A copper side-blown smelting process simulation system and method based on an image recognition method

Abstract

The application provides a copper side-blown smelting process simulation system and method based on an image recognition method, wherein the simulation system comprises a gas supply subsystem, a water simulation subsystem, an image recognition subsystem and a parameter control subsystem connected in sequence; the gas supply subsystem is used for storing and providing compressed air; the water simulation subsystem is used for simulating the copper side-blown smelting process and generating simulation images and numerical flow field images; the image recognition subsystem is used for recognizing and processing the simulation images and the numerical flow field images, and obtaining the multiphase material migration and transfer law of the copper side-blown smelting process simulation and the optimal gas flow; and the parameter control subsystem is used for adjusting the air flow and correcting the parameters of the copper side-blown smelting simulation process. The application exhibits the migration and transfer behavior of the multiphase material of the copper side-blown smelting process in a numerical way, and better reveals the migration and transfer law of the multiphase material of the copper side-blown smelting process.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of copper smelting process experiment equipment, in particular to a copper side-blown smelting process simulation system and method based on an image recognition method. BACKGROUND

[0002] The side-blown smelting process is a new copper smelting technology with independent intellectual property rights in China, which has many advantages such as strong raw material adaptability, low dust rate and slag copper content, low investment cost, and environmental friendliness, and has gradually been applied in the copper smelting industry in China.

[0003] In the copper side-blown smelting process, how to precisely control the migration and transmission kinetics conditions of the multiphase materials, and investigate the migration and transmission rules of the multiphase materials in the copper side-blown smelting process, is one of the key technical problems of the copper side-blown smelting process. At present, scholars at home and abroad mostly use CFD numerical simulation to carry out multi-physical field simulation analysis on the copper side-blown smelting process, so as to optimize the kinetics conditions and furnace type results of the copper side-blown smelting process, but it is often difficult to characterize the migration and transmission characteristics of the mass, momentum and energy of the multiphase materials in the copper side-blown furnace, and it is difficult to reveal the migration and transmission rules of the multiphase materials in the copper side-blown smelting process. SUMMARY

[0004] In view of this, the present application provides a copper side-blown smelting process simulation system and method based on an image recognition method to solve the above-mentioned problems in the prior art.

[0005] In one aspect to achieve the above object, the present application provides a copper side-blown smelting process simulation system based on an image recognition method, which comprises a gas supply subsystem, a water simulation subsystem, an image recognition subsystem and a parameter control subsystem connected in sequence.

[0006] The gas supply subsystem is used to store and provide compressed air.

[0007] The water simulation subsystem is used to simulate the copper side-blown smelting process and generate simulation images and numerical flow field images.

[0008] The image recognition subsystem is used to recognize and process the simulation images and numerical flow field images, and obtain the migration and transmission rules of the multiphase materials in the copper side-blown smelting process simulation and the optimal gas flow.

[0009] The parameter control subsystem is used to adjust the air flow and correct the parameters of the copper side-blown smelting simulation process.

[0010] Optionally, the air supply subsystem comprises an air compressor, an air tank, an air dryer and a control cabinet connected in sequence, the air compressor is used for compressing air and storing it into the air tank, the air tank is used for storing compressed air, the air dryer is used for drying the compressed air in the air tank and then conveying it to the control cabinet, and the control cabinet is used for conveying the dried compressed air to the water simulation subsystem.

[0011] Optionally, the water simulation subsystem comprises a copper side-blown smelting water model device, the size of the copper side-blown smelting water model device is 1 / 10 of the size of an industrial production reactor, the modified Froude number in the copper side-blown smelting simulation process of the copper side-blown smelting water model device is equal to the modified Froude number of the industrial production reactor, the copper side-blown smelting water model device comprises a plurality of air inlets and an air outlet, the height of the copper matte and the slag layer is set according to production requirements, and the interaction between the slag layer and the material and the influence of the chemical reaction are ignored; in the copper side-blown smelting simulation process of the copper side-blown smelting water model device, an air-water system is used to simulate an oxygen-enriched air-melt system to obtain the flow characteristics of the fluid in the simulation process.

[0012] The water simulation subsystem further comprises an industrial camera and a client, the simulation image in the simulation process of the copper side-blown smelting water model device is obtained through the industrial camera, the client adopts modeling software to perform digital modeling and grid division on the copper side-blown smelting water model device to obtain a device geometric model and a numerical flow field image of the device geometric model; and the water simulation subsystem transmits the simulation image and the numerical flow field image to the image recognition subsystem.

[0013] Optionally, the image recognition subsystem comprises an extraction module and an identification module, the extraction module adopts a convolutional neural network to construct a feature extraction model, divides the simulation image into a training set and a test set, trains the feature extraction model through the training set, inputs the test set into the feature extraction model after the training is completed, the feature extraction model extracts the feature image of the material in the test set image through a convolutional layer, saves the feature image in a picture format, classifies the feature image through a fully connected layer and a classifier, and finally outputs the feature image of the change of different kinds of materials;

[0014] The identification module identifies and classifies the feature image of the change of different kinds of materials, constructs a content change curve graph of the same material, obtains the migration and transfer rule of the multi-phase material in the copper side-blown smelting process through the content change curve graph, and simultaneously analyzes the numerical flow field image through the identification module to obtain the best gas flow in the simulation of the copper side-blown smelting process.

[0015] Optionally, the parameter control subsystem uses a feedforward and feedback calculation method to track material changes and calculates the material change rate during the copper side-blown smelting simulation process. By calculating the deviation between the actual material change rate and the theoretical change rate during the simulation process, the parameters of the copper side-blown smelting water model device are corrected.

[0016] The feedforward and feedback calculations include calculating the material change ratio based on the material quantity change and the interval time, and the parameter control subsystem adjusts the air flow rate of the air supply subsystem in real time based on the material change rate.

[0017] On the other hand, to achieve the above objectives, this invention proposes a method for simulating the copper side-blown smelting process based on image recognition, comprising the following steps:

[0018] Store and release compressed air;

[0019] Using the released compressed air, the copper side-blown smelting process was simulated using a copper side-blown smelting water model device. The air flow rate and parameters of the copper side-blown smelting simulation process were adjusted and corrected, and simulation images and numerical flow field images were generated.

[0020] The simulated images and numerical flow field images are identified and processed to obtain the multiphase material migration and transfer laws and the optimal gas flow rate in the simulated copper side-blown smelting process.

[0021] Optionally, the process of storing and releasing compressed air includes:

[0022] An air compressor is used to compress the air and store it in an air storage tank. An air dryer is used to dry the compressed air in the air storage tank. A control cabinet is used to deliver the dried compressed air to the copper side-blown smelting water model device.

[0023] Optionally, the process of generating simulated images and numerical flow field images includes:

[0024] Simulation images of the copper side-blown smelting water model device during the simulation process are acquired using an industrial camera. The copper side-blown smelting water model device is digitally modeled and meshed using modeling software to obtain the device's geometric model and numerical flow field images of the device's geometric model.

[0025] Optionally, the process of identifying and processing the simulated image and the numerical flow field image includes:

[0026] A feature extraction model is constructed using a convolutional neural network. The simulated images are divided into a training set and a test set. The feature extraction model is trained using the training set. After training, the test set is input into the feature extraction model. The feature extraction model extracts the feature images of materials in the test set images through convolutional layers. The feature images are saved as image formats and classified through fully connected layers and a classifier. Finally, feature images of changes in different types of materials are output.

[0027] The characteristic images of changes in different types of materials are identified and classified, and a content change curve of the same material is constructed. The migration and transfer law of multiphase materials in the copper side-blown smelting process is obtained through the content change curve. At the same time, the numerical flow field image is analyzed to obtain the optimal gas flow rate for simulating the copper side-blown smelting process.

[0028] Optionally, the process of adjusting the airflow and correcting parameters in the copper side-blown smelting simulation includes:

[0029] The material changes are tracked by using feedforward and feedback calculations, and the material change rate is calculated during the copper side-blown smelting simulation process. The deviation between the actual material change rate and the theoretical change rate during the simulation process is calculated, and the parameters of the copper side-blown smelting water model device are corrected.

[0030] The feedforward and feedback calculations include calculating the material change ratio based on the material quantity change and the interval time, and adjusting the air flow rate of the air supply subsystem in real time based on the material change rate.

[0031] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention uses image recognition to digitally analyze the migration and transfer characteristics of multiphase materials such as mass, momentum and energy in copper side-blown furnace, constructs a mathematical model of multiphase material migration and transfer in the copper side-blown smelting process, and uses a copper side-blown smelting water model experimental device as the sample collection and testing object to obtain experimental process parameters and migration and transfer behavior characteristic data. Then, multi-factor simulation and water model experiments are carried out to verify, adjust and optimize the mathematical model, and the migration and transfer behavior of multiphase materials in the copper side-blown smelting process is presented in a numerical way, which better reveals the migration and transfer law of multiphase materials in the copper side-blown smelting process. Attached Figure Description

[0032] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0033] Figure 1This is a schematic diagram of the copper side-blowing smelting process simulation system based on image recognition method in an embodiment of the present invention;

[0034] Figure 2 This is a schematic diagram of the copper side-blowing smelting process simulation method based on image recognition in an embodiment of the present invention. Detailed Implementation

[0035] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0036] Example 1

[0037] This embodiment proposes a simulation system for copper side-blown smelting process based on image recognition methods, such as... Figure 1 As shown, the system structure includes a gas supply subsystem, a water simulation subsystem, an image recognition subsystem, and a parameter control subsystem connected in sequence.

[0038] The system includes an air supply subsystem for storing and supplying compressed air; a water simulation subsystem for simulating the copper side-blown smelting process and generating simulated images and numerical flow field images; an image recognition subsystem for recognizing and processing the simulated images and numerical flow field images to obtain the multiphase material migration and transfer laws and the optimal gas flow rate in the copper side-blown smelting process simulation; and a parameter control subsystem for adjusting the air flow rate and correcting parameters in the copper side-blown smelting simulation process.

[0039] The specific functions and working principles of each component of the system are as follows:

[0040] The air supply subsystem includes an air compressor, an air storage tank, an air dryer, and a control cabinet connected in sequence. During the copper side-blowing smelting process, the air compressor compresses the air and stores it in the air storage tank. The compressed air is stored in the air storage tank, and the compressed air in the air storage tank is dried by the air dryer before being delivered to the control cabinet. Finally, the control cabinet delivers the dried compressed air to the water simulation subsystem to simulate the process.

[0041] The water simulation subsystem includes a copper side-blown smelting water model device. To ensure that the simulation device is geometrically similar to the reactor in normal industrial production, the size of the copper side-blown smelting water model device is designed to be one-tenth the size of the industrial production reactor.

[0042] Secondly, in the copper side-blown smelting water model device, the driving force for the copper side-blown smelting reaction system is mainly the impact force brought by the airflow blown in from the side of the reactor and the buoyancy brought by the generated bubbles. Therefore, the corrected Frode number in the device simulation process must be equal to the corrected Frode number when the industrial production reactor is running normally in order to ensure that the model is similar to the actual production reactor in terms of dynamics. At the same time, based on the equality of the corrected Frode number, the aeration rate of the model can be calculated.

[0043] Based on the above reasons, the copper side-blown smelting water model device is equipped with several air inlets and one air outlet. The height of the copper matte and slag layer is set according to the production requirements, and the interaction between the slag layer and the material and the influence of chemical reactions are ignored. During the copper side-blown smelting simulation process, the air-water system is used to simulate the oxygen-rich air-melt system to obtain the flow characteristics of the fluid during the simulation process.

[0044] To enable real-time recording and analysis of the copper side-blown smelting simulation process, the water simulation subsystem is equipped with an industrial camera and a client. The industrial camera acquires simulated images of the copper side-blown smelting water model device during the simulation process, facilitating the analysis of material changes during the simulation. The real-time client uses modeling software to digitally model and mesh the copper side-blown smelting water model device, acquiring the geometric model of the device and its numerical flow field image. After image acquisition, the water simulation subsystem transmits the real-time acquired simulation images and numerical flow field images to the image recognition subsystem for analysis and processing.

[0045] The image recognition subsystem includes an extraction module and a recognition module. The extraction module uses a convolutional neural network to construct a feature extraction model and divides the simulated images into a training set and a test set. The feature extraction model is trained using the training set. After training, the test set is input into the feature extraction model. The feature extraction model extracts the feature images of the materials in the test set images through convolutional layers, saves the feature images as image formats, and classifies the feature images through fully connected layers and a classifier. Finally, it outputs feature images of different types of materials.

[0046] The identification module identifies and classifies the characteristic images of different types of materials, constructs the content change curve of the same material, and obtains the migration and transfer law of multiphase materials in the copper side-blown smelting process through the content change curve. At the same time, the identification module analyzes the numerical flow field image to obtain the optimal gas flow rate for simulating the copper side-blown smelting process.

[0047] In the parameter control process of copper side-blown smelting simulation, the parameter control subsystem uses feedforward and feedback calculation to track material changes and calculates the material change rate in the copper side-blown smelting simulation process. The deviation between the actual calculated material change rate and the theoretical change rate is calculated, and the parameters of the copper side-blown smelting water model device are corrected based on the calculation results.

[0048] Specifically, the feedforward and feedback calculations include calculating the material change ratio based on changes in material quantity and time intervals, adjusting the air flow rate of the gas supply subsystem in real time based on the material change rate, comparing the material change rate with the change rate of the theoretical simulation process, and adjusting parameters such as the structure and inlet / outlet quantity of the copper side-blown smelting water model device based on the comparison deviation.

[0049] This embodiment uses image recognition to digitally analyze the migration and transfer characteristics of multiphase materials in a copper side-blown furnace, including mass, momentum, and energy. A mathematical model of multiphase material migration and transfer in the copper side-blown smelting process is constructed. A water model experimental device for copper side-blown smelting is used as the sample collection and testing object to obtain experimental process parameters and migration and transfer behavior characteristic data. Multi-factor simulation and water model experiments are then carried out to verify, adjust, and optimize the mathematical model. The migration and transfer behavior of multiphase materials in the copper side-blown smelting process is presented in a numerical way, which better reveals the migration and transfer law of multiphase materials in the copper side-blown smelting process.

[0050] Example 2

[0051] This embodiment proposes a simulation method for the copper side-blown smelting process based on image recognition, such as... Figure 2 As shown, it includes the following steps:

[0052] Store and release compressed air;

[0053] Using the released compressed air, the copper side-blown smelting process was simulated using a copper side-blown smelting water model device. The air flow rate and parameters of the copper side-blown smelting simulation process were adjusted and corrected, and simulation images and numerical flow field images were generated.

[0054] By identifying and processing simulated images and numerical flow field images, the migration and transport laws of multiphase materials and the optimal gas flow rate in the simulated copper side-blown smelting process are obtained.

[0055] In a preferred embodiment of this application, the process of storing and releasing compressed air includes:

[0056] An air compressor is used to compress the air and store it in an air tank. An air dryer is used to dry the compressed air in the air tank. A control cabinet is used to deliver the dried compressed air to the copper side-blown smelting water model device.

[0057] As a preferred embodiment of this application, the process of generating simulated images and numerical flow field images includes:

[0058] Simulation images of the copper side-blown smelting water model device were acquired using an industrial camera. The copper side-blown smelting water model device was digitally modeled and meshed using modeling software to obtain the device's geometric model and numerical flow field images.

[0059] In a preferred embodiment of this application, the process of identifying and processing simulated images and numerical flow field images includes:

[0060] A feature extraction model is constructed using a convolutional neural network. The simulated images are divided into a training set and a test set. The feature extraction model is trained using the training set. After training, the test set is input into the feature extraction model. The feature extraction model extracts the feature images of the materials in the test set images through convolutional layers. The feature images are saved as image formats and classified through fully connected layers and a classifier. Finally, the feature images of different types of materials are output.

[0061] The characteristic images of changes in different types of materials are identified and classified, and the content change curve of the same material is constructed. The migration and transfer law of multiphase materials in the copper side-blown smelting process is obtained through the content change curve. At the same time, the numerical flow field image is analyzed to obtain the optimal gas flow rate for simulating the copper side-blown smelting process.

[0062] As a preferred embodiment of this application, the process of adjusting the airflow and correcting parameters in the copper side-blown smelting simulation process includes:

[0063] The material changes are tracked by using feedforward and feedback calculations, and the material change rate is calculated in the copper side-blown smelting simulation process. The deviation between the actual material change rate and the theoretical change rate is calculated during the simulation process, and the parameters of the copper side-blown smelting water model device are corrected.

[0064] Feedforward and feedback calculations include calculating the material change ratio based on changes in material quantity and time intervals, and adjusting the air flow rate of the air supply subsystem in real time based on the material change rate.

[0065] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. The methods disclosed in the embodiments are described simply because they correspond to the systems disclosed in the embodiments; relevant details can be found in the method section.

[0066] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the scope of the technology disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. All should be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A simulation system for copper side-blown smelting process based on image recognition method, characterized in that, It includes a gas supply subsystem, a water simulation subsystem, an image recognition subsystem, and a parameter control subsystem connected in sequence; The air supply subsystem is used to store and supply compressed air; The water simulation subsystem is used to simulate the copper side-blown smelting process and generate simulation images and numerical flow field images. The image recognition subsystem is used to identify and process the simulated image and the numerical flow field image to obtain the multiphase material migration and transfer law and the optimal gas flow rate in the simulated copper side-blown smelting process. The parameter control subsystem is used to adjust the airflow and correct parameters in the copper side-blown smelting simulation process.

2. The copper side-blown smelting process simulation system based on image recognition method according to claim 1, characterized in that, The air supply subsystem includes an air compressor, an air storage tank, an air dryer, and a control cabinet connected in sequence. The air compressor is used to compress air and store it in the air storage tank. The air storage tank is used to store compressed air. The air dryer is used to dry the compressed air in the air storage tank and then deliver it to the control cabinet. The control cabinet is used to deliver the dried compressed air to the water simulation subsystem.

3. The copper side-blown smelting process simulation system based on image recognition method according to claim 1, characterized in that, The water simulation subsystem includes a copper side-blown smelting water model device, the size of which is 1 / 10 of the size of an industrial production reactor. The corrected Frode number used in the copper side-blown smelting simulation process of the copper side-blown smelting water model device is equal to that of the industrial production reactor. The copper side-blown smelting water model device includes several air inlets and one air outlet. The heights of the copper matte layer and slag layer are set according to production requirements, and the interaction between the slag layer and the material, as well as the influence of chemical reactions, are ignored. During the copper side-blown smelting simulation process, the copper side-blown smelting water model device uses an air-water system to simulate an oxygen-rich air-melt system, obtaining the flow characteristics of the fluid during the simulation process. The water simulation subsystem also includes an industrial camera and a client. The industrial camera acquires simulated images of the copper side-blown smelting water model device during the simulation process. The client uses modeling software to digitally model and mesh the copper side-blown smelting water model device to obtain the device's geometric model, and simultaneously acquires numerical flow field images of the device's geometric model. The water simulation subsystem transmits the simulated images and the numerical flow field images to the image recognition subsystem.

4. The copper side-blown smelting process simulation system based on image recognition method according to claim 1, characterized in that, The image recognition subsystem includes an extraction module and a recognition module. The extraction module uses a convolutional neural network to construct a feature extraction model and divides the simulated image into a training set and a test set. The feature extraction model is trained using the training set. After training, the test set is input into the feature extraction model. The feature extraction model extracts the feature images of the materials in the test set image through a convolutional layer, saves the feature images as image format, and classifies the feature images through a fully connected layer and a classifier. Finally, it outputs feature images of different types of materials. The identification module identifies and classifies the characteristic images of different types of materials, constructs a content change curve of the same material, and obtains the migration and transfer law of multiphase materials in the copper side-blown smelting process through the content change curve. At the same time, the identification module analyzes the numerical flow field image to obtain the optimal gas flow rate for simulating the copper side-blown smelting process.

5. The copper side-blown smelting process simulation system based on image recognition method according to claim 1, characterized in that, The parameter control subsystem uses a feedforward and feedback calculation method to track material changes and calculates the material change rate during the copper side-blown smelting simulation process. By calculating the deviation between the actual material change rate and the theoretical change rate during the simulation process, the parameters of the copper side-blown smelting water model device are corrected. The feedforward and feedback calculations include calculating the material change ratio based on the material quantity change and the interval time, and the parameter control subsystem adjusts the air flow rate of the air supply subsystem in real time based on the material change rate.

6. A method for simulating the copper side-blown smelting process based on image recognition, characterized in that, Includes the following steps: Store and release compressed air; Using the released compressed air, the copper side-blown smelting process was simulated using a copper side-blown smelting water model device. The air flow rate and parameters of the copper side-blown smelting simulation process were adjusted and corrected, and simulation images and numerical flow field images were generated. The simulated images and numerical flow field images are identified and processed to obtain the multiphase material migration and transport laws and the optimal gas flow rate in the simulated copper side-blown smelting process.

7. The method for simulating the copper side-blown smelting process based on image recognition according to claim 6, characterized in that, The process of storing and releasing compressed air includes: An air compressor is used to compress the air and store it in an air storage tank. An air dryer is used to dry the compressed air in the air storage tank. A control cabinet is used to deliver the dried compressed air to the copper side-blown smelting water model device.

8. The method for simulating the copper side-blown smelting process based on image recognition according to claim 6, characterized in that, The process of generating simulated images and numerical flow field images includes: Simulation images of the copper side-blown smelting water model device during the simulation process are acquired using an industrial camera. The copper side-blown smelting water model device is digitally modeled and meshed using modeling software to obtain the device's geometric model and numerical flow field images of the device's geometric model.

9. The method for simulating the copper side-blown smelting process based on image recognition according to claim 6, characterized in that, The process of identifying and processing the simulated images and numerical flow field images includes: A feature extraction model is constructed using a convolutional neural network. The simulated images are divided into a training set and a test set. The feature extraction model is trained using the training set. After training, the test set is input into the feature extraction model. The feature extraction model extracts the feature images of materials in the test set images through convolutional layers. The feature images are saved as image formats and classified through fully connected layers and a classifier. Finally, feature images of changes in different types of materials are output. The characteristic images of changes in different types of materials are identified and classified, and a content change curve of the same material is constructed. The migration and transfer law of multiphase materials in the copper side-blown smelting process is obtained through the content change curve. At the same time, the numerical flow field image is analyzed to obtain the optimal gas flow rate for simulating the copper side-blown smelting process.

10. The method for simulating the copper side-blown smelting process based on image recognition according to claim 6, characterized in that, The process of adjusting airflow and correcting parameters in the simulation of copper side-blown smelting includes: The material changes are tracked by using feedforward and feedback calculations, and the material change rate is calculated during the copper side-blown smelting simulation process. The deviation between the actual material change rate and the theoretical change rate during the simulation process is calculated, and the parameters of the copper side-blown smelting water model device are corrected. The feedforward and feedback calculations include calculating the material change ratio based on the material quantity change and the interval time, and adjusting the air flow rate of the air supply subsystem in the simulation system of claim 1 in real time based on the material change rate.

Images