CN111352968A

CN111352968A - Intelligent manufacturing element identification method based on block chain network

Info

Publication number: CN111352968A
Application number: CN202010128679.8A
Authority: CN
Inventors: 黄步添; 周伟华; 陈建海; 刘振广; 肖震; 杨正清
Original assignee: Hangzhou Yunxiang Network Technology Co Ltd
Current assignee: Hangzhou Yunxiang Network Technology Co Ltd
Priority date: 2020-02-28
Filing date: 2020-02-28
Publication date: 2020-06-30
Anticipated expiration: 2040-02-28
Also published as: CN111352968B

Abstract

The invention discloses an intelligent manufacturing element identification method based on a block chain network. And carrying out real-time analysis on worker state data uploaded by intelligent equipment in the intelligent manufacturing process in a frequency domain by using a highly universal Fourier transform algorithm. Through data mining, a sequence mode based on a time sequence and continuously occurring in the manufacturing process and correlation between production processes and items are found, meaningful items in the manufacturing process are extracted as representative subjects in a block chain network, so that intelligent manufacturing subject mining is realized, a side chain-based distributed consensus block chain network is adopted for solving the security problem that intelligent equipment is vulnerable, a subject packaging method is used in the network, an improved block chain distributed account book is applied to the manufacturing process, the information account book is distributed in a peer block chain network, information is recorded and managed together, and data security is improved.

Description

Intelligent manufacturing element identification method based on block chain network

Technical Field

The invention relates to a data mining and managing method, in particular to an intelligent manufacturing element identifying method based on a block chain network.

Background

Intelligent manufacturing utilizes the fusion of intelligent information technology and manufacturing technology to apply standardized industrial engineering software to various aspects of society. To maintain sustainable productivity, innovative intelligent information fusion manufacturing is seen as a new growth engine. In order to meet the fourth industrial revolution, the manufacturing industry is innovatively changed by the fusion of intelligent information, pattern recognition, data mining, machine learning, big data, the Internet of things and manufacturing technology. The intelligent information technology is combined with the intelligent manufacturing technology, not only in the logistics processes of production, logistics, distribution, consumption and the like, but also in the equipment process and the quality control process. For efficient development of systems, data mining analysis based on manufacturing big data, design and analysis of intelligent systems, and basic technology reserves such as manufacturing security and block chain networks, manufacturing service technologies, and the like are required. The production system changes due to aging population and population reduction, and a heterogeneous big data acquisition and preprocessing technology, a big data distributed storage and management technology and a big data analysis technology are utilized in the big data decision mining process of the manufacturing industry. To process unstructured manufacturing data into process structured data, process-based distributed integration is performed. In order to effectively process the acquired unstructured big data, it is necessary to study the development of fusion technology. In addition, in the financial field, the block chain technology is applied to the manufacturing industry and related processes, and the sharing and the safety of the workflow transaction distributed accounts are also ensured. In the internet of things, intelligent manufacturing makes transparent, extensible, and secure processes possible with blockchains, and changes the way network organizations change from centralized to distributed.

Block chains are classified into public, hybrid, federation, and private chains. These blockchain technologies are increasingly being applied to energy, logistics, distribution, finance, medicine, automotive and public services. IBM developed a project that created a large platform on which various blockchain platforms were interconnected and integrated. Such an integrated blockchain platform may connect blocks of transactions, establish them in a distributed manner, and share transactions on a block-by-block basis. The process of mining using transactions that occur in real time makes it possible to analyze large data, draw meaningful rules, and make reasonable decisions. In intelligent manufacturing, the technology can effectively save cost, improve productivity and quality control: the distribution and manufacturing process comprises product balance, warehouse loading, warehouse balance, wholesale balance and organization balance. From an information flow perspective, a trade transaction includes manufacturing production, production implementation, wholesale implementation, and wholesale implementation. In the element identification process, the production time can be shortened, the distribution time can be shortened, the optimized plan can be established and the time and the cost can be saved by redistributing the low-efficiency process. In the manufacture and distribution of products, the inventory can be reduced, thereby saving the cost, reducing the price of the products and increasing the sales. Accordingly, consumer satisfaction and market share may be improved.

The goal of factory automation systems is to minimize human resources and increase the efficiency of a process unit. It is therefore based on the high functionality and accurate processing capacity of automated machines and focuses on improving productivity and quality. In a more comprehensive concept, it utilizes business management models such as enterprise resource planning, customer relationship management and decision-making to integrate and systematize information. These models make it possible to predict demand and respond to production and supply plans efficiently and in a timely manner. However, the system is based on a bottom-up messaging system and generates different response periods based on demand. Therefore, it has a limitation in meeting the rapidly changing small lot production requirements. To overcome this problem, pull-based foyota production systems have been developed that target zero inventory, zero defects, and flexible production. At the national level, germany has established industry 4.0 for the cooperation of governments, companies and academia. In the united states, the common electrical based ICT enterprise has developed collaboration. Intelligent manufacturing development is rising at the government or corporate research level. In terms of implementation, smart manufacturing is based on industrial robot technology, requiring multiple device controllers and internet of things technology that supports data acquisition. In addition, tools such as Hadoop have been used to analyze and process the collected data. On the basis of statistical analysis of the preprocessed data, the further improved artificial intelligence analysis method can be applied to the control and development of the production line. Artificial intelligence with improved learning capabilities would facilitate the operation of stable intelligent manufacturing equipment, thereby greatly improving the quality and productivity of the product.

With the development of the fusion technology and the change of consumption trend, each industry increasingly pays more attention to personalized production, and product purchase relates to factors such as raw materials, native countries, production date and distribution channels. With the popularity of the internet, consumers can conveniently collect product information, and their needs have become more diverse. As a result, new products are becoming shorter and shorter in life cycle and release cycle, and consumers are becoming more and more demanding. Manufacturing is transitioning from existing mass-customized production to individualized production with demand analysis and trend prediction. The personalized production aims at creating new service value and providing differentiated service through cooperation of different industries. It also has the characteristics of high quality, low cost, product diversity, service practicability and the like, and provides wide product choices for consumers. People are more and more concerned about the element identification process of applying the fusion technology to the personalized production mining technology. In the element identification process, various technologies such as internet of things, cloud computing, artificial intelligence, big data, data mining and the like are required to be performed according to stages in the whole process from planning to selling. The element recognition process is a structure related to all objects (such as health, raw materials, energy, parts, machines, and health care) related to manufacturing in various fields. It also requires an environment for cooperative work through various connections: human to object, object to machine, machine to human, and industrial to industrial connections. This helps to maximise process flexibility and enables it to cope with variations in manufacturing processes, market demands etc. The element recognition process generates different analysis results based on the collected data. It needs to cooperate through a large network and transparent and reliable data. The blockchain based intelligent manufacturing process completely records data through distributed ledger and blockchain distributed consensus. With the development of the fusion technology, the amount of information used by consumers in selecting commodities is larger and higher, and the expectation value of the consumers is also higher, and correspondingly, the block chain technology is provided with a mining process based on distributed consensus and distributed ledgers. The flow is through the use of distributed accounts to prevent errors or pinching that may occur during planning, design, production, distribution, and sale. The distributed ledger is one of the basic concepts of blockchains, and is also a record of consensus achieved by participants. This is a method of managing product history by applying blockchain distributed ledgers and blockchain distributed consensus to mining business processes. Distributed consistency is a method of obtaining consistency among participants with respect to particular data to achieve greater reliability. In smart manufacturing, a distributed ledger displays characteristics of sequence data, and data is displayed chronologically and continuously according to the state of a product. The condition of the product changes continuously along with the change of the manufacturing process, and the product is made into blocks and recorded in the distributed ledger. For example, the distributed ledger can provide data about products to which similar raw materials or manufacturing processes are applied through mining and correlation analysis of data visualizations. The distributed ledger provided to the consumer should utilize association mining analysis and data visualization to help the consumer make use of in the actual purchase. In addition, the mining business process proposed in the intelligent manufacturing can be applied to improving the reliability and flexibility of the ledger. This allows transparent information to be provided to the consumer and improves the reliability of mining the business process. Based on the industrial and technical characteristics, how to apply the block chain technology to the intelligent manufacturing topic mining field is a problem worthy of research and solution.

Disclosure of Invention

In order to solve the problems, the invention provides an intelligent manufacturing element identification method based on a block chain network, and meanwhile, the analysis and mining of intelligent manufacturing big data and the safety of data transmission and storage are considered, so that the intelligent manufacturing level is improved.

The technical scheme of the invention comprises the following steps:

1) and carrying out frequency domain transformation on the time domain data acquired by the intelligent wearable equipment by adopting short-time Fourier transformation, and carrying out data analysis of a frequency domain.

2) Comparing the waveforms of the initial normal operation of the equipment and the current operation of the equipment, and capturing abnormal production elements.

3) And establishing an intelligent equipment network based on a block chain technology to ensure the security of industrial secrets such as identified manufacturing elements, employee life log data and the like.

4) Federation members (referring to miners or billers in the federation chain) participate in a distributed consensus of creating new blocks to produce new data blocks.

5) And constructing a transaction information ledger in the intelligent manufacturing element identification by adopting a distributed ledger technology, and carrying out joint recording and management on the transaction information through a distribution mode based on a peer-to-peer network.

6) The problem that data cannot be analyzed due to the fact that limited view data are analyzed by using a block chain network type transaction in a general data mining method and the problem that data in a block chain network cannot be analyzed in the general data mining method are solved based on the theme packaging method, and the natural defect can be overcome by the theme packaging method.

The short-time Fourier transform is adopted in the step 1) to carry out frequency domain transformation on the time domain data collected by the intelligent wearable equipment, and the reason and the method for carrying out frequency domain analysis are as follows:

with the popularization of intelligent equipment, various sensor information can be obtained from users, and process monitoring, error prediction and human resource dynamic conditions are deeply analyzed. Therefore, based on dynamic situation management, the production line fault can be predicted, the health information of production human resources can be extracted, and personalized health services can be provided. In particular, human resource sanitation services can help improve welfare, working conditions, and production efficiency. Detection sensors used for manufacturing process monitoring and medical services vary and have a wide range of connections. Typical sensors include sound and acceleration sensors mounted on the production equipment. If such a sensor is used to analyze the vibration waveform, a constant and repeating waveform pattern is observed. Therefore, equipment failure can be captured according to wave deformation, and some errors of large-scale connection production equipment can be predicted so as to select a maintenance object and optimize operation. Also, accelerometer sensors connected to human resources can find their movements and model the classification of the working conditions and the detection of safety conditions. In response to various information input by the sensor, the present invention employs a highly versatile Fourier transform that is capable of analyzing the conditions of the device and body motion. The movement pattern and posture of the user are discovered and analyzed by using the structured data measured by the sensor. In addition, the dynamic energy released during exercise, rest and exercise can also be measured. The input signal of the fourier transform algorithm used for preprocessing fluctuates over time. Thus, a particular form of frequency is easily analyzed, and a change in motion can be detected from the input of the sensor. The fast fourier transform, which is commonly used for frequency analysis, can analyze the correlation in time. Therefore, in order to perform frequency analysis, it is necessary to use a short-time fourier transform in practical implementation. The transformed short-time Fourier transform decomposes time according to a certain unit, and performs frequency domain analysis. The short-time fourier transform is shown below:

wherein ST represents a start time in the time period, ET represents an end time of the time period; t represents an independent variable time; f represents a frequency; i denotes the imaginary part of the imaginary number.

Comparing the waveforms of the initial normal operation of the equipment and the current operation of the equipment in the step 2) to capture abnormal production elements, wherein the adopted method comprises the following steps:

calculating an error value E of the waveform of the initial normal operation of the equipment and the current operation of the equipment by adopting the following formula:

when the Fourier transform of the separated frequency bins is divided by N, i denotes each frequency bin, N_iIs the first normal value input during normal operation of segment i. S_iRepresenting the frequency of the i-th part of the current input. The greater the value of E measured in real time, the higher the probability of observing an abnormal condition. Given the scale allowed by quality control is G, the value of G is constantly reduced to get closer to the production concept based on predictive maintenance, which means that the staff in working order to adjust G is subjected to the actual maintenance work frequency. Meanwhile, the production human resources have various motions and motions, n sections are needed to classify the frequencies, and the E value is used as an evaluation index of the similarity. Comparing the frequency of the unit time input with its past valueAnd then grouped according to the similarity. Accordingly, various kinds of classifications can be made, such as a fixed posture, a horizontal movement, a stair movement, a fast movement state, and the like of the employee.

And 3) establishing an intelligent equipment network based on a block chain technology to ensure the security of industrial secrets such as identified manufacturing elements, employee life log data and the like, and being characterized in that:

aiming at the problem of poor safety of intelligent equipment, a network based on a block chain technology is established. Block chains are classified into public, hybrid, federation, and private chains. The public chain is a completely open block chain, and large users such as virtual coin users participate in distributed accounts and distributed consensus. The alliance chain is a block chain in which alliance members participate in distributed accounts and distributed consensus, and is characterized by identifying users. A private blockchain is a blockchain that allows a small fraction of users to participate in distributed ledgers and distributed consensus. The intelligent device contains activity information of enterprises and individuals, and therefore the invention uses a block chain of alliances taking participants as management subjects, and only allowed users can access data.

smart devices continuously generate and collect large amounts of data. Due to the structural problems of the block chain technology, the block chain technology hardly contains massive data. To overcome the structural problem, the present invention uses block-chain technology that includes side chains. The side chain utilizes the characteristic that the traditional block chain is not operable, and can realize the creation of data storage space in the block chain. The method stores the data of the intelligent device in a single database, and the data mapping occurs in a side-chain transaction of a data block. Accordingly, data can be searched quickly. In the data block organization of the intelligent wearable device based on the side chain, workers form a union as a client and a manager with distributed consensus. The data block includes the creation date, hash information of its concatenated block, and side chain information. In a data block, the side chain has a set of multiple transactions. The transaction set includes the staff's intelligent device information, mapping information for the collected data, events and their process information, and the number of distributed consensus participants.

In the step 4), the coalition members participate in creating a distributed consensus of the new block to generate a new data block, and the specific strategy is as follows:

the alliance has human resources related to production, and distributed consensus is achieved through the agreement of most alliance participants. The distributed consistency of creating new blocks will occur when events occur or in units of work processes. If an event occurs, processing is performed according to the event rule. A new tile is created by distributed consensus and linked to a chain of tiles. Unless an event occurs until the work process completes its plan, a new block is created at the endpoint through distributed consensus and then linked to the blockchain. The created blockchain and smart device data are stored in a physically distributed store in a distributed ledger to achieve consistency and data management. If the event happens, processing according to the rule: data created before the event occurs is integrated and then a data block is created. The created data blocks are saved to a distributed database through distributed consistency.

The step 5) adopts the distributed account book technology to construct a transaction information account book in the intelligent manufacturing element identification, and performs joint recording and management on transaction information through a distribution mode based on a peer-to-peer network, which is specifically as follows:

the blocks selected by the coherency algorithm are linked to each other to create a block chain. Each block has two types of hash information. One is the hash of the previous block header and the other is the hash of the transaction. For a transaction information ledger in intelligent manufacturing, a distributed ledger technology is adopted, and the transaction information is jointly recorded and managed in a distribution mode based on a peer-to-peer network. In a blockchain network, the intelligent manufacturing process determines details of data read points, data transfer or processing based on P2P, and the like. Due to its original management, smart manufacturing processes may suffer from lack of valid statistics, neglect of supervision, inconsistency of physical and soft data, etc. The conventional physical intelligent manufacturing system has unclear information in the manufacturing process, and the proposed intelligent manufacturing process supports end-to-end tracking based on transaction data stored in a plurality of blockchains to prevent data loss at each step. The intelligent manufacturing process based on the block chain analyzes various data related to a tracing system, extension infrastructures of all bases and a staff working system by utilizing information exchange of real-time collected data. Accordingly, a block chain based distributed ledger mining technique provides innovative services in an intelligent manufacturing process.

The step 6) is based on a theme encapsulation method to solve the problem that the data cannot be analyzed due to the fact that the limited-view data is analyzed by using the transaction of the block chain network type in the general data mining method, and the method is specifically as follows:

data mining is the process of analyzing and exploring data to find meaningful rules and relationships from unstructured or structured data. The general data mining method can perform expression search on an intelligent manufacturing process to analyze data meaning, but it cannot obtain data analysis results in a data collection process. Particularly in the smart manufacturing element identification process, there is a problem in that data cannot be analyzed because limited-view data is analyzed using a blockchain network type transaction.

Therefore, the invention removes meaningless data variables and extracts main variables by a subject element packaging method of the affairs. The theme element packaging is a formatted statistical reasoning method for analyzing semantic environment from data, and can find potential data themes when block chain transaction data is applied. The subject matter element packaging process includes six steps, each step presenting a different data set for each manufacturing stage in a blockchain transaction:

step 1: collecting production element data in an intelligent manufacturing process by using intelligent devices such as wearable devices and intelligent sensors for building data blocks

Step 2: and carrying out data filtering and cleaning on the collected intelligent manufacturing production processes, and carrying out examination and verification on the collected data, wherein the examination and verification comprise data consistency checking, invalid value deletion, missing value supplement and the like.

And 3, step 3: stemming is a process of removing affixes to obtain roots. For the morphological root of a word, the stems do not need to be identical; mapping related words to the same stem can lead to satisfactory results even if the stem is not a valid root of a word. This step may perform data merging on synonyms in the production factor.

And 4, step 4: a matrix is created (process items) representing transactional data that may occur in the intelligent manufacturing process, including all elements that may occur in the intelligent manufacturing process, such as workers, equipment, product information, process steps, and process content.

And 5, step 5: and packaging the intelligent manufacturing theme elements, namely the (process) item matrix into the data blocks.

And 6, step 6: and classifying the data blocks according to the intelligent manufacturing production element word stems and connecting the data blocks into a block chain.

Topic element packaging is a formatted statistical reasoning method for analyzing semantic environment from data, and can find potential data topics when block chain data is applied.

The invention has the beneficial effects that:

the method can analyze and mine big data collected by intelligent equipment in intelligent manufacturing, extracts meaningful items in the manufacturing process as representative subjects, realizes element mining and identification of intelligent manufacturing, and simultaneously ensures the safety of industrial data transmission and management by applying a block chain technology. The method can be applied to intelligent manufacturing processes such as production quality prediction, trend prediction, production monitoring, fault diagnosis, data distortion analysis and the like, and is beneficial to improving the intelligent manufacturing level.

Description of the figures (tables)

FIG. 1 is a schematic diagram of a side-chain based data block organization in a smart device;

FIG. 2 is a schematic diagram of a process for creating and managing a federation chain center data block;

fig. 3 is a schematic diagram of a data transmission and storage method based on a block chain distributed ledger;

FIG. 4 is a diagram illustrating an element encapsulation process for a blockchain transaction;

fig. 5 is a schematic diagram of experimental results of effectiveness experiment 1 and experiment 2.

DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION

The invention is described in further detail below with reference to the figures and the embodiments.

The specific embodiment of the invention is as follows:

example 1

An intelligent manufacturing element identification method based on a block chain network is characterized by comprising the following steps:

(1) the method comprises the following steps of carrying out frequency domain transformation on time domain data collected by the intelligent wearable equipment by adopting short-time Fourier transformation, carrying out frequency domain data analysis, and carrying out frequency domain data analysis:

in order to perform frequency analysis, it is necessary to use a short-time fourier transform in actual implementation, and the short-time fourier transform after transformation is decomposed in time units and is subjected to frequency domain analysis, and the short-time fourier transform is expressed by the following equation:

(2) Comparing the waveforms of the initial normal operation of the equipment and the current operation of the equipment, and capturing abnormal production elements, wherein the specific method comprises the following steps:

when the Fourier transform of the separated frequency bins is divided by N, i denotes each frequency bin, N_iIs the first normal value, S, input during normal operation of the i segment_iRepresenting the frequency of the i-th part of the current input, the greater the value of E measured in real time, the higher the probability of observing an abnormal state. Frequency of unit time input and the samePast values are compared to be grouped according to the degree of similarity, and according to this, various classifications can be made, such as a stationary posture, a horizontal movement, a step movement, a fast movement state, and the like of the employee.

(3) An intelligent equipment network is established based on a block chain technology, the security of industrial secrets such as identified manufacturing elements, employee life log data and the like is ensured, and the network structure specifically comprises the following steps:

as shown in fig. 1, the present invention uses federation chain technology including side chains. The side chain stores the data of the intelligent device in a separate database, and data mapping occurs in side chain transactions of the data block. Accordingly, data can be searched quickly. According to the data block organization scheme based on the side chain in the intelligent wearable device, workers form a union as a client and a manager in distributed consensus to form a union network. The data block includes the creation date, hash information of its concatenated block, and side chain information.

(4) The strategy for a coalition member to participate in creating a distributed consensus of new blocks to generate new data blocks is designed as follows:

the alliance has human resources related to production, distributed consensus is achieved through the agreement of most alliance participants, and when an event occurs or a work process is taken as a unit, distributed consistency for creating a new block occurs; fig. 2 shows the creation and management flow of a new block in a federation chain, and if an event occurs, the process is processed according to the rules: data created before the event occurs is integrated and then a data block is created, and the created data block is saved to a distributed database through distributed consistency.

(5) The method comprises the following steps of constructing a transaction information account book in intelligent manufacturing element identification by adopting a distributed account book technology, and carrying out joint recording and management on transaction information through a distribution mode based on a peer-to-peer network, wherein the specific method comprises the following steps:

by using the intelligent manufacturing data transmission and storage method based on the block chain distributed account book shown in fig. 3, for the transaction information account book in the intelligent manufacturing, the distributed account book technology is adopted, and the transaction information is jointly recorded and managed in a distribution mode based on the peer-to-peer network.

(6) The problem that data cannot be analyzed due to the fact that a block chain network type transaction is used for analyzing data of a limited view in a general data mining method is solved based on a theme packaging method. The topic element packaging is a formatted statistical reasoning method for analyzing semantic environment from data, and can find potential data topics when block chain transaction data is applied. FIG. 4 illustrates an element encapsulation process for a blockchain transaction. The subject matter element packaging process includes six steps, each step presenting a different data set for each manufacturing stage in a blockchain transaction.

Example 2:

selecting a certain textile production line to carry out the effect test of the invention. The testing environment comprises 7 textile workers (respectively marked as W1-W7), wherein each person wears a happy 5S bracelet and is used for sensing the heart rate (reflecting the health condition and the working strength) and the step number (reflecting the health condition and the working strength) of the textile workers, each person distributes a smart phone for event and transaction processing confirmation, the model of the smart phone is Redmi Note8 Pro, and the memory is 6 GB. 7 textile workers play the role of workers and the role of managers at the same time, and participate in the system consensus process. And 3, installing flatness test sensors at product outlets of the assembly lines, detecting the quality of the produced cloth, and giving an alarm to a system when a flaw is found.

Based on the production system, the effectiveness and safety of the invention were tested separately, and the following experiments were performed:

effectiveness test 1: worker status monitoring

In the normal working process, the worker W2 takes off the intelligent bracelet at the time t 1. The system suggests the presence of worker condition anomaly-stop movement, heart rate 0.

Effectiveness experiment 2: product quality monitoring

In the normal working process, the product quality of the production line 2 is intentionally damaged at the time t2, the cloth is uneven, and a worker W3 is assigned to perform processing and affair confirmation.

Effectiveness experiment 3: statistics of production efficiency change

The production conditions of the textile production line in the traditional production mode (without carrying the system of the invention) and the production mode carrying the intelligent manufacturing element identification method system are compared.

Safety experiment 1: false transaction validation

A worker is designated to process a flaw in the production process, but the worker performs a transaction confirmation on the smart phone terminal without performing a relevant inspection process.

Safety experiment 2: malicious tampering, deletion of data

If a worker intelligent mobile phone terminal is illegally occupied or invaded, an illegal person tries to tamper and delete system data.

The results of the effectiveness tests 1 and 2 are shown in fig. 5. At time t1, the system detects that there is a significant abnormality in the heart rate of worker W2, which is reduced to 0, and can alert that worker W2 is abnormal in working condition. At time t2, the system detects that worker W3 is handling abnormal problems of products, the body load is increased, the heart rate is increased, and information uploaded by a flatness test sensor installed at a product outlet can prove that problems occur in a corresponding production line and are handled by worker W3. Effectiveness experiment 3 results are shown in table 1. In the new mode, the total production amount is slightly increased, the defective rate is obviously reduced, because quality abnormality can be found and processed in time, the fault and error production conditions are reduced, and the defective rate is reduced. In addition, the worker can optimize the production behavior of the worker according to the system prompt, the working accuracy is improved, the invalid movement is reduced, and the working strength is obviously reduced.

Table 1 effectiveness test 3 test results

TABLE 1

1. Is the sum of three production lines; 2. the average value of three production lines; 3. the average of the 8-hour exercise amounts of 7 workers.

The safety experiment 1 showed the following experimental results: because the abnormity is not processed, a flatness test sensor arranged at the outlet of the production line continuously uploads product abnormity alarm data, and a certain false affair is confirmed to be incapable of obtaining system consensus and not to be recorded into a system account book. The system sends an alarm for the abnormal working state of the worker, and continuously alarms the abnormal product quality until other workers smoothly treat the abnormal product quality and then stops the abnormal quality alarm.

The safety experiment 2 has the following experimental results: the reason why the system data is not tampered and deleted is that the intelligent manufacturing data transmission and storage method based on the block chain distributed account book is adopted, all system data are not stored in a single terminal, and the system data cannot be changed under the condition that the system consensus is not obtained.

The foregoing detailed description is intended to illustrate and not limit the invention, which is intended to be within the spirit and scope of the appended claims, and any changes and modifications that fall within the true spirit and scope of the invention are intended to be covered by the following claims.

Claims

1. An intelligent manufacturing element identification method based on a block chain network is characterized by comprising the following steps:

(1) carrying out frequency domain transformation on the acquired time domain data by adopting short-time Fourier transformation in the intelligent wearable equipment, and carrying out frequency domain data analysis;

(2) comparing the waveforms of the initial normal operation of the equipment and the current operation of the equipment, and capturing abnormal production elements;

(3) in order to ensure the security of industrial confidential data, an intelligent equipment network is established based on a alliance chain and a side chain technology, intelligent manufacturing data and data are stored in a side chain in a mapping mode, and hash information and side chain information of a data block are stored in the alliance chain;

(4) the alliance member participates in creating a distributed consensus of the new block to generate a new data block;

(5) a transaction information ledger in intelligent manufacturing element identification is constructed by adopting a distributed ledger technology, and the transaction information is jointly recorded and managed in a distribution mode based on a peer-to-peer network;

(6) finite view data generated by blockchain network type transactions is analyzed based on a subject encapsulation method.

2. The blockchain network based intelligent manufacturing factor identification method according to claim 1, wherein: the method for carrying out frequency domain data analysis on the time domain data collected by the intelligent wearable device by adopting short-time Fourier transform in the step (1) comprises the following steps:

in practical implementation, a short-time fourier transform is used for frequency analysis, and the transformed short-time fourier transform decomposes time according to a certain unit and performs frequency domain analysis, wherein the short-time fourier transform is represented by the following formula:

3. The blockchain network based intelligent manufacturing factor identification method according to claim 1, wherein: comparing the waveforms of the initial normal operation of the equipment and the current operation of the equipment in the step (2) to capture abnormal production elements, wherein the specific method comprises the following steps:

when the Fourier transform of the separated frequency bins is divided by N, i denotes each frequency bin, N_iIs the first normal value, S, input during normal operation of the i segment_iRepresenting the frequency of the i-th part of the current input, the greater the E value measured in real time, the higher the probability of observing an abnormal state, and the frequency of the input per unit time is compared with its past values to group them according to the degree of similarity.

4. The blockchain network based intelligent manufacturing factor identification method according to claim 1, wherein: in the step (3), an intelligent device network is established based on a block chain technology to ensure the security of industrial secrets such as identified manufacturing elements and employee life log data, and the intelligent device network specifically comprises the following steps:

the data of the intelligent device is stored in a single database by the side chain, data mapping occurs in side chain transactions of data blocks, data block organization schematic based on the side chain is established in the intelligent wearable device, workers form a alliance as distributed consensus clients and managers to form an alliance network, and the data blocks comprise creation dates, hash information of connection blocks of the data blocks and side chain information.

5. The blockchain network based intelligent manufacturing factor identification method according to claim 1, wherein: in the step (3), the strategy design that the coalition members participate in creating the distributed consensus of the new block to generate the new data block is as follows:

the alliance has human resources related to production, and by the agreement of most alliance participants, a distributed consensus is achieved, namely data created before an event occurs are integrated, then data blocks are created, the created data blocks are stored in a distributed database through distributed consistency, and when the event occurs or a work process is taken as a unit, the distributed consistency for creating a new block occurs.

6. The blockchain network based intelligent manufacturing factor identification method according to claim 1, wherein: the step (6) removes meaningless data variables and extracts main variables by a subject element encapsulation method of the transaction, and comprises the following specific steps:

6.1 collecting production element data in the intelligent manufacturing process by using intelligent equipment such as wearable equipment and intelligent sensors to construct a data block;

6.2, filtering and cleaning the collected data of the intelligent manufacturing production process, and checking and verifying the collected data, wherein the checking comprises checking data consistency, deleting invalid values and supplementing missing values;

6.3 extracting the word stem, and merging data of synonyms in the production elements;

6.4 creating a [ process items ] matrix comprising all elements that may occur in the intelligent manufacturing process, such as workers, equipment, product information, process steps and process content;

6.5, packaging the intelligent manufacturing theme elements, namely the (process items) matrix into the data blocks;

6.6 classifying and connecting the data blocks into a block chain according to the intelligent manufacturing production element word stems.