CN114897286A - Fault model establishing method based on alliance chain - Google Patents

Abstract

The invention provides a fault model building method based on a alliance chain, which comprises the following steps: training and establishing a production fault recognition model by using historical production data; establishing a alliance chain for fault identification, and carrying out data encryption transmission among all nodes through the alliance chain; acquiring real-time production data of the operation of the sewing machine, transmitting the real-time production data to a production fault identification model, judging whether the production fault identification model successfully identifies the fault type, and if so, feeding the fault type back to an enterprise node; if not, extracting the production data corresponding to the unknown fault and feeding the production data back to the enterprise node, and feeding the production data of the unknown fault back to the operator node by the enterprise node through the alliance chain. The invention encrypts the fault data by constructing the alliance chain, prevents the data from being lost or tampered, ensures that the subsequent fault analysis can be safely carried out, and simultaneously transmits the subsequent fault solution by the alliance chain, and prevents the solution from being tampered.

Description

A method for establishing fault model based on consortium chain

technical field

The invention relates to the technical field of fault judgment, in particular to a method for establishing a fault model based on a consortium chain.

Background technique

With the development of the textile industry and the improvement of automatic control technology, the demand for automation equipment in the sewing industry is also getting higher and higher, which puts forward higher requirements for the automation control level of sewing equipment. Industrial sewing machines are typical automation equipment in the sewing industry, and are mainly used in sewing factories or other industrial sectors that mass-produce sewn products, such as sewing factories for various quilts, sleeping bags, and curtains. To realize the automatic control of industrial sewing machines, there must be a control system. The control system includes the main control unit MCU. The main control unit MCU transmits signals to the motor controller through the electrical control connection interface, and the motor controller controls the motor to complete the sewing work. The automatic control of industrial sewing machines in the industrial field is to increase product output and improve work efficiency. Therefore, in the existing working environment, many industrial sewing machines are often working at the same time. The work of industrial sewing machines must be managed by staff in order to deal with them in time. some unexpected situations. If each industrial sewing machine is equipped with a staff, it will undoubtedly be a waste of human resources; if the working status of the industrial sewing machine is monitored and managed through the flow of staff, the workload of the staff is very large, and it is also possible Accidents occur due to insufficient monitoring; at the same time, once a problem occurs, the staff responsible for monitoring and management may not know where the problem lies, and it is still necessary to communicate with the manufacturer of the industrial sewing machine and have professional maintenance personnel to repair the industrial sewing machine. For users who use industrial sewing machines, it not only wastes human resources, but also cannot solve problems in time, which brings great inconvenience to users who use industrial sewing machines and delays production. With reference to the Chinese patent authorization announcement number CN103451866B, an industrial sewing machine remote monitoring and fault diagnosis system includes a motor, a motor controller, an electric control connection interface and a main control unit installed on the industrial sewing machine and connected in sequence, and also includes a connection with the main control unit. The data collector connected to the unit, the data collector communicates with the repeater through the data transmission device, the repeater is provided with a GPRS/GSM module, the repeater is connected in communication with the network server, and the network server is respectively connected with the display terminal It is connected to the fault diagnosis system, and the information such as the machine setting parameters of the motor is transmitted to the repeater through the data transmission device through the data collector, and the repeater transmits the received information to the customer service website through the built-in GPRS/GSM module. Server, customer service personnel remotely monitor the working status of the industrial sewing machine through the display terminal and fault diagnosis system connected to the website server, so as to help solve the problem in time. However, due to the lack of effective security measures during network transmission, data may be lost or tampered with during network transmission, resulting in failure to perform correct fault diagnosis in the future.

SUMMARY OF THE INVENTION

The invention solves the problem of unsafe network transmission for remote monitoring and fault diagnosis, and proposes a method for establishing a fault model based on a consortium chain. By constructing a consortium chain, the fault data is encrypted to prevent data loss or tampering, so that subsequent fault analysis can be performed. It is carried out safely, and the subsequent fault solutions are also transmitted through the alliance chain to prevent the solutions from being tampered with.

In order to achieve the above purpose, the following technical solutions are proposed:

A method for establishing a fault model based on a consortium chain, comprising the following steps:

S1, obtain the historical production data of sewing machine operation, use the historical production data to train and establish a production fault identification model, and the production fault identification model is provided with a fault type library;

S2, constructing a consortium chain for fault identification, the consortium chain includes several enterprise nodes and operator nodes, and data encryption transmission is performed between each node through the consortium chain;

S3, obtain the real-time production data of sewing machine operation and transmit it to the production fault identification model, and judge whether the production fault identification model successfully identifies the fault type, if so, feed back the fault type to the enterprise node; if not, go to S4;

S4, extract the production data corresponding to the unknown fault and feed it back to the enterprise node, and the enterprise node feeds back the production data of the unknown fault to the operator node through the alliance chain;

S5, the operator node identifies the production data of the unknown fault, obtains a new fault type, and sends the data of the new fault type to each enterprise node through the alliance chain, and the fault identification model updates the fault type database.

In the present invention, one enterprise node corresponds to one production fault identification model, and one production fault identification model can simultaneously identify several sewing machines corresponding to the enterprise node. When an unknown fault occurs and needs to be solved by the operator node, the fault data can be encrypted through the alliance chain Uploaded to the operator node, due to the immutability of the transmission of the alliance chain, the fault data is the original collected and real data, which provides convenience for the operator to find a solution. At the same time, due to the data sharing within the alliance chain, when a sewing machine appears For new faults, after the operator node solves the fault, it can synchronize the solution of the new fault to each enterprise node, and the production fault identification model of each enterprise node is updated regularly, so that the same fault can be dealt with in time.

Preferably, the production fault identification model is provided with a fault processing library, and the fault processing library records the processing method corresponding to the fault, and the operator node provides a new fault processing method corresponding to the new fault type, and transmits it to the Each enterprise node, then the fault identification model updates the fault processing library.

Preferably, the S1 specifically includes the following steps:

S101. Simultaneously collect and de-noise several types of vibration data when the sewing machine fails, and sample the several types of vibration data according to time series to obtain historical production data;

S102, according to the set time step, generate a training set and a test set based on the historical production data;

S103, setting a label representing the failure type of mechanical equipment for the training set;

S104, train the double-layer LSTM neural network through the training set and the test set; the double-layer LSTM neural network is used for judging the failure type of mechanical equipment failure; the test set is used for verifying the trained LSTM neural network; verified The trained LSTM neural network is the production fault identification model.

Preferably, the double-layer LSTM neural network includes: a first hidden layer, a second hidden layer, and a flattening module;

When the i-th two-layer LSTM network is trained, i∈[1,L-w+1], where L is a constant, including:

Input the time series data x _i ,...,x _i+w in the test set into the first hidden layer in sequence, and after the LSTM standard module of the first hidden layer performs operations for w time steps, the first output result of the corresponding time step is obtained ; where x _i ,...,x _i+w are used as the input of the LSTM standard module of the first hidden layer at the w time steps;

The first output results are sequentially input into the second hidden layer, and after the LSTM standard module of the second hidden layer performs operations for w time steps, the second output results of the corresponding time steps are obtained, wherein the first output results are respectively used as the second output results. The input of the LSTM standard module of the hidden layer at the w time steps;

The second output result is input into the flattening module to obtain a classification result of the failure type of the mechanical equipment, and the flattening module includes a classifier.

Preferably, the process of encrypting data transmission between the nodes through the alliance chain is as follows:

S201, each node stores a separate key, when a node uploads shared data, the node converts the data to be shared into data blocks, and the node annotates the data blocks in plaintext, and uploads them to the alliance chain intranet platform ;

S202, the alliance chain intranet platform divides the data block into several data blocks, and each data block is individually distributed to several nodes according to a certain order; several nodes use keys to encrypt the data blocks, and after several nodes encrypt The encrypted data blocks are combined into encrypted shared data blocks according to a certain order, the encrypted shared data blocks are remarked with the plaintext, and the number of nodes participating in the encryption is more than half of the number of all nodes;

S203, the alliance chain intranet platform displays the encrypted shared data block, and allows the encrypted shared data block to be copied to any node.

Preferably, the present invention also includes the decryption step of the shared data block:

When the computer application in the intranet environment of the alliance chain requests to open the shared data, it sends a request to the intranet platform of the alliance chain, and the intranet platform of the alliance chain verifies whether the application is in the whitelist. In the whitelist, the intranet platform of the alliance chain sends a decryption request to the nodes participating in the encryption. When each node is decrypted, it puts the decrypted shared data into the designated buffer, and replaces the data read address of the application with For the buffer address, when the application is closed, the encryption/decryption monitoring node clears the buffer, and if the application is not in the whitelist, no operation is performed.

The beneficial effects of the present invention are: the fault data is encrypted and uploaded to the operator node through the alliance chain. Due to the non-tampering of the transmission of the alliance chain, the fault data is originally collected and real data, and it is convenient for the operator to find a solution. At the same time, due to the data sharing in the alliance chain, when a sewing machine has a new fault, after the operator node solves the fault, it can synchronize the new fault solution to each enterprise node, and the production fault identification model of each enterprise node is regularly updated, so that When the same fault occurs, it can respond in time.

Description of drawings

FIG. 1 is a flow chart of a method of an embodiment.

Detailed ways

Example:

This embodiment proposes a method for establishing a fault model based on a consortium chain. Referring to FIG. 1 , the method includes the following steps:

S1, obtain the historical production data of sewing machine operation, use the historical production data to train and establish a production fault identification model, the production fault identification model has a fault type library; the production fault identification model has a fault processing library, and the fault processing library records the corresponding faults In the processing method, the operator node provides a new fault processing method corresponding to a new fault type, and transmits it to each enterprise node, and then the fault identification model updates the fault processing library.

S101. Simultaneously collect and de-noise several types of vibration data when the sewing machine fails, and sample several types of vibration data according to time series to obtain historical production data; the sampling data set is recorded as E, E={e _r }r∈[1,num ]; num is the total number of sampling times; _{er = {e′ r1 ,} …,e′ _rp }; er ∈R ^m , e′ _rp is the p-th vibration data sampled for the _rth time, p∈[1, m], where m is the total number of categories of vibration data.

S102. According to the set time step, generate a training set and a test set based on historical production data; let x _i =[e _(i-1)×s+1 ,e _(i-1)×s+2 ,..., e _i×s ]′, x _{i ∈R} ^sm where i∈[ ₁ ,num/s], s is the set time step, [ ]′ represents the transpose of the matrix; The training set, x _L+1 ~ x _num/s will be used as the test set, and L is a set constant.

S103, setting a label representing the failure type of mechanical equipment for the training set;

S104. Train the two-layer LSTM neural network through the training set and the test set; the two-layer LSTM neural network is used to judge the failure type of mechanical equipment failure; the test set is used to verify the trained LSTM neural network; the verified trained LSTM The neural network is the production fault identification model.

The two-layer LSTM neural network includes: the first hidden layer, the second hidden layer, and the flattening module;

其中x_i,…,x_i+w分别作为第一隐层的LSTM标准模块在该w个时间步的输入；将第一输出结果

依序输入第二隐层，经过第二隐层的LSTM标准模块进行w个时间步的运算后，得到对应时间步的第二输出结果

其中第一输出结果

分别作为第二隐层的LSTM标准模块在该w个时间步的输入；When training the two-layer LSTM network for the i-th time, i∈[1,L-w+1], where L is a constant, including: input the time-series data x _i ,...,x _i+w in the test set into the first A hidden layer, after the LSTM standard module of the first hidden layer performs operations for w time steps, the first output result of the corresponding time step is obtained

where x _i ,...,x _i+w are used as the input of the LSTM standard module of the first hidden layer at the w time steps; the first output result is

Input the second hidden layer in sequence, and after the LSTM standard module of the second hidden layer performs operations for w time steps, the second output result of the corresponding time step is obtained.

where the first output result

Respectively as the input of the LSTM standard module of the second hidden layer at the w time steps;

输入扁平化模块得到机械设备故障类型的分类结果，扁平化模块包含分类器。the second output

Input the flattening module to get the classification result of the mechanical equipment failure type, and the flattening module contains the classifier.

The LSTM standard module performs a time step operation including:

S141. Denote the number of hidden units of the LSTM standard module as n, and s is the set time step; the current time step t, time step t-1 is the previous time step of time step t; x _t ∈ R ^sm represents the input at time step t, h _t-1 represents the hidden state at time step t-1, h _t-1 ∈ R ⁿ , the forget gate f _t and the memory gate i _t of the LSTM standard module at time step t are respectively :

f _t =σ(W _f ·[h _t-1 ,x _t ]+b _f );

i _t =σ(W _i ·[h _t-1 ,x _t ]+ _bi );

Among them, f _t and i _t ∈ R ^sn ; represent matrix operations, σ is the sigmod activation function, W _f and W _t ∈ R ^(n+m)s are weight parameters, and b _f and b _i ∈ R ⁿ are biases parameter;

S142, calculate the new state candidate quantity at time step t:

Among them, tanh is the activation function, W _c is the weight parameter, and b _c is the bias parameter;

S143. Calculate the update gate C _t and the output gate o _{t of the time step t} :

o _t =σ(W _o ·[h _t-1 ,x _t ]+b _o );

Among them, C _t and o _t ∈ R ^sn , C _t-1 is the update gate at time step t-1, C _t-1 ∈ R ^sn , W _o is the weight parameter, and b _o is the bias parameter;

S144, h _t is the hidden state of the LSTM standard module at time step t, and is also used as the output result of the LSTM standard module at time step t:

h _t =o _t *tanh(C _t )

where h _t ∈ R ^sn .

S2, build a consortium chain for fault identification. The consortium chain includes several enterprise nodes and operator nodes, and data encryption transmission is performed between each node through the consortium chain;

S3, obtain the real-time production data of sewing machine operation and transmit it to the production fault identification model, and judge whether the production fault identification model successfully identifies the fault type, if so, feed back the fault type to the enterprise node; if not, go to S4;

S4, extract the production data corresponding to the unknown fault and feed it back to the enterprise node, and the enterprise node feeds back the production data of the unknown fault to the operator node through the alliance chain;

S5, the operator node identifies the production data of the unknown fault, obtains a new fault type, and sends the data of the new fault type to each enterprise node through the alliance chain, and the fault identification model updates the fault type database.

The process of data encryption transmission between nodes through the alliance chain is as follows:

S201, each node stores a separate key, when a node uploads shared data, the node converts the data to be shared into data blocks, and the node annotates the data blocks in plaintext, and uploads them to the alliance chain intranet platform ;

S202, the intranet platform of the alliance chain divides the data block into several data blocks, and each data block is individually distributed to several nodes according to a certain order; several nodes use the key to encrypt the data blocks, and the data encrypted by several nodes The blocks are combined into encrypted shared data blocks according to a certain order, the encrypted shared data blocks are remarked with plaintext, and the number of nodes participating in the encryption is more than half of the number of all nodes;

S203, the intranet platform of the alliance chain displays the encrypted shared data block, and allows the encrypted shared data block to be copied to any node.

The present invention also includes the decryption step of the shared data block:

When the computer application in the intranet environment of the alliance chain requests to open the shared data, it sends a request to the intranet platform of the alliance chain, and the intranet platform of the alliance chain verifies whether the application is in the whitelist. If the application is in the whitelist, the alliance The intra-chain network platform sends a decryption request to the nodes participating in the encryption. When the decryption of each node is completed, it puts the decrypted shared data into the designated buffer, and replaces the data read address of the application with the buffer address. When the program is closed, the encryption and decryption monitoring node clears the buffer. If the application is not in the whitelist, it does nothing.

In order to clearly express the encryption and decryption process of the alliance chain, the following describes the data sharing process of a specific node: the shared data includes fault data and fault solution data.

In the first step, node A converts the data to be shared into data blocks D _k , and node A annotates the data blocks in plaintext, and uploads them to the intranet platform of the alliance chain;

In the second step, the intranet platform of the alliance chain divides the data block into several data blocks, and each data block is individually distributed to several nodes according to a certain order; several nodes use the key to encrypt the data blocks, and after several nodes encrypt The encrypted data blocks are combined into encrypted shared data blocks according to a certain order. The encrypted shared data blocks are remarked with plaintext, and the number of nodes participating in the encryption is more than half of the number of all nodes. Specifically, the following steps are included:

a. The alliance chain intranet platform divides the data block D _k into 6/2=3 data blocks d _{ki, i∈[1,3]} , k is the data block label, at this time k is A; when n is an odd number Divided into (n+1)/2 data blocks;

b, establish the encrypted node sequence group corresponding to node A,

Select nodes other than node A, arrange and combine them according to the time of entering the intranet of the alliance chain, and select the sequence combination whose number is the set threshold in the arrangement and combination as the encrypted node sequence group;

Each sequence in the encrypted node sequence group encrypts the data block d _{ki, i ∈ [1, 3]} ;

c. The intranet platform of the alliance chain displays the encrypted shared data block D' _A and allows the encrypted shared data block D' _A to be copied to any node.

Decryption steps:

When the computer application in the intranet environment of the alliance chain requests to open the shared data, it sends a request to the intranet platform of the alliance chain, and the intranet platform of the alliance chain verifies whether the application is in the whitelist. If the application is in the whitelist, the alliance The intra-chain network platform sends a decryption request to the nodes participating in the encryption. When the decryption of each node is completed, it puts the decrypted shared data into the designated buffer, and replaces the data read address of the application with the buffer address. When the program is closed, the encryption and decryption monitoring node clears the buffer. If the application is not in the whitelist, it does nothing.

Taking node B's acquisition of the shared data block D' _A uploaded by node A as an example, the specific decryption process is as follows:

d. In the intranet environment, node B uses a computer application to request to open the shared data, and sends a request to the intranet platform of the alliance chain;

e. The alliance chain intranet platform verifies whether the application is in the whitelist, if so, go to S6, if not, do nothing;

f. The intranet platform of the alliance chain sends a decryption request to the nodes participating in the encryption. When the decryption of each node is completed, the decrypted shared data is put into the designated buffer, which includes the following steps:

The alliance chain intranet platform extracts the encrypted data block combination of the shared data block D'_A; traverses the encrypted data block, divides the encrypted data block into encrypted data blocks, and sends them to each encryption node for decryption until the encrypted data block can be completely decrypted When , the decrypted block combination is restored to the data block D _k , and the data block D _k is put into the specified buffer.

In the present invention, one enterprise node corresponds to one production fault identification model, and one production fault identification model can simultaneously identify several sewing machines corresponding to the enterprise node. When an unknown fault occurs and needs to be solved by the operator node, the fault data can be encrypted through the alliance chain Uploaded to the operator node, due to the immutability of the transmission of the alliance chain, the fault data is the original collected and real data, which provides convenience for the operator to find a solution. At the same time, due to the data sharing within the alliance chain, when a sewing machine appears For new faults, after the operator node solves the fault, it can synchronize the solution of the new fault to each enterprise node, and the production fault identification model of each enterprise node is updated regularly, so that the same fault can be dealt with in time.

Claims (6)

1. a method for establishing a fault model based on a consortium chain, is characterized in that, comprises the following steps: S1, obtain the historical production data of sewing machine operation, use the historical production data to train and establish a production fault identification model, and the production fault identification model is provided with a fault type library; S2, constructing a consortium chain for fault identification, the consortium chain includes several enterprise nodes and operator nodes, and data encryption transmission is performed between each node through the consortium chain; S3, obtain the real-time production data of sewing machine operation and transmit it to the production fault identification model, and judge whether the production fault identification model successfully identifies the fault type, if so, feed back the fault type to the enterprise node; if not, go to S4; S4, extract the production data corresponding to the unknown fault and feed it back to the enterprise node, and the enterprise node feeds back the production data of the unknown fault to the operator node through the alliance chain; S5, the operator node identifies the production data of the unknown fault, obtains a new fault type, and sends the data of the new fault type to each enterprise node through the alliance chain, and the fault identification model updates the fault type database. 2. The method for establishing a fault model based on a consortium chain according to claim 1, wherein the production fault identification model is provided with a fault processing library, and the fault processing library records the processing method corresponding to the fault, and the operation The business node provides a new fault processing method corresponding to the new fault type, and transmits it to each enterprise node, and then the fault identification model updates the fault processing library. 3. The method for establishing a fault model based on a consortium chain according to claim 1, wherein the S1 specifically comprises the following steps: S101. Simultaneously collect and de-noise several types of vibration data when the sewing machine fails, and sample the several types of vibration data according to time series to obtain historical production data; S102, according to the set time step, generate a training set and a test set based on the historical production data; S103, setting a label representing the failure type of mechanical equipment for the training set; S104, train the double-layer LSTM neural network through the training set and the test set; the double-layer LSTM neural network is used for judging the failure type of mechanical equipment failure; the test set is used for verifying the trained LSTM neural network; verified The trained LSTM neural network is the production fault identification model. 4. The method for establishing a fault model based on a consortium chain according to claim 3, wherein the double-layer LSTM neural network comprises: a first hidden layer, a second hidden layer, and a flattening module; When the i-th two-layer LSTM network is trained, i∈[1,L-w+1], where L is a constant, including: Input the time series data x _i ,...,x _i+w in the test set into the first hidden layer in sequence, and after the LSTM standard module of the first hidden layer performs operations for w time steps, the first output result of the corresponding time step is obtained ; where x _i ,...,x _i+w are used as the input of the LSTM standard module of the first hidden layer at the w time steps; The first output results are sequentially input into the second hidden layer, and after the LSTM standard module of the second hidden layer performs operations for w time steps, the second output results of the corresponding time steps are obtained, wherein the first output results are respectively used as the second output results. The input of the LSTM standard module of the hidden layer at the w time steps; The second output result is input into the flattening module to obtain a classification result of the failure type of the mechanical equipment, and the flattening module includes a classifier. 5. a kind of fault model establishment method based on consortium chain according to claim 1, is characterized in that, the process of carrying out data encryption transmission by consortium chain between described each node is as follows: S201, each node stores a separate key, when a node uploads shared data, the node converts the data to be shared into data blocks, and the node annotates the data blocks in plaintext, and uploads them to the alliance chain intranet platform ; S202, the alliance chain intranet platform divides the data block into several data blocks, and each data block is individually distributed to several nodes according to a certain order; several nodes use keys to encrypt the data blocks, and after several nodes encrypt The encrypted data blocks are combined into encrypted shared data blocks according to a certain order, the encrypted shared data blocks are remarked with the plaintext, and the number of nodes participating in the encryption is more than half of the number of all nodes; S203, the alliance chain intranet platform displays the encrypted shared data block, and allows the encrypted shared data block to be copied to any node. 6. a kind of fault model establishment method based on alliance chain according to claim 1, is characterized in that, also comprises the decryption step of shared data block: When the computer application in the intranet environment of the alliance chain requests to open the shared data, it sends a request to the intranet platform of the alliance chain, and the intranet platform of the alliance chain verifies whether the application is in the whitelist. In the whitelist, the intranet platform of the alliance chain sends a decryption request to the nodes participating in the encryption. When each node is decrypted, it puts the decrypted shared data into the designated buffer, and replaces the data read address of the application with For the buffer address, when the application is closed, the encryption/decryption monitoring node clears the buffer, and if the application is not in the whitelist, no operation is performed.

Images