Application linkage control method based on information fusion technology

Abstract

The invention belongs to the technical field of mine application linkage control, and particularly relates to an application linkage control method based on an information fusion technology, which comprises a visual arrangement module, a data acquisition module, a fusion processing module, a fusion decision analysis module, a linkage control execution module and an archiving analysis module. Comprising the following steps: S1, carrying out visual arrangement on dragging, pulling and dragging combination operations of service combination, service governance, data circulation and other components, and realizing a complete linkage control application through modes of parameter and interface service configuration, event visual arrangement and the like; s2, acquiring sensing data and basic data of safety monitoring, industrial video, water disaster prevention and control systems and unstructured data such as graphs, videos, documents and voices, and storing acquisition results into a data storage system supporting the structured data and the unstructured data;

Description

Technical Field

[0001] The present invention relates to the technical field of mine application linkage control, and in particular to an application linkage control method based on information fusion technology. Background Art

[0002] There are many sudden natural disasters such as gas, water, fire, and rock burst in mining production, which are very harmful and have a wide range of impact. Usually, when a major accident occurs and emergency evacuation of personnel is required, telephone, broadcast or underground mobile phone calls are often used to notify and guide on-site personnel to evacuate dangerous areas. However, due to the harsh environment and complex terrain of the mine, the number of telephone installation points and the number of people holding underground mobile phones is limited, and the broadcast system cannot cover all areas. It is impossible to fully and timely notify every underground worker, which will result in untimely and incomplete notification of personnel, delaying the opportunity to escape from danger and causing major accidents.

[0003] The mining industry has established various communication and safety assurance systems such as program-controlled dispatching, wireless communication, broadcasting systems, personnel positioning and safety monitoring, which play an important role in coal mine production safety, dispatching and command, emergency rescue, etc. However, each system is independent of each other and cannot be controlled in a coordinated manner, which restricts the development of active mine safety warning and cannot provide effective means and safety assurance.

[0004] Information fusion is a new technology for comprehensive processing of multi-source information. It can intelligently synthesize multi-source information from a certain target to produce more accurate and complete estimates and judgments than a single information source. In recent years, with the development of big data, artificial intelligence, and communication technology, this technology has developed rapidly and has attracted widespread attention worldwide. Research on the multi-system fusion linkage control platform of coal mines, the integration and interconnection of coal mine program control scheduling, wireless communication, and emergency broadcasting, and the linkage of communication systems with personnel positioning, safety monitoring, and automated control systems can significantly improve the efficiency of notifying underground personnel in emergency situations.

[0005] To this end, the present invention provides an application linkage control method based on information fusion technology. Summary of the invention

[0006] In order to make up for the deficiencies of the prior art and solve at least one technical problem raised in the background technology, the present invention proposes an application linkage control method based on information fusion technology.

[0007] The technical solution adopted by the present invention to solve its technical problems is: the application linkage control method based on information fusion technology described in the present invention includes S1: visually arranging the drag, drop and drag combination operations of components such as service combination, service governance, and data flow, and realizing a complete linkage control application through parameters, interface service configuration and event visualization orchestration.

[0008] S2: Collect perception data and basic data, graphics, video, documents, voice and other unstructured data from security monitoring, industrial video, flood control and other systems, and store the collected results in a data storage system that supports structured and unstructured data.

[0009] S3: Clean and process the data resources integrated from heterogeneous systems, implement data quality verification rules for different types of business systems, verify the consistency, completeness, timeliness and accuracy of the data, catalog the data resources and store them accordingly.

[0010] S4: The standardized data after governance of the collected data from each subsystem is calculated in the linkage control model, the results of each event are evaluated, and the linkage control plan is selected and recommended based on expert knowledge and previous decision-making plans for similar events in the prior database.

[0011] S5: Identify according to service functions, such as fire extinguishing system, safety monitoring system, etc., and execute according to the linkage control plan. At the same time, record and feedback the existing execution status.

[0012] S6: Record, combine, analyze and archive the linkage control as a whole. At the same time, store relevant cases in the knowledge base and extract and store process rules.

[0013] The beneficial effects of the present invention are as follows:

[0014] 1. The application linkage control method based on information fusion technology described in the present invention has robustness and parallel processing capabilities, and enhances the real-time and accuracy of control operation. By referring to various corresponding instructions set in advance by people, through the correlation analysis of various response measures, such as linkage relationships, irrelevant relationships, and contradictory relationships, various events can be comprehensively processed quickly and orderly, and the losses caused by various concurrent events can be minimized. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention will be further described below in conjunction with the accompanying drawings.

[0016] Figure 1 It is a flow chart of the application linkage control method based on information fusion technology provided by the present invention;

[0017] Figure 2 It is a fusion decision structure diagram of the application linkage control method based on information fusion technology provided by the present invention;

[0018] Figure 3 It is a linkage control execution structure diagram of the application linkage control method based on information fusion technology provided by the present invention.

[0019] Figure 4 It is an application linkage control mode diagram based on information fusion technology provided by the present invention. DETAILED DESCRIPTION

[0020] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0021] like Figure 1 to Figure 2 As shown, an application linkage control method based on information fusion technology described in an embodiment of the present invention includes a visualization orchestration module, a data acquisition module, a fusion processing module, a fusion decision analysis module, a linkage control execution module, and an archiving analysis module.

[0022] The visual orchestration module visually orchestrates the drag, drop and combination operations of components such as service composition, service governance and data flow, and realizes complete linkage control application through parameter, interface service configuration and event visual orchestration.

[0023] Specifically, visual orchestration includes service management components, operator components, and model orchestration components. The service management component configures and tunes the engine's address, port, number of slots, and other parameters. The operator component includes various components, such as input operators, output operators, parsing operators, conversion operators, query operators, and execution operators. The model orchestration component connects and manages the input and output of the operator, transfers the output data to the input interface of the next operator, and connects data, events, and actions in series.

[0024] Furthermore, orchestration refers to an orchestration engine with functions such as service combination and service governance. Through a visual orchestration interface, existing microservices and interfaces are arranged according to business requirements and process rules, and automatically scheduled or re-aggregated into new services or APIs through a process engine and released to the outside world. It focuses on re-combining the services provided by the system, which is an enhancement of service capabilities. Through orchestration, the newly combined services are governed and enhanced in function, while atomic services and their systems only need to focus on their internal business logic.

[0025] Optionally, kubescheduler is a scheduler component of Kubernetes that handles the issue of which node a Pod is scheduled to in the cluster. The scheduler schedules the Pod to the optimal node based on certain scheduling policies (such as node resource utilization, node affinity, taints, etc.), thereby improving the resource utilization and performance of the cluster.

[0026] Furthermore, components are any objects that can be dragged to the canvas for layout during interface arrangement, such as text boxes, tables, buttons, two-dimensional and three-dimensional map containers, system REST services and other service functions. Arrangement is used to implement component events and behaviors after event triggering. Component events include ordinary component click events, map click events, page loading events, data change events, drawing end events, etc.; behaviors after event triggering include operating data, operating components, operating procedures, pop-up window opening, page jump, variable operation, logic definition, etc. Set the total number of components to l, the total number of component event enumerations to e, and the total number of all behavior enumerations to b.

[0027]

[0028] Represents a many-to-many relationship between a component set and an event set. Any element y i，j ={y|y∈{0,1}} when y i，j When =1, it means that the i-th component has the j-th event; otherwise, it does not.

[0029]

[0030] Represents a many-to-many relationship between a collection of components and the operations that can be performed after an event is triggered. Any element z i，j = {z|z∈{0,1}} when z i，j =1, it means that the jth behavior can be performed after the i-th event is triggered; otherwise, it is not possible.

[0031] The data acquisition module collects perception data and basic data, graphics, video, documents, voice and other unstructured data from systems such as security monitoring, industrial video, flood control, and emergency broadcasting, and stores the collection results in a data storage system that supports structured and unstructured data.

[0032] Furthermore, in the mining industry, the equipment and business systems of different manufacturers and different communication methods make it impossible to achieve data integration between various data sources, which has become a key point restricting the networking, informatization and intelligence of mining enterprises. Collect, convert and forward multi-source heterogeneous data to interconnect and integrate system data.

[0033] Among them, the data conversion protocol is defined, and the source data object is converted into the target data object according to the protocol. The data collection and forwarding mechanism is defined, and the source data is collected and forwarded to the data target periodically or non-periodically according to the collection and forwarding mechanism. The driver includes commonly used protocol drivers, such as OPC, MODBUS, InSQL, WebService, etc. Peripheral devices and systems are data generators and data consumers. Breaking the limitations of inconsistent data structures of various business data systems can realize unified interface collection of various business systems, so that the business subsystems are open and the "data island" problem can be solved through information fusion.

[0034] The fusion processing module cleans and processes the data resources integrated from heterogeneous systems, implements data quality verification rules for different types of business systems, verifies data consistency, completeness, timeliness and accuracy, and catalogs and stores data resources accordingly.

[0035] Specifically, for the video data of the industrial video system of the coal mining industry, the point table data of the rock burst system, the mine map data of the coal mining industry, the basic data of the safety monitoring system, etc., these data from different sources or different systems are stored, integrated and merged. In the process of data aggregation, various types and forms of data are gathered into a unified data storage or data warehouse.

[0036] Furthermore, the data is classified and graded according to the differences in the nature of the data source categories, and the data is analyzed and eliminated based on similarity and duplicate data records, and the data analysis recall rate and data analysis accuracy are often used as quantitative evaluation indicators of data calculation efficiency. Various types of data cleaning algorithms are treated as a black box and run in a sorting or cross method to clean the data.

[0037] Furthermore, we will conduct quality monitoring and management on the processed data, conduct more comprehensive and scientific effective monitoring, evaluation and implementation of improvement management process on data analysis information, so as to further ensure the high accuracy, integrity, uniformity, timeliness and authenticity of the information data content, improve the reliability and availability of data, and thus support correct decision-making and reliable analytical decision-making.

[0038] Furthermore, by means of multimodal information extraction, valuable information can be accurately extracted from various information data sets. By using classification information extraction, sequential labeling information extraction and rule-based information extraction, the boundaries of special types of information can be detected to extract intermediate information.

[0039] Optionally, build a model for AI learning to make predictions by labeling observation sequences. Mainstream models include hidden Markov models, maximum entropy Markov models, conditional random fields, etc.

[0040] Furthermore, multimodal data fusion is performed on the data to merge and integrate multiple data from different data sources or modalities to enhance the understanding, description or inference of the target. By integrating information from multiple data sources, more comprehensive, accurate and reliable results can be obtained than from a single data source. In the fusion model of multi-sensor modal data, the data sources include multiple sensors from various different types, such as gas sensors, noise sensors, dust sensors, etc. The fused data can include raw sensor data, high-level features or semantic information.

[0041] Optionally, rule-based methods, statistical regression methods, deep learning methods, etc. can be used to perform data fusion. Common multimodal data fusion techniques include feature layer fusion, trust weight fusion, and deep learning network fusion.

[0042] The integrated decision analysis module calculates the governed and standardized data of the collected data from each subsystem in the linkage control model, evaluates the results of each event, and selects and recommends linkage control plans based on expert knowledge and previous decision plans for similar events in the prior database.

[0043] like Figure 2 As shown in the figure, fusion decision analysis mainly includes joint decision fusion, evaluation analysis, and scheme selection. It performs decision fusion on the data from sensor subsystems and data collection. Evaluation analysis evaluates and analyzes the fusion results, and takes the degree of harm caused by each event as an evaluation indicator. Scheme selection uses the results of joint decision fusion and evaluation analysis, and draws on the expert knowledge in the knowledge base and previous decision plans of similar events in the prior database to formulate, select and recommend control plans.

[0044] Optionally, most general methods of information fusion are applicable, and commonly used methods include Kalman filtering, Bayesian estimation, fuzzy reasoning theory and neural network theory, genetic algorithm, wavelet analysis theory, and support vector machine.

[0045] The linkage control execution module is identified according to the service functions, such as fire extinguishing system, safety monitoring system, etc., and is executed according to the linkage control plan. At the same time, the existing execution status is recorded and fed back.

[0046] Furthermore, the linkage control execution system mainly includes two parts: function identification and resource allocation, such as Figure 4, is a linkage control execution structure diagram of the application linkage control method based on information fusion technology provided by the present invention. Function identification identifies the actuators according to their functions, including emergency control systems, fire control systems, etc., and these control systems also contain some smaller subsystems. Classifying the actuators can facilitate the implementation of control functions. Resource allocation is for limited actuator resources. It is necessary to prioritize the actuators according to the control scheme so that they can give priority to the control requirements under the most dangerous level. Under this condition, the actuator resources are allocated. At the same time, the existing actuator resources are fed back to the decision-making fusion analysis system to achieve comprehensive information fusion.

[0047] Optionally, for emergency control systems, functions such as disaster warning, automatic sprinkler protection, carbon dioxide protection linkage control, etc. are often integrated, so it is more effective to adopt intelligent methods, such as knowledge graph-based methods.

[0048] The archiving and analysis module records, combines, analyzes and archives the overall linkage control. At the same time, it stores relevant cases in the knowledge base and extracts and stores process rules.

[0049] Furthermore, the linkage control template is used to initialize the data creation of the entire linkage control process, integrate electronic documents, fill in data, compile text data, and form a complete record.

[0050] Furthermore, the linkage control data is classified and stored. To ensure the integrity of the stored data and the accuracy of the classification, it is particularly important to introduce a verification mechanism at the end of this link. With the help of the verification mechanism, problems such as missing items in data reporting and classification errors can be discovered and corrected in a timely manner.

[0051] Optional, closed-loop management can improve the quality and efficiency of rectification. To achieve closed-loop management of linkage control, common problems are classified and a knowledge case library is formed through continuous summary for subsequent optimization. The accuracy of decision analysis is improved through decision-making, control, feedback, re-decision-making, re-control, and re-feedback.

[0052] like Figure 4As shown in the figure, it is the functional target information and overall level hierarchy of the application linkage control scheme based on information fusion technology. Each functional block diagram may also contain corresponding sub-bodies or subsystems, and the sub-bodies or subsystems may still contain lower-level sub-bodies or subsystems! In addition, the figure also shows the direction of information flow. From the perspective of data extraction, the direction of data flow is from bottom to top, while from the perspective of control signals, its flow tends to be downward. In this case, the perception data information of the sensor system will be extracted in sequence in the upward direction and finally reach the fusion decision analysis system, while the control signal is transmitted downward. Finally, the main control signal acts on the controlled target through the linkage execution system to achieve complete closed-loop control.

[0053] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An application linkage control method based on information fusion technology, comprising a visual arrangement module, a data acquisition module, a fusion processing module, a fusion decision analysis module, a linkage control execution module, and an archiving analysis module; comprising the following steps: S1: Visually orchestrate the drag, drop, and combination operations of components such as service composition, service governance, and data flow, and implement complete linkage control applications through parameter, interface service configuration, and event visual orchestration; S2: Collect the perception data and basic data, graphics, video, documents, voice and other unstructured data of security monitoring, industrial video, flood control and other systems, and store the collection results in a data storage system that supports structured and unstructured data; S3: Clean and process the data resources integrated from heterogeneous systems, verify the data quality rules of different types of business systems, verify the consistency, completeness, timeliness and accuracy of the data, catalog the data resources and store them accordingly; S4: Calculate the standardized data after governance of the collected data from each subsystem in the linkage control model, evaluate the results of each event, and select and recommend the linkage control plan based on expert knowledge and previous decision-making plans for similar events in the prior database; S5: Identify according to service functions, such as fire extinguishing system, safety monitoring system, etc., and execute according to the linkage control plan. At the same time, record and feedback the existing execution status; S6: Record, combine, analyze and archive the linkage control as a whole. At the same time, store relevant cases in the knowledge base and extract and store process rules.

2. The application linkage control method based on information fusion technology according to claim 1 is characterized in that: The visual orchestration module includes a service management component, an operator component, and a model orchestration component. The service management component configures and tunes the engine's address, port, number of slots, etc.

3. The application linkage control method based on information fusion technology according to claim 1 is characterized in that: The data acquisition module collects perception data and basic data, graphics, video, documents, voice and other unstructured data of systems such as security monitoring, industrial video, flood control, and emergency broadcasting, and stores the collection results in a data storage system that supports structured data and unstructured data.

4. The application linkage control method based on information fusion technology according to claim 1 is characterized in that: The fusion processing module cleans and processes the data resources integrated from heterogeneous systems, verifies the data quality rules of different types of business systems, verifies the consistency, completeness, timeliness and accuracy of the data, and catalogs and stores the data resources accordingly.

5. The application linkage control method based on information fusion technology according to claim 1 is characterized in that: The fusion decision analysis module calculates the managed normative data of the collected data from each subsystem in the linkage control model, evaluates the results produced by each event, and selects and recommends linkage control plans based on expert knowledge and previous decision plans for similar events in the prior database.

6. The application linkage control method based on information fusion technology according to claim 1 is characterized in that: The linkage control execution module mainly includes two parts: function identification and resource configuration.

7. The application linkage control method based on information fusion technology according to claim 1 is characterized in that: The archiving and analysis module records, combines, analyzes and archives the linkage control as a whole, and at the same time, stores relevant cases in the knowledge base and extracts and stores process rules.

Images