the method provided by the present invention is significantly better than that of other methods currently available.

Table 5 five methods diagnosis accuracy data table.

It should also be understood that the above description is only a preferred embodiment of the present invention, and not intended to limit the present invention, and that the technical matters in the present invention are included in the technical contents of the present invention.

Claims

1. A multi-channel time series data fault diagnosis method based on a convolutional neural network is characterized by comprising the following steps:

(1) collecting multi-channel one-dimensional time series fault data of a monitored object; arranging sensors on a monitored object, and acquiring one-dimensional time sequence data monitoring signals generated when the monitored object runs in multiple health states by using the sensors, wherein the number of measuring points arranged on the monitored object by the sensors is set to be T (namely T channels) and used for acquiring fault data of T parts of the monitored object, each measuring point is provided with one sensor, the data acquired by each sensor is a continuous one-dimensional time sequence original data segment, and the sample length of the one-dimensional time sequence original data segment is L, namely L data points are included; the health state of the monitoring object is set to be N health states, wherein the N health states comprise a normal state and N-1 fault states, and each health state comprises a one-dimensional time sequence original fault data segment of T channels;

(3) preprocessing data; acquiring a multichannel one-dimensional time sequence original fault data set (phi)_{Original source}Carrying out data preprocessing operation on the N multiplied by T one-dimensional time sequence original data segments, wherein the preprocessing comprises normalization and data truncation; firstly, carrying out normalization data processing on the data of the NxT one-dimensional time sequence original data segments one by one, and converting the magnitude of the values of all data points in each one-dimensional time sequence original data segment into 0-1, wherein the normalization method comprises the following steps:X={x _i}=(x _i-x _min)/(x _max-x _min) (ii) a Secondly, normalizing the one-dimensional time sequence of each channel in the T channels after the treatment of each health stateSegmenting, equally dividing and truncating an original data segment to equally divide each one-dimensional time sequence original data segment containing L data points into h one-dimensional time sequence original small data segments with equal length (assuming that the length of each small data segment with equal length is k data points, k multiplied by h = L, and the value range of k is between 100 and 10000), obtaining T multiplied by h one-dimensional time sequence original small data segments by T channels, thereby obtaining h one-dimensional time sequence original small data segment groups with T channels by each health state, forming a fault sample by each one-dimensional time sequence original small data segment group with T channels, wherein each channel comprises T channels, and dividing and equally dividing a multichannel one-dimensional time sequence original fault data set { phi } after truncation operation into k channels_{Original source}Is marked as { phi }_{Cutting block}And therefore { φ }_{Cutting block}The method comprises the steps that N x h one-dimensional time sequence small data segment groups with T channels are contained, and each one-dimensional time sequence original small data segment group with T channels is marked as a fault sampleX _(k)}^(T)That is, each fault sample contains T channels, each channel containing k data points;

(4) constructing a multi-channel two-dimensional feature map fault data set; (phi) obtained after the equal division truncation operation_{Cutting block}Is last fault sampleX _(k)}^(T)The one-dimensional time sequence data of each channel is further reconstructed into a two-dimensional characteristic diagram in a data format, and the specific data reconstruction method comprises the following steps: first, a fault sample is mappedX _(k)}^(T)K length one-dimensional time series small data segment component of each channel in the channelx _(k)Is reconstructed into a matrix form of a two-dimensional feature map in a one-dimensional vector format of (2)x]_m×n(i.e., the size of each feature map is m × n = k), the construction method is set as follows: dividing a data segment containing k data points into m parts, wherein each part contains n data points, and the arrangement sequence is as follows: placing the 1 st n data point on the 1 st row, the 2 nd n data point on the 2 nd row, the 3 rd n data point on the 3 rd row, and sequentially sequencing, … …, wherein the mth n data point is placed on the mth row, thereby obtaining an m multiplied by n two-dimensional characteristic diagram; in the same way, each failure sampleInstant dictionaryX _(k)}^(T)T two-dimensional characteristic graphs with the size of m multiplied by n can be obtained through the middle T channels; secondly, the fault sample is mappedX _(k)}^(T)The T two-dimensional feature maps are sequentially superposed according to the sequence of a channel 1, a channel 2, a channel … … and a channel T, and then a fault sample tone containing the two-dimensional feature map of the T channel can be constructedX _[m×n]}^(T)The sample format is a readable sample form for the fault diagnosis of the convolutional neural network designed by the invention in the subsequent step; according to the same sample reconstruction method, the { phi } can be obtained_{Cutting block}N x h one-dimensional time series small data segment group fault samples with T channelX _(k)}^(T)Reconstructed into N x h two-dimensional characteristic graphs with T channels superposed and with size of m x NX _[m×n]}^(T)(ii) a And reconstructs the data format to { phi }_{Cutting block}Recording as a multi-channel two-dimensional feature map fault data set (phi)_2D；

(5) Dividing a data set; a multi-channel two-dimensional feature map fault data set (phi)_2DThe h samples in each health state type are divided into a training set, a verification set and a test set, and the division method comprises the following steps: firstly, a multichannel two-dimensional feature map fault data set { phi }_2DRandomly selecting 30% of samples as a test set from h fault samples in each health state type, and randomly taking out 80% of the rest 70% of samples to be classified into a training set and 20% of samples to be a verification set; and finally, a multi-channel two-dimensional feature map fault data set { phi }_2DAll the sets of fault samples classified as training sets in the N-type health states in (1) are the total training set { D }_{Training device}Will { phi }_2DAll the failure sample group set classified as verification set in the N-type health states are set as a total verification set { D }_{Test (experiment)}Will { phi }_2DAll the fault sample groups classified as test sets of the N-type health states of (1) are set as a total test set { D }_Measuring；

(6) Constructing a multi-channel deep learning fault diagnosis model; the multi-channel deep learning fault diagnosis model comprises an input layer, a feature extraction layer, a dimension reduction and parameter reduction layer, a softmax classification layer and a support vector machine output layer, wherein the feature extraction layer comprises a first layerThe dimensionality reduction parameter layer comprises a transition convolution layer with a convolution kernel of 1 multiplied by 1 and a global mean value pooling layer, and is characterized in that: a dimension reduction parameter reduction layer is arranged between the feature extraction layer and the softmax classification layer and is used for replacing a fully-connected network layer part of a traditional convolutional neural network, and the input layer is used for receiving a multi-channel two-dimensional feature map fault data set (phi)_2DTotal training set of (D) }_{Training device}Total verification set { D }_{Test (experiment)}And the total test set { D }_MeasuringThe multi-channel two-dimensional feature map data are input into a first convolution layer in a feature extraction layer, a 1 x 1 transition convolution layer is used for receiving an output feature map of a second pooling layer of the feature extraction layer, the input layer, the first convolution layer, the first pooling layer, the second convolution layer, the second pooling layer, the 1 x 1 transition convolution layer and a global mean pooling layer are sequentially connected in series, a softmax classification layer and a support vector machine output layer are connected in parallel behind the global mean pooling layer, the multi-channel deep learning fault diagnosis model comprises two model combinations of a model training phase model combination and a model testing phase model combination, and the design is as follows: in the training stage of the multi-channel deep learning fault diagnosis model, the softmax classification layer is connected behind the global mean pooling layer and used for training model parameters of each layer of the convolutional neural network, and in the testing stage of the multi-channel deep learning fault diagnosis model, the support vector machine output layer is connected behind the global mean pooling layer and used for further improving and outputting a final diagnosis result.

2. Training a multi-channel deep learning fault diagnosis model; firstly, adopting the model combination of the training phase of the multi-channel deep learning fault diagnosis model in the step (6), connecting a softmax classification layer behind a global mean pooling layer to execute CNN model parameter training, initializing model parameters, and collecting the total training set { D }_{Training device}Inputting the data samples of the multi-channel two-dimensional characteristic diagram into the multi-channel deep learning fault diagnosis model for training and learning the parameters of the deep learning model, repeatedly executing the iterative calculation process of forward propagation and backward propagation, and performing the iterative calculation on the parameters of the deep learning modelTraining model parameters of each of the first convolution layer, the first pooling layer, the second convolution layer, the second pooling layer, the 1 × 1 transition convolution layer and the global mean pooling layer;

(10) testing a multi-channel deep learning fault diagnosis model; firstly, adopting the model combination of the testing stage of the multi-channel deep learning fault diagnosis model in the step (6), connecting a support vector machine output layer after a global mean pooling layer, and secondly, combining the total training set { D }_{Training device}The multichannel two-dimensional characteristic diagram data samples are input into a trained multichannel deep learning fault diagnosis model and are processed by a trained first volumeThe method comprises the following steps that a characteristic value is sequentially provided for an input sample by an accumulation layer, a first pooling layer, a second convolution layer, a second pooling layer, a 1 x 1 transition convolution layer and a global mean pooling layer, then, an output value of the global mean pooling layer is input into a support vector machine, training of model parameters of the support vector machine is completed, and the model parameters of the support vector machine are stored; finally, the total test set { D }_MeasuringThe method comprises the steps of inputting sample data into a trained model of a test stage model combination of the multi-channel deep learning fault diagnosis model, sequentially providing features for input samples by a trained first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer, a 1 x 1 transition convolution layer, a global mean value pooling layer and a support vector machine output layer, and finally outputting a final fault diagnosis result by the support vector machine output layer.

3. The convolutional neural network-based multi-channel time series data fault diagnosis method as claimed in claim 1, wherein the sensor types are set to include one or more combinations of vibration acceleration sensors, noise sensors, pressure sensors, displacement sensors, rotation speed sensors, voltage sensors and current sensors.

4. The convolutional neural network-based multi-channel time series data fault diagnosis method as claimed in claim 1, wherein the sensors for monitoring the object data acquisition of different types and numbers are set to have the same sampling frequency and sampling time.

5. The convolutional neural network-based multi-channel time series data fault diagnosis method as claimed in claim 1, wherein the monitored object comprises conventional rotating electromechanical equipment commonly used in the technical field, such as an engine, a motor, a rolling bearing, a gear and the like.