A simulation control method and device for coal heavy medium

Abstract

The present invention discloses a simulation control method and device for coal heavy medium, which relates to the technical field of heavy medium separation. The method includes: an AI system automatically identifies all effective processes of the current process and records data corresponding to the effective processes; the above data is reversely sorted to obtain the type of product coal, and the product coal is split and then integrated into an AI learning file, the AI ​​learning file includes the type, composition, precision, density, moisture content, and temperature of the current process and product coal; the AI ​​learning file is analyzed based on a deep learning algorithm to reconstruct the current process data, obtain an optimized process, and apply it to the next process; repeat the process of the AI ​​system automatically identifying all effective processes of the current process until the process with the lowest energy consumption corresponding to the product coal is obtained. The method uses data reconstruction and deep learning methods to gradually replace and improve the process flow and method with AI, thereby reducing the energy consumption of coal heavy medium.

Description

Technical Field

[0001] The invention relates to the technical field of heavy medium separation, in particular to a simulation control method and device for coal heavy medium separation. Background Art

[0002] In the traditional heavy medium separation process, the relevant data of heavy medium separation can only be obtained through the observation and accumulated experience of technicians. For example, the density of the circulating medium in the heavy medium separation has always been controlled according to the density given by the operator, and the operator has to determine the given density based on the quick ash once an hour. Therefore, it is impossible to obtain more accurate data manually, which also means that the addition of materials in the heavy medium process is inaccurate.

[0003] Secondly, in the traditional heavy medium separation process, the operator cannot accurately control the heavy medium machine valve or the opening and closing degree of the hopper, resulting in separation failure. Therefore, how to carry out comprehensive intelligent transformation of the heavy medium production system, reduce the intensity of manual labor, and improve the stability of product quality is a technical problem to be solved. Summary of the invention

[0004] In view of the above problems and technical requirements, the inventors have proposed a simulation control method and device for coal heavy medium. The technical solution of the present invention is as follows:

[0005] A simulation control method for coal heavy medium, comprising the following steps:

[0006] The AI ​​system automatically identifies all valid processes of the current process and records the data corresponding to the valid processes. The current process refers to the flow direction of coal output from the raw coal bunker to the flow before the heavy medium machine. The valid process includes the process flow and process method.

[0007] The above data are reversely sorted to obtain the type of product coal, and the product coal is split and then integrated into an AI learning file. The AI ​​learning file includes the current process and the type, composition, precision, density, moisture content, and temperature of the product coal;

[0008] The AI ​​system analyzes AI learning files based on deep learning algorithms to reconstruct current process data, obtain optimized processes, and apply them to the next process;

[0009] Repeat the process of the AI ​​system automatically identifying all effective processes of the current process until the process with the lowest energy consumption corresponding to the product coal is obtained;

[0010] The AI ​​system controls the opening and closing degree of each valve of the heavy medium machine under each process based on a deep learning algorithm.

[0011] A further technical solution is to record data corresponding to the effective process, including:

[0012] Sensors are installed at various locations where data monitoring is required in the current process. The sensor data forms a data stream and is stored in a remediation server for archiving. The remediation server is connected to the AI ​​system. The sensors include pressure sensors, ash meters, and belt scales.

[0013] A further technical solution is to reversely sort the above data to obtain the type of product coal, and then split the type of product coal, including:

[0014] The AI ​​system obtains the types of product coal based on the known component density of the product coal through comparative analysis, and intelligently splits the product coal according to different classifications. The classifications include the current process and the composition, precision, density, and moisture content of the product coal.

[0015] Its further technical solution is that the AI ​​system analyzes the AI ​​learning file based on the deep learning algorithm to reconstruct the current process data and obtain the optimized process, including:

[0016] The AI ​​system retrieves the AI ​​learning file and the data released by the AI ​​system itself for real-time comparison, analysis and verification, and feeds the results back to the AI ​​system. The AI ​​system reconstructs the process of each coal flow with the purpose of reducing energy consumption; the comparison and analysis content includes real-time analysis of the temperature changes of the product coal, real-time analysis of the moisture content of the product coal, real-time analysis of the composition of the product coal, real-time analysis of the size of the product coal, and real-time analysis of the coal quality of the product coal.

[0017] Its further technical solution is that the AI ​​system controls the opening and closing degree of each valve of the heavy medium machine in each process based on the deep learning algorithm, including:

[0018] The AI ​​system segments the loading and unloading data of the current process based on a deep learning algorithm, obtains the composition of the product coal and the opening and closing degree of each valve of the heavy medium machine, and transmits it to the controllers of each valve of the heavy medium machine;

[0019] The AI ​​system obtains the operating conditions of each valve pushed by the controller, and repeats the process of segmenting the loading and unloading data of the next process based on the deep learning algorithm, to obtain the composition of the coal product and the opening and closing degree of each valve of the optimized heavy medium machine, until the control accuracy of each valve meets the predetermined requirements.

[0020] Its further technical solution is that the method also includes: the AI ​​system performs real-time three-dimensional modeling of the entire factory.

[0021] The further technical solution is that the AI ​​system performs real-time 3D modeling of the entire plant, including:

[0022] Based on the data collected by cameras distributed on equipment, objects and factories, the AI ​​system calculates and models data streams in real time and organizes the data;

[0023] The AI ​​system uses front-end devices to perform real-time calculations, performs three-dimensional modeling based on real-time calculation data within the three-dimensional platform, and automatically shares it with the client.

[0024] A simulation control device for coal heavy medium, which is mounted in an AI system and includes:

[0025] The data identification module is used to automatically identify all valid processes of the current process and record the data corresponding to the valid processes. The current process refers to the process from the coal flow direction output from the raw coal bunker to the heavy medium machine. The valid process includes the process flow and process method.

[0026] The data sorting module is used to reverse sort the above data to obtain the type of product coal, split the product coal, and then integrate it into an AI learning file. The AI ​​learning file includes the current process and the type, composition, precision, density, and moisture content of the product coal;

[0027] The process optimization module is used to analyze AI learning files based on deep learning algorithms to reconstruct the current process data, obtain the optimized process, and apply it to the next process;

[0028] The optimization discrimination module is used to re-call the data identification module until the process with the lowest energy consumption corresponding to the product coal is obtained;

[0029] The heavy medium optimization module is used to control the opening and closing degree of each valve of the heavy medium machine under each process based on the deep learning algorithm.

[0030] Its further technical solution is that the data sorting module includes:

[0031] The sorting and splitting unit is used to obtain the type of product coal based on the known component density of the product coal through comparative analysis and sorting, and to intelligently split the product coal according to different classifications. The classifications include the current process and the composition, precision, density, moisture content and temperature of the product coal.

[0032] Its further technical solution is that the process optimization module includes:

[0033] The comparison and analysis unit is used to retrieve the AI ​​learning file and the data released by the AI ​​system itself for real-time comparison and analysis and verification, and feed back the results to the process reconstruction unit; the comparison and analysis content includes real-time analysis of the temperature change of the product coal, real-time analysis of the moisture content of the product coal, real-time analysis of the composition of the product coal, real-time analysis of the size of the product coal, and real-time analysis of the coal quality of the product coal;

[0034] The process reconstruction unit is used to reconstruct the process of each coal flow based on the purpose of reducing energy consumption.

[0035] The beneficial technical effects of the present invention are:

[0036] The simulation control method proposed in this application is based on the idea of ​​data-driven, and uses data reconstruction and deep learning to gradually replace and improve the process flow and process methods with AI. The AI ​​intelligent technology used is suitable for normal medium addition, heavy medium, concentration, filtration, and other processes and equipment that can be automatically controlled by AI. It overcomes the problems caused by traditional manual records and manual operations, realizes unmanned operation, and can quickly improve the equipment, optimize the process, and accurately control the heavy medium machine, thereby improving the stability of the clean coal product quality and the clean coal yield. Furthermore, through AI three-dimensional modeling of the factory building, the operator can see the operation content and details of each part of the equipment at a remote end. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flow chart of the simulation control method of coal heavy medium provided in this application.

[0038] Figure 2 It is a flow chart of step 5 in the simulation control method of coal heavy medium provided in the present application.

[0039] Figure 3 It is a flow chart of step 6 in the simulation control method of coal heavy medium provided in the present application.

[0040] Figure 4 It is a schematic diagram of the simulation control device of coal heavy medium provided in this application. DETAILED DESCRIPTION

[0041] The specific implementation of the present invention will be further described below in conjunction with the accompanying drawings.

[0042] The present application provides a simulation control method for coal heavy medium, the flow chart of which is as follows: Figure 1 As shown, the specific steps include:

[0043] Step 1: The AI ​​system automatically identifies all valid processes in the current process and records the data corresponding to the valid processes.

[0044] Among them, the current process refers to the process from the coal flow direction output from the raw coal bin to the heavy medium machine, such as the situation of the product coal reaching a certain machine from the raw coal bin; the effective process includes the process flow and process method.

[0045] Among them, recording the data corresponding to the effective process includes: installing sensors at various parts of the current process that need to monitor data, storing the sensor data in the data stream and creating files on the heavy media server, and connecting the heavy media server to the AI ​​system. Sensors include pressure sensors, ash analyzers and belt scales.

[0046] Step 2: Reversely sort the above data to obtain the types of product coal, split the product coal, and then integrate it into an AI learning file, including:

[0047] The AI ​​system obtains the types of product coal based on the known component density of the product coal through comparative analysis, and intelligently splits the product coal according to different classifications. The classifications include the current process and the composition, precision, density, and moisture content of the product coal.

[0048] The AI ​​learning file includes the type, composition, precision, density, moisture content, and temperature of the current process and product coal. Optionally, it also includes coal loss and energy consumption of the product coal.

[0049] Step 3: The AI ​​system analyzes the AI ​​learning file based on the deep learning algorithm to reconstruct the current process data, obtain the optimized process, and apply it to the next process, including:

[0050] Step 31: The AI ​​system retrieves the AI ​​learning file and compares and analyzes the data released by the AI ​​system itself in real time, and verifies the results, and feeds them back to the AI ​​system.

[0051] Among them, the comparative analysis content includes real-time analysis of the temperature change of the product coal, real-time analysis of the moisture content of the product coal, real-time analysis of the composition of the product coal, real-time analysis of the size of the product coal, and real-time analysis of the coal quality of the product coal.

[0052] Step 32: The AI ​​system restructures the process for each coal flow based on the purpose of reducing energy consumption.

[0053] Step 4: Repeat the process of the AI ​​system automatically identifying all valid processes of the current process, that is, repeat step 1, until the process with the lowest energy consumption corresponding to the product coal is obtained.

[0054] In this embodiment, data is collated by analyzing coal loss, energy consumption, material quantity, moisture content, temperature, etc. during the coal flow process. Through continuous collation and calculation, AI can eventually infer the process of the required product coal, such as the speed reaching a certain machine for filtration, the range of filtered materials, the degree of screening, etc. The entire coal washing process is rebuilt through the data it has introduced, realizing fully automated coal washing production with higher accuracy than the current semi-manual coal washing.

[0055] The method also includes step 5: the AI ​​system controls the opening and closing degree of each valve of the heavy medium machine under each process based on the deep learning algorithm, such as Figure 2 As shown, specifically including:

[0056] Step 51: The AI ​​system segments the loading and unloading data of the current process based on the deep learning algorithm, obtains the composition of the product coal and the opening and closing degree of each valve of the heavy medium machine, and transmits it to the controllers of each valve of the heavy medium machine.

[0057] Step 52: The AI ​​system obtains the operating conditions of each valve pushed by the controller, and repeats the process of segmenting the loading and unloading data of the next process based on the deep learning algorithm, that is, repeats step 51, to obtain the composition of the product coal and the opening and closing degree of each valve of the optimized heavy medium machine, until the control accuracy of each valve meets the predetermined requirements.

[0058] In this embodiment, the controller feeds back the result data of the content of the loading and unloading materials and the content of the material mixture to the AI ​​system, which gradually improves all the data, obtains the optimized opening and closing degree of each valve, and sends it to the controller, thereby realizing precise control of the gripper action according to data such as the amount of loading and unloading materials and the content of the loading and unloading materials, so as to complete a heavy medium process.

[0059] The method also includes step 6: the AI ​​system performs real-time three-dimensional modeling of the entire plant, such as Figure 3 As shown, specifically including:

[0060] Step 61: Based on the data collected by cameras distributed on equipment, objects and factories, the AI ​​system calculates and models the data flow in real time and organizes the data.

[0061] Step 62: The AI ​​system uses the front-end device to perform real-time calculations, performs three-dimensional modeling based on the real-time calculated data in the three-dimensional platform, and automatically shares it with the client, such as automatically sharing it with the control room and system-related personnel.

[0062] In this embodiment, since the existing methods cannot enable operators to master the overall operation in the centralized control room, and cannot see the real-time changes of the machine / equipment during the operation, this application uses AI to perform three-dimensional modeling of the machine equipment, so that the operator can see the operation content and details of each part of the machine remotely.

[0063] Based on the same inventive concept, the present application also provides a simulation control device for coal heavy medium, which is mounted in an AI system, and its principle diagram is as follows: Figure 4 As shown, it includes a data identification module, a data sorting module, a process optimization module, a heavy medium optimization module and a three-dimensional modeling module, wherein:

[0064] The data identification module is used to automatically identify all valid processes of the current process and record the data corresponding to the valid processes. The current process refers to the process from the coal flow direction output from the raw coal bin to the heavy medium machine. The valid process includes the process flow and process method.

[0065] The data sorting module is used to reverse sort the above data to obtain the type of product coal, split the product coal, and then integrate it into an AI learning file. The AI ​​learning file includes the current process and the type, composition, precision, density, and moisture content of the product coal.

[0066] The process optimization module is used to analyze AI learning files based on deep learning algorithms to reconstruct the current process data, obtain the optimized process, and apply it to the next process.

[0067] The optimization discrimination module is used to re-call the data identification module until the process with the lowest energy consumption corresponding to the product coal is obtained.

[0068] The heavy medium optimization module is used to control the opening and closing degree of each valve of the heavy medium machine under each process based on the deep learning algorithm.

[0069] The 3D modeling module is used to perform real-time 3D modeling of the entire factory.

[0070] Among them, the data sorting module includes a sorting and splitting unit, which is used to obtain the type of product coal based on the known component density of the product coal through comparative analysis and sorting, and intelligently split the product coal according to different classifications. The classifications include the current process and the composition, precision, density, moisture content, and temperature of the product coal.

[0071] Among them, the process optimization module includes a comparison analysis unit and a process reconstruction unit, specifically:

[0072] The comparison and analysis unit is used to retrieve the AI ​​learning file and the data released by the AI ​​system itself for real-time comparison and analysis and verification, and feed back the results to the process reconstruction unit; the comparison and analysis content includes real-time analysis of the temperature change of the product coal, real-time analysis of the moisture content of the product coal, real-time analysis of the composition of the product coal, real-time analysis of the size of the product coal, and real-time analysis of the coal quality of the product coal;

[0073] The process reconstruction unit is used to reconstruct the process of each coal flow based on the purpose of reducing energy consumption.

[0074] Among them, the heavy medium optimization module includes a data segmentation unit and a control optimization unit, specifically:

[0075] The data segmentation unit is used to segment the loading and unloading data of the current process based on the deep learning algorithm, obtain the composition of the product coal and the opening and closing degree of each valve of the heavy medium machine, and transmit them to the controllers of each valve of the heavy medium machine.

[0076] The control optimization unit is used to obtain the operating conditions of each valve pushed by the controller, and repeatedly call the data segmentation unit to obtain the composition of the product coal and the opening and closing degree of each valve of the optimized heavy medium machine until the control accuracy of each valve meets the predetermined requirements.

[0077] The 3D modeling module includes a data sorting unit and a modeling unit, specifically:

[0078] The data sorting unit is used to calculate and model data streams and sort data in real time based on the data collected by cameras distributed on equipment, objects and factories.

[0079] The modeling unit is used to perform real-time calculations using front-end equipment, perform three-dimensional modeling based on real-time calculated data within a three-dimensional platform, and automatically share the data with the client, such as automatically sharing it with the control room and system-related personnel.

[0080] The solution to the problem provided by the device is similar to the solution recorded in the above method. Therefore, the specific limitations in the above-mentioned simulation control device for coal preparation heavy medium can be found in the above-mentioned limitations on the simulation control method for coal preparation heavy medium, which will not be repeated here.

[0081] The above is only a preferred embodiment of the present application, and the present invention is not limited to the above embodiments. It is understood that other improvements and changes directly derived or associated by those skilled in the art without departing from the spirit and concept of the present invention should be considered to be included in the protection scope of the present invention.

Claims

1. A simulation control method for coal heavy medium, characterized in that: The method comprises: The AI ​​system automatically identifies all effective processes of the current process and records the data corresponding to the effective process. The current process refers to the process from the coal flow direction output from the raw coal bunker to the heavy medium machine. The effective process includes the process flow and process method. The above data are reversely sorted to obtain the type of product coal, and the product coal is split and then integrated into an AI learning file, which includes the current process and the type, composition, precision, density, moisture content, and temperature of the product coal; The AI ​​system analyzes the AI ​​learning file based on a deep learning algorithm to reconstruct current process data, obtain an optimized process, and apply it to the next process; Repeat the process of automatically identifying all effective processes of the current process by the AI ​​system until the process with the lowest energy consumption corresponding to the product coal is obtained; The AI ​​system controls the opening and closing degree of each valve of the heavy medium machine under each process based on a deep learning algorithm; The reverse sorting of the above data to obtain the type of product coal and splitting the type of product coal includes: The AI ​​system obtains the types of product coals through comparative analysis based on the known component density of product coals, and intelligently splits the product coals according to different categories, including the composition, precision, density, moisture content, and temperature of the current process and product coals; The AI ​​system controls the opening and closing degree of each valve of the heavy medium machine under each process based on the deep learning algorithm, including: The AI ​​system segments the loading and unloading data of the current process based on the deep learning algorithm, obtains the composition of the product coal and the opening and closing degree of each valve of the heavy medium machine, and transmits them to the controller of each valve of the heavy medium machine; The AI ​​system obtains the operating conditions of the valves pushed by the controller, and repeats the process of segmenting the loading and unloading data of the next process based on the deep learning algorithm, to obtain the composition of the product coal and the opening and closing degree of each valve of the optimized heavy medium machine, until the control accuracy of each valve meets the predetermined requirements.

2. The simulation control method of coal heavy medium according to claim 1, characterized in that: The data corresponding to the effective process is recorded, including: Sensors are installed at various locations where data monitoring is required in the current process. The sensor data forms a data stream and is stored in a remediation server for archiving. The remediation server is connected to the AI ​​system. The sensors include pressure sensors, ash meters, and belt scales.

3. The simulation control method of coal heavy medium according to claim 1, characterized in that: The AI ​​system analyzes the AI ​​learning file based on a deep learning algorithm to reconstruct current process data and obtain an optimized process, including: The AI ​​system retrieves the AI ​​learning file and the data released by the AI ​​system itself for real-time comparison, analysis and verification, and feeds back the results to the AI ​​system. The AI ​​system reconstructs the process through which each coal flow passes for the purpose of reducing energy consumption; the comparison and analysis content includes real-time analysis of the temperature change of the product coal, real-time analysis of the water content of the product coal, real-time analysis of the composition of the product coal, real-time analysis of the size of the product coal, and real-time analysis of the coal quality of the product coal.

4. The simulation control method for coal heavy medium according to any one of claims 1 to 3, characterized in that: The method also includes: the AI ​​system performs real-time three-dimensional modeling of the entire factory building.

5. The simulation control method of coal heavy medium according to claim 4 is characterized in that: The AI ​​system performs real-time 3D modeling of the entire plant, including: Based on the data collected by cameras distributed on equipment, objects and factories, the AI ​​system calculates and models data streams in real time and organizes the data; The AI ​​system uses front-end devices to perform real-time calculations, performs three-dimensional modeling on a three-dimensional platform based on real-time calculated data, and automatically shares it with the client.

6. A simulation control device for coal heavy medium, characterized in that: The device is mounted in an AI system and includes: A data identification module is used to automatically identify all valid processes of the current process and record data corresponding to the valid processes. The current process refers to the process from the coal flow direction output from the raw coal bunker to the heavy medium machine. The valid process includes process flow and process method. A data sorting module is used to reverse sort the above data to obtain the type of product coal, split the product coal, and then integrate it into an AI learning file, which includes the current process and the type, composition, precision, density, and moisture content of the product coal; A process optimization module, which is used to analyze the AI ​​learning file based on a deep learning algorithm to reconstruct the current process data, obtain an optimized process, and apply it to the next process; An optimization and discrimination module is used to call the data identification module again until a process with the lowest energy consumption corresponding to the product coal is obtained; The heavy medium optimization module is used to control the opening and closing degree of each valve of the heavy medium machine under each process based on the deep learning algorithm; Wherein, the data sorting module includes: The sorting and splitting unit is used to obtain the type of product coal through comparative analysis based on the known component density of the product coal, and to intelligently split the product coal according to different classifications, including the composition, precision, density, moisture content, and temperature of the current process and product coal; Wherein, the control of the opening and closing degree of each valve of the heavy medium machine under each process based on the deep learning algorithm includes: The AI ​​system segments the loading and unloading data of the current process based on the deep learning algorithm, obtains the composition of the product coal and the opening and closing degree of each valve of the heavy medium machine, and transmits them to the controller of each valve of the heavy medium machine; The AI ​​system obtains the operating conditions of the valves pushed by the controller, and repeats the process of segmenting the loading and unloading data of the next process based on the deep learning algorithm, to obtain the composition of the product coal and the opening and closing degree of each valve of the optimized heavy medium machine, until the control accuracy of each valve meets the predetermined requirements.

7. The simulation control device for coal heavy medium according to claim 6, characterized in that: The process optimization module includes: A comparison and analysis unit is used to retrieve the AI ​​learning file and the data launched by the AI ​​system itself for real-time comparison and analysis and verification, and feed back the results to the process reconstruction unit; wherein the comparison and analysis content includes real-time analysis of the temperature change of the product coal, real-time analysis of the moisture content of the product coal, real-time analysis of the composition of the product coal, real-time analysis of the size of the product coal, and real-time analysis of the coal quality of the product coal; The process reconstruction unit is used to reconstruct the process of each coal flow based on the purpose of reducing energy consumption.

Images