CN112508382A - Industrial control system based on big data - Google Patents

Industrial control system based on big data Download PDF

Info

Publication number
CN112508382A
CN112508382A CN202011395699.8A CN202011395699A CN112508382A CN 112508382 A CN112508382 A CN 112508382A CN 202011395699 A CN202011395699 A CN 202011395699A CN 112508382 A CN112508382 A CN 112508382A
Authority
CN
China
Prior art keywords
data
big data
layer
target
module
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Withdrawn
Application number
CN202011395699.8A
Other languages
Chinese (zh)
Inventor
黄瑜丹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenyang Sport University
Original Assignee
Shenyang Sport University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenyang Sport University filed Critical Shenyang Sport University
Priority to CN202011395699.8A priority Critical patent/CN112508382A/en
Publication of CN112508382A publication Critical patent/CN112508382A/en
Withdrawn legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2465—Query processing support for facilitating data mining operations in structured databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06313—Resource planning in a project environment
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/10—Office automation; Time management
    • G06Q10/103—Workflow collaboration or project management

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • General Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Educational Administration (AREA)
  • Databases & Information Systems (AREA)
  • Mathematical Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Pure & Applied Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Development Economics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Algebra (AREA)
  • Fuzzy Systems (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biodiversity & Conservation Biology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to an industrial control system based on big data, which comprises a data acquisition module, an inquiry service module, a snapshot service module, a big data processing module, a terminal module and a cloud module, wherein big data dimensionality and application scene dimensionality are set, different elements are set in each dimensionality, and the three-dimensional spaces of different layers of the different elements are integrated to construct an integral system structure of the big data dimensionality and the application scene dimensionality. By the big data processing method, internal cause-and-effect relationships and key factors of the complex system are found out, and the logical relationship and the interaction degree among the factors in the complex system are analyzed, so that technical support is provided for the industrial control system.

Description

Industrial control system based on big data
Technical Field
The invention relates to the technical field of big data, in particular to an industrial control system based on big data.
Background
In recent years, informatization and industrialization are deeply fused to become a great trend, mass production data have important values on the real-time monitoring of the enterprise process, the evaluation of production process, the business operation decision and the like, and the method is a powerful weapon for upgrading, increasing efficiency and upgrading of the enterprise, has the capability of efficiently storing and processing the mass data, and becomes a core competitiveness of modern industrial enterprises. Industrial process data may be referred to as big data because it is difficult to perform capture, analysis, and visualization. In addition, it is a large data with high capacity and low density value, which is helpful for the enterprise to fully understand the process. As a comprehensive data platform of the process industry automatic control system, the real-time database plays a key role in the integration, storage and processing links of enterprise production information and is a nuclear primary infrastructure of process industry informatization. The real-time database has a strong multi-source heterogeneous data acquisition interface, high-efficiency historical data compression and condition retrieval capability, and high reliability and high practicability of data and service. Compared with a relational database, the real-time database and the relational database can better integrate information resources of various heterogeneous systems, comprehensively integrate production process information of enterprises, and provide efficient data service for upper-layer applications such as benefit evaluation, process optimization, management decision and the like.
Traditional data storage, centralized data preprocessing, and modeling approaches face the dilemma of computational inefficiency or memory scarcity. To provide a fast and low cost solution, there is a need for a parallel and distributed based big data industrial control system that leverages distributed computing resources and translates the heavy computational burden into parallel small scale processing. Furthermore, the conventional data driving method should be converted into a scalable form to handle the increasing data size.
For process industry application scenarios, conventional process monitoring information is presented as continuously collected time series data and discretely occurring event series data. Because the monitored system variable is in continuous change, a certain specific variable value can only represent the state of the data acquisition moment, and therefore, the integrity of data semantics is ensured by carrying a time attribute. Meanwhile, the upper application processing must be completed as far as possible before the transaction deadline, so that the collected data used in the processing process is ensured not to have obvious difference from the current real state, and a timely and correct decision is formed. Discrete alarm incidents generated in the industrial process need to be processed in real time according to the priority of the discrete alarm incidents, so that serious production accidents are avoided. The invention provides a control system established aiming at distributed parallel probability of big data, and an application module for process monitoring and intelligent prediction is further developed in the system.
Disclosure of Invention
The invention provides an industrial control system based on big data, which can comprehensively integrate the production process information of enterprises and provide high-efficiency data service for upper-layer applications such as benefit evaluation, process optimization, management decision and the like.
An industrial control system based on big data comprises a data acquisition module, an inquiry service module, a snapshot service module, a big data processing module, a terminal module and a cloud module; the big data processing module comprises data integration and processing, data modeling and analysis, data decision and data control, and specifically comprises the following processing steps:
s1, a calculation manager is responsible for initializing calculation tasks, and the initialization work comprises the following steps: generating a calculation object to uniformly manage the full life cycle process of the calculation task, and acquiring and setting relevant parameters of the calculation task according to the configuration information;
s2, respectively setting clustering labels and clustering data of nodes for a user presentation layer, a service analysis layer, a data storage layer, a network transmission layer and an intelligent sensing layer;
s3, optimizing node data;
s4, refining different elements of the user presentation layer, the service analysis layer, the data storage layer, the network transmission layer and the intelligent perception layer, and respectively listing the different elements as A1,A2,….,A12；
S5, element AiThe influence between the other elements is graded and scored, and is marked as RiThe scoring rules are as follows: score 0 indicates no effect between the two factors, score 1 indicates very little effect, score 2 indicates little effect, score 3 indicates medium effect, score 4 indicates high effect, score 5 indicates very high effect, a sufficient number of experts are invited to score, the results are averaged to obtain a score set Ri；
S6, optimizing the industrial control system through multi-dimensional intelligent cooperation and optimization technology manufactured in the big data and cloud computing environment based on the cloud service planning of the industrial big data;
defining a collection of industrial items as U ═ U1,u2,…,umN is an integer, u1,u2,…,umRespectively being a sub-item in an item, the drive control item parameter of the control action is Ci(i＝1,2,…,n)，
Ci,u1＝{A1,u1,A2,u1…,A12,u1}
Ci,u2＝{A1,u2,A2,u2…,A12,u2}
Figure BDA0002814995600000021
Ci,um＝{A1,um,A2,um…,A12,um}
Ci,u1,Ci,u2,…,Ci,umRespectively representing control actions to sub-items u1,u2,…,umDrive control item parameter of (1);
s7, calculating control behavior and driving control item parameters C by adopting a correlation analysis methodi,umDegree of association of (E) is expressed as ∈iIf epsiloni>Beta, then the drive control item parameter is called Ci,umIs the main driver sequence, beta is the relevance threshold,
Figure BDA0002814995600000031
sim (u1, u2) is the sub-item u1,u2The degree of similarity of (a) to (b),
Figure BDA0002814995600000032
represents the average score, R, of all the sub-items scoredu1，Ru2Respectively represent a pair sub-item u1,u2Scoring of (4);
s8, predicting a system behavior series according to the driving factors;
for a given multivariable control system, there often exists a complex non-linear relationship between its sequence of drive factors and the sequence of system behavior, in terms of εi>β, determining a new sequence of primary drivers, note
Figure BDA0002814995600000033
The original sequence of drive factors is
Figure BDA0002814995600000034
After one driving factor, the sequence is generated by accumulating the driving factors once
Figure BDA0002814995600000035
For a sequence of system behaviors, there are:
Figure BDA0002814995600000036
wherein d isi+1For driving control item parameter CiFor determining the mechanism of action of the driver on the system behaviour, bi、bi+1Respectively starting time and ending time of the driving factors, wherein gamma is a power exponent and reflects the nonlinear action relation of the driving factors on the system behavior;
and S9, analyzing the rationality and significance of the prediction result according to the system behavior prediction value, and optimizing and adjusting the existing problems by adopting various evaluation indexes.
Further, in step S4, the user presentation layer, the service analysis layer, the data storage layer, the network transport layer, and the intelligent sensing layer respectively include:
A1: the client requirements are used for identifying the client requirements, intelligently predicting the client requirements, intelligently analyzing the client behaviors and public opinions and intelligently acquiring the clients;
A2: the big data innovation chain is based on a big data collaborative interaction design platform, and product design, process design, collaborative design and intelligent research and development are innovatively designed by using big data;
A3：R&d, performing cross-industry and cross-region industrial chain cooperative operation based on big data;
A4: the big data management chain is used for enterprise strategic planning and market intelligent prediction, equipment intelligent management, material intelligent management, logistics intelligent optimization and warehousing intelligent optimization;
A5: the industrial operation management chain is used for acquiring data of an enterprise operation process;
A6: the big data manufacturing chain is used for intelligently sensing, intelligently predicting and analyzing based on a networked manufacturing resource collaboration platform of big data;
A7: the industrial manufacturing chain is used for acquiring data of enterprise equipment and production manufacturing environment;
A8: the big data service chain is used for intelligently extracting the intention of a customer, analyzing the behavior and habit of the customer, analyzing the trend of service requirements and intelligently interacting with the customer on the basis of a product life cycle analysis management platform of big data;
A9: the industrial service chain is used for acquiring data of the enterprise service chain;
A10: the big data support chain is used for processing and analyzing big data;
A11: big data processing for data analysis and optimization, network mining and decision making;
A12: the big data platform provides a support technology, a shared resource, an access system and a cloud service platform.
Furthermore, the data acquisition module is used for acquiring data of an equipment layer, a process layer, a production line layer, a vehicle interlayer, an enterprise layer and a cooperation layer respectively.
Further, the big data processing module performs dynamic resource optimization on the tracking feedback information of the multiple sensors, and the specific optimization process is as follows:
defining time k the movable sensor (1,2 …, i, …, N)s) As a function of schedule
Figure BDA0002814995600000041
Defining the state of the target at the time k as
Figure BDA0002814995600000042
Where x (k), y (k) represent the position of the target in the x and y directions at time k, respectively,
Figure BDA0002814995600000043
representing the velocity of the target in the x and y directions at time k, respectively, the state transition equation for the target can be described as
x(k+1)＝Fkx(k)+v(k)；
Wherein x (k +1) represents the position of the target in the x direction at the time k +1, and FkSelecting different models for a target state transition matrix at the moment k according to different motion modes of the target (such as uniform motion, uniform acceleration motion and turning motion), wherein v (k) is process noise at the moment k, and the Gaussian distribution with the mean value of 0 and the variance of Q is obeyed;
suppose NsThe angle measurement sensor carries out cooperative positioning tracking on the target, and the position and the speed of the target in the two-dimensional space are estimated by using the arrival angle measured by the sensor. In order to meet the observability of the tracking problem of multiple passive sensors, at least more than two passive sensors are required to track one target.
Azimuthal measurement z of sensor i to targeti(k) Is composed of
Figure BDA0002814995600000044
vi(k) The measurement noise of sensor i follows a gaussian distribution with mean 0 and variance Q.
The azimuth angle measurement of the multisensor system is z (k) ═ z1(k),z2(k),…,zNs(k)]Is modeled as
Figure BDA0002814995600000051
And after obtaining new target measurement information, the sensor obtains the state estimation of the target by weighted least squares. Positioning the target by using the relation between the target azimuth information measured by each sensor and the target position through direction finding positioning:
Figure BDA0002814995600000052
let xtan (z)i) -y is taken as an equivalent observation,
Figure BDA0002814995600000053
is equivalent noise, therefore, the multi-sensor measurement equation can be converted into a matrix form
Z is HX + epsilon, wherein,
Figure BDA0002814995600000054
through the measurement characteristics of the multiple sensors, the multiple sensors are controlled to move to the optimal observation positions, operation set scheduling and information transmission are carried out, a signal transmission model of the multimedia sensor network is built, data information of the sensor network is scheduled, and signal modulation output of coherent points of the multimedia sensor network is obtained.
Further, step S2 sets a cluster label and cluster data of the node for the user presentation layer, the service analysis layer, the data storage layer, the network transmission layer, and the intelligent sensing layer, and step S3 optimizes the node data, which specifically includes the following steps:
suppose that a user presentation layer, a service analysis layer, a data storage layer, a network transmission layer and an intelligent perception layer are provided with M nodes which communicate through a network, and each node is provided with N nodesjSet of data points { xji|i＝1,…,NjAnd the total number of data stored in the network is N ═ Σ NjIn which N isjN data points are samples of a random variable X with a probability distribution p (X), each node j having N of the random variable XjThe data points are aggregated into at most K different classes, each data point is allocated with a clustering label omega, the probability distribution of the clustering labels is p (x), K belongs to the sampling of random variables Y of {1, …, K }, and each node aims to search the clustering label and the clustering data of a condition model p;
minimizing network routing information between X and Y by:
Figure BDA0002814995600000061
Figure BDA0002814995600000062
where F (ω) is the cross entropy loss function, p (k) is the conditional model function of the node, yjiIs the label of the ith data point of node j, G (y)ji) Is an indicator function; each node is in cooperative communication with adjacent nodes, local optimal solution is obtained by minimizing cross entropy loss function, the size of the model is compressed, and overfitting is avoided.
Furthermore, the data acquisition module receives industrial data of data sources such as industrial field devices and third-party control systems through acquisition services and uploads the industrial data to the data storage unit in real time.
Further, the terminal module comprises a client UI module and a visualization module, and the client UI module is suitable for collecting terminal user information.
Furthermore, the cloud module comprises a signal receiving and processing module, and the signal receiving and processing module is suitable for receiving and processing the terminal user information and the related big data collected by the client UI module.
Further, the query service module is used for uniformly scheduling and managing all query requests, executing query tasks according to a specified algorithm and priority, and returning query results to the client.
Furthermore, the snapshot service module provides real-time data services to the outside by using the process variable as a structural unit and using the memory mapping file as a cache database of the carrier, including real-time query and change subscription of the current state of the process variable.
The method is based on industrial application scene planning, combines other dimensions of a big data analysis model, forms a reference system of big data application in product service systems in different industrial stages, and achieves the technical effects of customer demand identification, intelligent customer demand prediction, intelligent customer behavior and public opinion analysis, intelligent customer acquisition, intelligent customer demand analysis, intelligent customer demand mining, intelligent interaction with customers and the like.
The big data processing module comprises data acquisition and exchange, data integration and processing, data modeling and analysis, and data decision and control. The invention sets the big data dimension and the application scene dimension, sets different elements in each dimension, and constructs the whole system structure of the big data dimension and the application scene dimension by integrating the three-dimensional spaces of different layers of different elements. Therefore, by the big data processing method, the internal cause-and-effect relationship and key factors of the complex system are found out, and the logical relationship and the interaction degree among the factors in the complex system are analyzed, so that technical support is provided for the industrial control system.
According to the method, each node is in cooperative communication with the adjacent nodes, and the local optimal solution is obtained by minimizing the cross entropy loss function, so that the effect of model order adjustment is achieved, the size of the model is reduced, and the problem of overfitting is avoided.
Detailed Description
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. In order to make those skilled in the art better understand the technical solutions of the embodiments of the present invention, the following will clearly and completely describe the technical solutions of the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
And the data acquisition module receives industrial data of data sources such as industrial field equipment and a third-party control system through the acquisition service and uploads the industrial data to the data storage unit in real time. The acquisition service provides basic functions of acquisition on demand, protocol mapping, breakpoint resume, range conversion and the like.
The query service module is a uniform dispatcher and an agent for real-time query, change subscription and historical query requests, and is the only query interface of the sampling result data. The query service is used for uniformly scheduling and managing all query requests, executing query tasks according to a specified algorithm and priority, and returning query results to the client.
And the snapshot service module is used for providing real-time data service for the outside by taking the process variable as a structural unit and taking the memory mapping file as a cache database of a carrier, and comprises real-time inquiry and change subscription of the current state of the process variable.
In the intelligent manufacturing system, the architecture is mainly divided into three parts, from bottom to top, which are basic commonalities, key technologies and industrial applications. Basic commonality mainly refers to the intelligent and sustainability basis supporting the whole system, including universality, security, reliability, detection, evaluation. The key technology is divided into four parts: intelligent equipment, intelligent factories, intelligent services, intelligent enabling technologies, and industrial networks. The industrial big data is a part of an intelligent manufacturing technology and plays an important role in popularization of other technologies. The product service system is embodied as intelligent equipment service, factory intelligent product service, operation intelligent innovation service and intelligent network communication service in an industrial intelligent network. Industrial applications refer to the application of intelligent manufacturing techniques in key industrial areas. The industrial control system of the invention is an organic whole with input and output sustainability, has specific functions and is composed of various interdependent and interactive subsystems. The industrial control system based on big data is formed by a plurality of subsystems together, and the industrial control system based on big data specifically comprises: industry, big data and application scenarios.
The big data processing module comprises data acquisition and exchange, data integration and processing, data modeling and analysis, data decision and data control. A big data dimension and an application scenario dimension are set, each dimension corresponding to a different level. In the big data dimension, a user display layer, a business analysis layer, a data storage layer, a network transmission layer and an intelligent perception layer exist, each layer comprises different elements, and three-dimensional spaces of different layers of the different elements are integrated to construct an integral system structure of the big data dimension and the application scene dimension. Thus, enterprises need to consider exploring and evaluating which elements are the most core elements to build through relationships between the various levels of elements. In order to find out internal causal relationships and key factors of the complex system, the method adopts the following method to analyze the logical relationship and the interaction degree among the factors in the complex system.
1) The computing manager is responsible for initializing computing tasks, and the initialization work comprises the following steps: and generating a calculation object to uniformly manage the full life cycle process of the calculation task, and acquiring and setting relevant parameters of the calculation task according to the configuration information.
2) Respectively setting clustering labels and clustering data of nodes for a user presentation layer, a service analysis layer, a data storage layer, a network transmission layer and an intelligent sensing layer;
suppose that a user presentation layer, a service analysis layer, a data storage layer, a network transmission layer and an intelligent perception layer are provided with M nodes which communicate through a network, and each node is provided with N nodesjSet of data points { xji|i＝1,…,NjAnd the total number of data stored in the network is N ═ Σ NjIn which N isjN data points are samples of a random variable X with a probability distribution p (X), each node j having N of the random variable XjThe data points are gathered into at most K different classes, each data point is allocated with a clustering label omega, the probability distribution of the clustering labels is also p (x), K belongs to the sampling of random variables Y of {1, …, K }, and each node aims to find the clustering label and the clustering data of the condition model p.
3) Optimizing node data
In a big data network, an intelligent sensing layer is a 'manager' of the whole network, namely a central node and is responsible for receiving and sending global information to other nodes. Once the central node fails or crashes due to an attack, the entire network will stop working, and the central node will need special maintenance. Since all working nodes need the central node to send and receive information, the links near the central node can be very congested. This may cause problems such as a communication time extension and a packet loss, and further, the efficiency of processing data by the network may be reduced. Meanwhile, since the centralized network needs to set a route so that each working node knows a path to the central node, each node needs to store information about the route, which increases resource overhead of the node, and routing information needs to be recalculated once a link is damaged due to some node failures.
Minimizing network routing information between X and Y by:
Figure BDA0002814995600000081
Figure BDA0002814995600000082
where F (ω) is the cross entropy loss function, p (k) is the conditional model function of the node, yjiIs the label of the ith data point of node j, G (y)ji) Is an indicator function.
Each node is in cooperative communication with adjacent nodes, so that a local optimal solution is obtained by minimizing a cross entropy loss function, a model order adjusting effect is achieved, the size of the model is reduced, and the problem of overfitting is avoided.
4) Refining different elements of a user presentation layer, a service analysis layer, a data storage layer, a network transmission layer and an intelligent perception layer, and respectively listing the different elements as A1,A2,….,A12。
A1: the client requirements are used for identifying the client requirements, intelligently predicting the client requirements, intelligently analyzing the client behaviors and public opinions and intelligently acquiring the clients;
A2: the big data innovation chain is based on a big data collaborative interaction design platform, and product design, process design, collaborative design and intelligent research and development are innovatively designed by using big data;
A3：R&d, performing cross-industry and cross-region industrial chain cooperative operation based on big data;
A4: the big data management chain is used for enterprise strategic planning and market intelligent prediction, equipment intelligent management, material intelligent management, logistics intelligent optimization and warehousing intelligent optimization;
A5: industrial operation management chain for collecting enterprisesRunning the data of the flow;
A6: the big data manufacturing chain is used for intelligently sensing, intelligently predicting and analyzing based on a networked manufacturing resource collaboration platform of big data;
A7: the industrial manufacturing chain is used for acquiring data of enterprise equipment and production manufacturing environment;
A8: the big data service chain is used for intelligently extracting the intention of a customer, analyzing the behavior and habit of the customer, analyzing the trend of service requirements and intelligently interacting with the customer on the basis of a product life cycle analysis management platform of big data;
A9: the industrial service chain is used for acquiring data of the enterprise service chain;
A10: the big data support chain is used for processing and analyzing big data;
A11: big data processing for data analysis and optimization, network mining and decision making;
A12: the big data platform provides a support technology, a shared resource, an access system and a cloud service platform.
5) For element AiThe influence between the other elements is graded and scored, and is marked as RiThe scoring rules are as follows: a score of 0 indicates no effect between the two factors, a score of 1 indicates very little effect, a score of 2 indicates small effect, a score of 3 indicates medium effect, a score of 4 indicates high effect, and a score of 5 indicates very high effect. Inviting a sufficient number of experts to score, and averaging the results to obtain a score set Ri。
6) And (3) optimizing the industrial control system by using a multi-dimensional intelligent cooperation and optimization technology manufactured in the environment of big data and cloud computing based on the cloud service planning of the industrial big data.
Defining a collection of industrial items as U ═ U1,u2,…,umN is an integer, u1,u2,…,umRespectively being a sub-item in an item, the drive control item parameter of the control action is Ci(i＝1,2,…,n)，
Ci,u1＝{A1,u1,A2,u1…,A12,u1}
Ci,u2＝{A1,u2,A2,u2…,A12,u2}
Figure BDA0002814995600000107
Ci,um＝{A1,um,A2,um…,A12,um}
Ci,u1,Ci,u2,…,Ci,umRespectively representing control actions to sub-items u1,u2,…,umDrive control item parameter of (1).
7) Calculating control behavior and driving control item parameters C by adopting correlation analysis methodi,umDegree of association of (E) is expressed as ∈iIf epsiloni>Beta, then the drive control item parameter is called Ci,umIs the main driver sequence, beta is the relevance threshold,
Figure BDA0002814995600000101
sim (u1, u2) is the sub-item u1,u2The degree of similarity of (a) to (b),
Figure BDA0002814995600000102
represents the average score, R, of all the sub-items scoredu1，Ru2Respectively represent a pair sub-item u1,u2The score of (1).
8) And predicting the system behavior series according to the driving factors.
For a given multivariable control system, there often exists a complex non-linear relationship between its sequence of drive factors and the sequence of system behavior, in terms of εi>β, determining a new sequence of primary drivers, note
Figure BDA0002814995600000103
The original sequence of drive factors is
Figure BDA0002814995600000104
After one driving factor, the sequence is generated by accumulating the driving factors once
Figure BDA0002814995600000105
For a sequence of system behaviors, there are:
Figure BDA0002814995600000106
wherein d isi+1For driving control item parameter CiFor determining the mechanism of action of the driver on the system behaviour, bi、bi+1The time of the start and the end of the driving factor respectively, and gamma is a power exponent, and reflects the nonlinear action relation of the driving factor to the system behavior.
9) And analyzing the rationality and significance of the prediction result according to the system behavior prediction value, and optimizing and adjusting the existing problems by adopting various evaluation indexes.
The data acquisition module is used for acquiring data of the equipment layer, the process layer, the production line layer, the vehicle interlayer, the enterprise layer and the cooperation layer respectively. Therefore, the scheduling of the sensors is the optimal embodiment of a series of decision processes, the method considers the constraint conditions and the tracking capacities of different sensors to the target, takes the optimal tracking performance of the system as the target, and cooperatively controls the operation parameters and the working mode of the sensors to optimize the overall performance of the system. The multi-sensor cooperative tracking is established on the basis of independent motion and perception fusion of each sensor, and dynamic resource optimization is performed on tracking feedback information of the system. The sensor dispatching object comprises the steps of changing the position, the direction, the internal parameters or the working mode of a sensor platform, performing long-term dispatching based on single-step decision and long-term dispatching based on multi-step decision, and obtaining the optimal cooperative motion scheme of the multiple sensors by optimally controlling the motion track of the movable sensor, so that the optimal target tracking performance is obtained.
Defining time k the movable sensor (1,2 …, i, …, N)s) As a function of schedule
Figure BDA0002814995600000111
Defining the state of the target at the time k as
Figure BDA0002814995600000112
Where x (k), y (k) represent the position of the target in the x and y directions at time k, respectively,
Figure BDA0002814995600000113
representing the velocity of the target in the x and y directions at time k, respectively, the state transition equation for the target can be described as
x(k+1)＝Fkx(k)+v(k)；
Wherein x (k +1) represents the position of the target in the x direction at the time k +1, and FkAnd (v) is process noise at the moment k, and follows Gaussian distribution with the mean value of 0 and the variance of Q.
Suppose NsThe angle measurement sensor carries out cooperative positioning tracking on the target, and the position and the speed of the target in the two-dimensional space are estimated by using the arrival angle measured by the sensor. In order to meet the observability of the tracking problem of multiple passive sensors, at least more than two passive sensors are required to track one target.
Azimuthal measurement z of sensor i to targeti(k) Is composed of
Figure BDA0002814995600000114
vi(k) The measurement noise of sensor i follows a gaussian distribution with mean 0 and variance Q.
The azimuth angle measurement of the multisensor system is z (k) ═ z1(k),z2(k),…,zNs(k)]Is modeled as
Figure BDA0002814995600000115
And after obtaining new target measurement information, the sensor obtains the state estimation of the target by weighted least squares. Positioning the target by using the relation between the target azimuth information measured by each sensor and the target position through direction finding positioning:
Figure BDA0002814995600000121
let xtan (z)i) -y is taken as an equivalent observation,
Figure BDA0002814995600000122
is equivalent noise, therefore, the multi-sensor measurement equation can be converted into a matrix form
Z＝HX+ε
In the formula
Figure BDA0002814995600000123
And controlling the plurality of sensors to move to the optimal observation position through the measurement characteristics of the plurality of sensors, and scheduling the job set and transmitting information. And constructing a signal transmission model of the multimedia sensor network, and scheduling data information of the sensor network to obtain signal modulation output of the coherent point of the multimedia sensor network.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications, equivalent changes and modifications made to the above embodiment according to the technical spirit of the present invention are within the scope of the technical solution of the present invention.

Claims (10)

1. An industrial control system based on big data is characterized by comprising a data acquisition module, an inquiry service module, a snapshot service module, a big data processing module, a terminal module and a cloud end module; the big data processing module comprises data integration and processing, data modeling and analysis, data decision and data control, and specifically comprises the following processing steps:
s1, a calculation manager is responsible for initializing calculation tasks, and the initialization work comprises the following steps: generating a calculation object to uniformly manage the full life cycle process of the calculation task, and acquiring and setting relevant parameters of the calculation task according to the configuration information;
s2, respectively setting clustering labels and clustering data of nodes for a user presentation layer, a service analysis layer, a data storage layer, a network transmission layer and an intelligent sensing layer;
s3, optimizing node data;
s4, refining different elements of the user presentation layer, the service analysis layer, the data storage layer, the network transmission layer and the intelligent perception layer, and respectively listing the different elements as A1,A2,....,A12；
S5, element AiThe influence between the other elements is graded and scored, and is marked as RiThe scoring rules are as follows: score 0 indicates no effect between the two factors, score 1 indicates very little effect, score 2 indicates little effect, score 3 indicates medium effect, score 4 indicates high effect, score 5 indicates very high effect, a sufficient number of experts are invited to score, the results are averaged to obtain a score set Ri；
S6, optimizing the industrial control system through multi-dimensional intelligent cooperation and optimization technology manufactured in the big data and cloud computing environment based on the cloud service planning of the industrial big data;
defining a collection of industrial items as U ═ U1,u2,…,umN is an integer, u1,u2,…,umRespectively being a sub-item in an item, the drive control item parameter of the control action is Ci(i＝1,2,…,n)，
Ci,u1＝{A1,u1,A2,u1…,A12,u1}
Ci,u2＝{A1,u2,A2,u2…,A12,u2}
Figure FDA0002814995590000011
Ci,um＝{A1,um,A2,um…,A12,um}
Ci,u1,Ci,u2,…,Ci,umRespectively representing control actions to sub-items u1,u2,…,umDrive control item parameter of (1);
s7, calculating control behavior and driving control item parameters C by adopting a correlation analysis methodi,umDegree of association of (E) is expressed as ∈iIf epsiloni>Beta, then the drive control item parameter is called Ci,umIs the main driver sequence, beta is the relevance threshold,
Figure FDA0002814995590000012
sim (u1, u2) is the sub-item u1,u2The degree of similarity of (a) to (b),
Figure FDA0002814995590000013
represents the average score, R, of all the sub-items scoredu1，Ru2Respectively represent a pair sub-item u1,u2Scoring of (4);
s8, predicting a system behavior series according to the driving factors;
for a given multivariable control system, there often exists a complex non-linear relationship between its sequence of drive factors and the sequence of system behavior, in terms of εi>β, determining a new sequence of primary drivers, note
Figure FDA0002814995590000021
The original sequence of drive factors is
Figure FDA0002814995590000022
After one driving factor, the sequence is generated by accumulating the driving factors once
Figure FDA0002814995590000023
For a sequence of system behaviors, there are:
Figure FDA0002814995590000024
wherein d isi+1For driving control item parameter CiFor determining the mechanism of action of the driver on the system behaviour, bi、bi+1Respectively starting time and ending time of the driving factors, wherein gamma is a power exponent and reflects the nonlinear action relation of the driving factors on the system behavior;
and S9, analyzing the rationality and significance of the prediction result according to the system behavior prediction value, and optimizing and adjusting the existing problems by adopting various evaluation indexes.
2. The big-data-based industrial control system according to claim 1, wherein in step S4, the different elements of the user presentation layer, the business analysis layer, the data storage layer, the network transport layer, and the intelligent sensing layer are respectively:
A1: the client requirements are used for identifying the client requirements, intelligently predicting the client requirements, intelligently analyzing the client behaviors and public opinions and intelligently acquiring the clients;
A2: the big data innovation chain is based on a big data collaborative interaction design platform, and product design, process design, collaborative design and intelligent research and development are innovatively designed by using big data;
A3：R&d, performing cross-industry and cross-region industrial chain cooperative operation based on big data;
A4: the big data management chain is used for enterprise strategic planning and market intelligent prediction, equipment intelligent management, material intelligent management, logistics intelligent optimization and warehousing intelligent optimization;
A5: the industrial operation management chain is used for acquiring data of an enterprise operation process;
A6: the big data manufacturing chain is used for intelligently sensing, intelligently predicting and analyzing based on a networked manufacturing resource collaboration platform of big data;
A7: worker's toolThe industrial manufacturing chain is used for acquiring data of enterprise equipment and production manufacturing environment;
A8: the big data service chain is used for intelligently extracting the intention of a customer, analyzing the behavior and habit of the customer, analyzing the trend of service requirements and intelligently interacting with the customer on the basis of a product life cycle analysis management platform of big data;
A9: the industrial service chain is used for acquiring data of the enterprise service chain;
A10: the big data support chain is used for processing and analyzing big data;
A11: big data processing for data analysis and optimization, network mining and decision making;
A12: the big data platform provides a support technology, a shared resource, an access system and a cloud service platform.
3. The industrial control system based on big data according to claim 1-2, characterized in that the data acquisition module respectively acquires data of an equipment layer, a process layer, a production line layer, a vehicle layer, an enterprise layer and a collaboration layer.
4. The big data based industrial control system according to claims 1-3, wherein the big data processing module further performs dynamic resource optimization on the tracking feedback information of the multiple sensors, and the specific optimization process is as follows:
defining time k the movable sensor (1,2 …, i, …, N)s) As a function of schedule
Figure FDA0002814995590000031
Defining the state of the target at the time k as
Figure FDA0002814995590000032
Where x (k), y (k) represent the position of the target in the x and y directions at time k, respectively,
Figure FDA0002814995590000033
representing the velocity of the target in the x and y directions at time k, respectively, the state transition equation for the target can be described as
x(k+1)＝Fkx(k)+v(k)；
Wherein x (k +1) represents the position of the target in the x direction at the time k +1, and FkSelecting different models for a target state transition matrix at the moment k according to different motion modes of the target (such as uniform motion, uniform acceleration motion and turning motion), wherein v (k) is process noise at the moment k, and the Gaussian distribution with the mean value of 0 and the variance of Q is obeyed;
suppose NsThe angle measuring sensors are used for carrying out cooperative positioning tracking on the target, the position and the speed of the target in a two-dimensional space are estimated by using the arrival angles measured by the sensors, and at least more than two passive sensors are required to be ensured to track one target in order to meet the observability of the tracking problem of the multiple passive sensors;
azimuthal measurement z of sensor i to targeti(k) Is composed of
Figure FDA0002814995590000034
vi(k) The measurement noise of the sensor i is subjected to Gaussian distribution with the mean value of 0 and the variance of Q;
the azimuth angle measurement of the multisensor system is z (k) ═ z1(k),z2(k),…,zNs(k)]Is modeled as
Figure FDA0002814995590000041
The sensors obtain new target measurement information and then obtain the state estimation of the target through weighted least squares, and the target is positioned through direction-finding positioning by utilizing the relation between target azimuth information measured by each sensor and the target position:
Figure FDA0002814995590000042
let xtan (z)i) -y is taken as an equivalent observation,
Figure FDA0002814995590000043
is equivalent noise, therefore, the multi-sensor measurement equation can be converted into a matrix form
Z is HX + epsilon, wherein,
Figure FDA0002814995590000044
through the measurement characteristics of the multiple sensors, the multiple sensors are controlled to move to the optimal observation positions, operation set scheduling and information transmission are carried out, a signal transmission model of the multimedia sensor network is built, data information of the sensor network is scheduled, and signal modulation output of coherent points of the multimedia sensor network is obtained.
5. The big-data-based industrial control system according to any one of claims 1 to 4, wherein step S2 is to respectively present the clustering labels and clustering data of the nodes to the user, the service analysis layer, the data storage layer, the network transmission layer, and the intelligent sensing layer, and step S3 is to optimize the node data, and specifically comprises the following steps:
suppose that a user presentation layer, a service analysis layer, a data storage layer, a network transmission layer and an intelligent perception layer are provided with M nodes which communicate through a network, and each node is provided with N nodesjSet of data points { xji|i＝1,…,NjAnd the total number of data stored in the network is N ═ Σ NjIn which N isjN data points are samples of a random variable X with a probability distribution p (X), each node j having N of the random variable XjClustering data points into at most K different classes, and allocating a cluster label omega to each data point, wherein the probability distribution of the cluster label is also p (x), K belongs to the samples of random variables Y of {1, …, K }, and each data point is sampled by random variables Y of {1, …, K }Each node aims at searching a clustering label and clustering data of the condition model p;
minimizing network routing information between X and Y by:
Figure FDA0002814995590000051
Figure FDA0002814995590000052
where F (ω) is the cross entropy loss function, p (k) is the conditional model function of the node, yjiIs the label of the ith data point of node j, G (y)ji) Is an indicator function; each node is in cooperative communication with adjacent nodes, local optimal solution is obtained by minimizing cross entropy loss function, the size of the model is compressed, and overfitting is avoided.
6. The big data based industrial control system according to any one of claims 1 to 4, wherein the data collection module receives industrial data of data sources such as industrial field devices and third party control systems through collection services, and uploads the industrial data to the data storage unit in real time.
7. The big data based industrial control system according to any of claims 1-4, wherein the terminal module comprises a client UI module, a visualization module, the client UI module is adapted to collect terminal user information.
8. The big data based industrial control system according to claim 7, wherein the cloud module comprises a signal receiving and processing module, and the signal receiving and processing module is adapted to receive and process the end user information and the related big data collected by the client UI module.
9. The big data-based industrial control system according to any one of claims 1 to 3, wherein the query service module is configured to perform unified scheduling and management on all query requests, execute query tasks according to a specified algorithm and priority, and return query results to the client.
10. The industrial control system based on big data according to any one of claims 1 to 3, wherein the snapshot service module provides real-time data services to the outside by using the process variable as a structural unit and using the memory mapping file as a cache database of the carrier, and the real-time data services include real-time query and change subscription of the current state of the process variable.
CN202011395699.8A 2020-12-03 2020-12-03 Industrial control system based on big data Withdrawn CN112508382A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011395699.8A CN112508382A (en) 2020-12-03 2020-12-03 Industrial control system based on big data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011395699.8A CN112508382A (en) 2020-12-03 2020-12-03 Industrial control system based on big data

Publications (1)

Publication Number Publication Date
CN112508382A true CN112508382A (en) 2021-03-16

Family

ID=74969554

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011395699.8A Withdrawn CN112508382A (en) 2020-12-03 2020-12-03 Industrial control system based on big data

Country Status (1)

Country Link
CN (1) CN112508382A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114518963A (en) * 2022-04-21 2022-05-20 中国航空工业集团公司沈阳飞机设计研究所 Edge information cooperative processing method and system for airborne end system

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114518963A (en) * 2022-04-21 2022-05-20 中国航空工业集团公司沈阳飞机设计研究所 Edge information cooperative processing method and system for airborne end system

Similar Documents

Publication Publication Date Title
Pan et al. A BIM-data mining integrated digital twin framework for advanced project management
US20190244129A1 (en) Data orchestration platform management
Zhao et al. Digital twin-enabled dynamic spatial-temporal knowledge graph for production logistics resource allocation
Bin et al. Research on data mining models for the internet of things
US20060235742A1 (en) System and method for process evaluation
CN110990461A (en) Big data analysis model algorithm model selection method and device, electronic equipment and medium
Lujic et al. Efficient edge storage management based on near real-time forecasts
US11593735B2 (en) Automated and efficient personal transportation vehicle sharing
CN109213752A (en) A kind of data cleansing conversion method based on CIM
Li et al. Parallelizing skyline queries over uncertain data streams with sliding window partitioning and grid index
CN116703303A (en) Warehouse visual supervision system and method based on multi-layer perceptron and RBF
CN117172509B (en) Construction project distribution system based on decoration construction progress analysis
CN117768469B (en) Cloud service management method and system based on big data
CN116881744A (en) Operation and maintenance data distribution method, device, equipment and medium based on Internet of things
Tian et al. RETRACTED ARTICLE: Research on emergency vehicle routing planning based on short-term traffic flow prediction
CN112508382A (en) Industrial control system based on big data
CN111125450A (en) Management method of multilayer topology network resource object
CN114706994A (en) Operation and maintenance management system and method based on knowledge base
Lujic et al. Resilient edge data management framework
CN117807377A (en) Multidimensional logistics data mining and predicting method and system
CN106779245B (en) Event-based civil aviation demand prediction method and device
CN114757448B (en) Manufacturing inter-link optimal value chain construction method based on data space model
Ye et al. Application of time series method to the passenger flow prediction in the intelligent bus transportation system with big data
Azevedo et al. A reduced network traffic method for IoT data clustering
CN112990647A (en) Distribution network project equipment portrait index evaluation system and evaluation method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WW01 Invention patent application withdrawn after publication
WW01 Invention patent application withdrawn after publication

Application publication date: 20210316