An intelligent dust removal system for electric furnace based on flow field simulation and image recognition

Abstract

The present invention relates to an electric furnace intelligent dust removal system based on flow field simulation and image recognition, and belongs to the field of industrial dust removal. The system includes an image acquisition device, a flow field acquisition device, a flow field simulation model, a machine learning model and a fan adjustment device; wherein the flow field simulation model is used to determine the correlation between smoke characteristics, smoke parameters and fan parameters according to smoke and dust characteristics; the machine learning model is used to obtain the last fan parameters corresponding to the last fan operation, and determine the smoke characteristics according to the smoke image, and determine the rationality of the last fan operation parameters according to the smoke characteristics, smoke parameters and the flow field simulation model. The present invention integrates the various operating parameters of the dust removal system into the control module according to the smoke thermal imaging characteristics, flow field boundary parameters, and dust removal system, and can realize the adjustment of the dust removal system fan parameters in an automated manner, improve work efficiency, reduce the limitations of manual operation, improve the dust removal effect in the factory, and save fan energy consumption.

Description

Technical Field

[0001] The invention belongs to the field of industrial dust removal, and relates to an electric furnace intelligent dust removal system based on flow field simulation and image recognition. Background Art

[0002] Many pollutants will be emitted during electric furnace smelting, including solid metal, metal compound smoke, carbon monoxide (CO), volatile organic compounds, dioxins and furans and other particles. Smelting one ton of steel will produce 15 to 25 kg of dust. The emissions generated by the electric arc furnace are controlled by the dust removal system. The most common situation is to use a primary flue gas suction control system and a large roof hood to collect emissions. The primary flue gas is extracted from the electric furnace through a pipeline and enters the combustion settling chamber to burn CO and separate large particles of dust by gravity. After cooling, the fine dust is removed by a bag filter. Then the escaping flue gas from the electric furnace steel plant is difficult to capture. The dust emission research of the German electric furnace steel plant shows that about 66% of the emissions are escaping flue gas. In addition to the primary flue gas, other flue gas emissions are called secondary flue gas. The collection and treatment of these secondary flue gases include large roof hoods, side suction vents, etc. The capture efficiency of secondary flue gas is reduced due to the disturbance caused by the movement of vehicles and the pollutants that escape during the feeding of the basket. The horizontal airflow, other building openings, moving vehicles in the workshop, temperature gradient in the workshop, other ventilation devices in the workshop, low pressure, high humidity and strong wind weather will affect the capture of secondary flue gas. The current secondary dust removal control system is mainly controlled according to the electric furnace smelting cycle, which is too extensive and is not ideal for dust removal effect and fan energy saving.

[0003] Therefore, there is an urgent need for an electric furnace intelligent dust removal system that can improve work efficiency. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide an electric furnace intelligent dust removal system based on flow field simulation and image recognition. According to the flue gas thermal imaging characteristics, flow field boundary parameters, and various operating parameters of the dust removal system are integrated into the controller, the fan parameters of the dust removal system can be adjusted in an automated manner, thereby improving work efficiency. Workers only need to pay attention to the working conditions of the intelligent dust removal system, further reducing the limitations of manual operation, improving the dust removal effect in the factory, and saving fan energy consumption.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] Solution 1: An electric furnace intelligent dust removal system based on flow field simulation and image recognition, comprising a collection device, a controller and a fan adjustment device; the collection device comprises an image collection device and a flow field collection device; the controller comprises a flow field simulation model and a machine learning model;

[0007] The image acquisition device is installed above the roof hood, with its camera passing through the roof hood and facing downward, and is used to collect smoke and dust images in the roof hood;

[0008] The flow field acquisition device is installed between the roof hood and the furnace mouth close to the roof hood, and its probe extends to the center position below the roof hood to collect flow field simulation boundary parameters, that is, flue gas parameters;

[0009] The flow field simulation model is used to determine the correlation between smoke and dust characteristics, smoke parameters and fan parameters according to smoke and dust characteristics;

[0010] The machine learning model is used to obtain the last fan parameters corresponding to the last fan operation, determine the smoke characteristics according to the smoke image, and determine the rationality of the last fan operation parameters according to the smoke characteristics, smoke parameters and flow field simulation model;

[0011] The fan adjustment device is used to use the previous fan parameters as execution adjustment parameters when the fan parameter determination device determines that the result is reasonable, and perform the fan frequency conversion operation according to the execution adjustment parameters.

[0012] Preferably, the image acquisition device adopts an industrial camera, and the industrial camera is equipped with a fill light and a lens cleaning device.

[0013] Preferably, the image acquisition device is also used to determine the fan frequency conversion parameters according to the smoke form and smoke concentration.

[0014] Preferably, the flow field acquisition device comprises a probe and a probe driving mechanism, and the probe is driven by the probe driving mechanism to measure the airflow or airflow signals at different positions in the airflow channel.

[0015] Preferably, the system also includes a data storage device for storing historical data, including: flue gas parameters (obtained from the flow field acquisition device, including flue gas temperature, flue gas flow rate and flue gas pressure), smoke dust characteristics (obtained from the image acquisition device), smoke dust images, this time's recommended fan parameters, execution fan parameters, fan parameter adjustment information, flue gas parameter adjustment information, acquisition time of fan parameter adjustment information, acquisition time of flue gas parameter adjustment information, input source of fan parameter adjustment information, input source of flue gas parameter adjustment information.

[0016] Solution 2: An intelligent dust removal method for an electric furnace based on flow field simulation and image recognition, specifically comprising the following steps:

[0017] S1: Record the operation status of the electric furnace, the travel position of the roof hood, the flue gas flow rate at the top of the electric furnace, the dust form and the fan speed during the secondary dust removal during the production operation of the electric furnace under different smelting conditions, and build a historical operation database.

[0018] S2: According to the actual geometric structure of the furnace roof hood, the inlet conditions of the flue gas and the dust content, a physical model of the furnace roof dust removal in the flow field simulation model is established (referring to the electric furnace and roof hood physical model, which belongs to the flow field simulation model).

[0019] After the physical model is established, simulations are performed for different flow rates, fan speeds and smoke concentrations, the flow field information, dust removal efficiency and fan model in the dust removal area are calculated, and a flow field simulation database is constructed.

[0020] S3: Data fusion is performed on the historical operation database and the flow field simulation database to construct a big data model sample database. The big data model sample database mainly includes information such as fan speed, flue gas flow, smoke concentration, flow field information, dust removal efficiency and smelting conditions.

[0021] S4: The model sample data in the big data model sample database is trained through the machine learning model to calculate the fan operating parameters in the dust removal area.

[0022] S5: Based on the real-time operation data of the reaction dust removal area obtained from the real-time operation database, the real-time operation data is matched with the model sample data for similarity through the machine learning model, and the machine learning model is continuously corrected online to obtain the optimal fan operation parameters; when the real-time operation data on site is transmitted to the trained machine learning model, it is first matched with the big data model sample database for similarity. If the similarity is high, such as 90% or above, the machine learning model directly outputs the fan parameters; if the matching degree is low, such as 0% to 90%, the flow field simulation model simulates and calculates the corresponding fan parameters under the same working conditions, and transmits them to the machine learning model for output. After that, the real-time operation data with a low matching degree is transmitted to the model sample database for data fusion to obtain new model sample data. After that, the machine learning model performs offline learning and training on the new model sample data, updates the machine learning model, and provides a more accurate model for the follow-up.

[0023] S6: The controller calls the optimal fan operating parameters obtained by the machine learning model, and controls the fan speed in the dust removal area according to the fan operating parameters to achieve precise frequency conversion of the dust removal area; at the same time, the result data obtained under the optimal parameter conditions is fed back to the real-time operation database to supplement the richness of the data, making the machine learning training more accurate.

[0024] Furthermore, in step S2, a physical model of dust removal on the roof of the electric furnace is established, specifically including: using a two-fluid model coupled with a particle model to numerically simulate the movement of flue gas and dust in the dust removal area between the furnace mouth and the smoke hood on the roof of the electric furnace, determine the relationship between flow conditions and dust removal efficiency, and obtain sample data beyond the historical operation database.

[0025] The beneficial effects of the present invention are as follows: the present invention adopts an intelligent dust removal control system that combines flow field numerical simulation and image recognition. The relationship between the flow field information from the top of the electric furnace to the roof hood and the dust removal effect is given from a mechanistic perspective through flow field numerical simulation. The present invention expands the original historical operation data set through flow field numerical simulation and increases the dimension of the image recognition model input data, which is conducive to establishing a better ventilation and dust removal model and realizing precise control of the dust removal system.

[0026] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0028] Figure 1 This is a schematic diagram of the installation of the electric furnace intelligent dust removal system in this embodiment;

[0029] Figure 2 This is a structural block diagram of the electric furnace intelligent dust removal system in this embodiment.

[0030] Figure numerals: 1-image acquisition device, 2-flow field acquisition device, 3-electric furnace, 4-roof hood, 5-dust removal flue, 6-four-hole dust removal flue, 7-sedimentation chamber, 8-water-cooled flue, 9-quenching tower, 10-flue gas duct, 11-bag dust collector, 12-fan, 13-chimney. DETAILED DESCRIPTION

[0031] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0032] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0033] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "front", "rear", etc. indicate the orientation or position relationship, they are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0034] See also Figure 1-2 This embodiment provides an electric furnace intelligent dust removal system based on flow field simulation and image recognition, which specifically includes an acquisition device, a controller, a fan adjustment device and a data storage device; wherein the acquisition device includes an image acquisition device and a flow field acquisition device; the controller includes a flow field simulation model and a machine learning model.

[0035] The image acquisition device is installed above the roof hood, with its camera passing through the roof hood and facing downward, and is used to collect smoke images inside the roof hood; it is also used to determine the fan frequency conversion parameters based on the smoke shape and smoke concentration.

[0036] As a preferred embodiment, the image acquisition device in this embodiment adopts an industrial camera, and the industrial camera is equipped with a fill light and a lens cleaning device.

[0037] The flow field collection device is installed between the roof hood and the furnace mouth close to the roof hood. Its probe extends to the center position below the roof hood and is used to collect flow field simulation boundary parameters, namely, flue gas parameters.

[0038] As a preferred embodiment, the flow field acquisition device in this embodiment includes a probe and a probe driving mechanism. The probe is driven by the probe driving mechanism to measure the airflow or airflow signals at different positions in the airflow channel.

[0039] The flow field simulation model is used to determine the correlation between smoke and dust characteristics, smoke parameters and fan parameters according to the smoke and dust characteristics.

[0040] The machine learning model is used to obtain the previous fan parameters corresponding to the previous fan operation, determine the smoke characteristics based on the smoke image, and judge the rationality of the previous fan operation parameters based on the smoke characteristics, smoke parameters and flow field simulation model.

[0041] The fan adjustment device is used to use the previous fan parameters as execution adjustment parameters when the fan parameter determination device determines that the result is reasonable, and perform the fan frequency conversion operation according to the execution adjustment parameters.

[0042] A data storage device, used to store historical data, including at least one of the following: smelting time, smelting cycle, arc furnace smelting status, flue gas parameters, smoke characteristics, smoke images, this time recommended fan parameters, execution fan parameters, fan parameter adjustment information, flue gas parameter adjustment information, fan parameter adjustment information acquisition time, flue gas parameter adjustment information acquisition time, fan parameter adjustment information input source, flue gas parameter adjustment information input source.

[0043] In conjunction with the electric furnace intelligent dust removal system of this embodiment, the dust removal method thereof specifically includes the following steps:

[0044] S1: Record the operation status of the electric furnace, the travel position of the roof hood, the flue gas flow rate at the top of the electric furnace, the dust form and the fan speed during the secondary dust removal during the production operation of the electric furnace under different smelting conditions, and build a historical operation database.

[0045] S2: According to the actual geometric structure of the furnace roof hood, the inlet conditions of the flue gas and the dust content, a physical model of the furnace roof dust removal in the flow field simulation model is established (referring to the electric furnace and roof hood physical model, which belongs to the flow field simulation model).

[0046] A physical model of dust removal on the roof of an electric furnace is established, specifically including: using a two-fluid model coupled with a particle model to numerically simulate the movement of flue gas and dust in the dust removal area between the furnace mouth and the smoke hood on the roof of the furnace, determine the relationship between flow conditions and dust removal efficiency, and obtain sample data beyond the historical operation database.

[0047] After the physical model is established, simulations are performed for different flow rates, fan speeds and smoke concentrations, the flow field information, dust removal efficiency and fan model in the dust removal area are calculated, and a flow field simulation database is constructed.

[0048] S3: Data fusion is performed on the historical operation database and the flow field simulation database to construct a big data model sample database. The big data model sample database mainly includes information such as fan speed, flue gas flow, smoke concentration, flow field information, dust removal efficiency and smelting conditions.

[0049] S4: The model sample data in the big data model sample database is trained through the machine learning model to calculate the fan operating parameters in the dust removal area.

[0050] S5: Based on the real-time operation data of the reaction dust removal area obtained from the real-time operation database, the real-time operation data is matched with the model sample data for similarity through the machine learning model, and the machine learning model is continuously corrected online to obtain the optimal fan operation parameters; when the real-time operation data on site is transmitted to the trained machine learning model, it is first matched with the big data model sample database for similarity. If the similarity is high, such as 90% or above, the machine learning model directly outputs the fan parameters; if the matching degree is low, such as 0% to 90%, the flow field simulation model simulates and calculates the corresponding fan parameters under the same working conditions, and transmits them to the machine learning model for output. After that, the real-time operation data with a low matching degree is transmitted to the model sample database for data fusion to obtain new model sample data. After that, the machine learning model performs offline learning and training on the new model sample data, updates the machine learning model, and provides a more accurate model for the follow-up.

[0051] S6: The controller calls the optimal fan operating parameters obtained by the machine learning model, and controls the fan speed in the dust removal area according to the fan operating parameters to achieve precise frequency conversion of the dust removal area; at the same time, the result data obtained under the optimal parameter conditions is fed back to the real-time operation database to supplement the richness of the data, making the machine learning training more accurate.

[0052] Compared with the prior art, the present invention adopts an intelligent control system that combines flow field numerical simulation and machine learning. The relationship between the flow field information in the dust removal area and the dust removal effect is given from a mechanistic perspective through flow field numerical simulation. The present invention expands the original historical operation data set through flow field numerical simulation and increases the dimension of the machine learning model input data, which is conducive to machine learning to establish a better prediction model and realize precise control of the fan.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. An intelligent dust removal method for an electric furnace based on flow field simulation and image recognition, characterized in that: The method specifically comprises the following steps: S1: Record the operation status of the electric furnace, the travel position of the roof hood, the flue gas flow rate at the top of the electric furnace, the dust form and the fan speed during the secondary dust removal under different smelting conditions during the production operation of the electric furnace, and build a historical operation database; S2: According to the actual geometric structure of the furnace roof hood, the inlet conditions of the flue gas and the dust content, a physical model of the furnace roof dust removal in the flow field simulation model is established; After the physical model is established, simulations are performed for different flow rates, fan speeds, and smoke concentrations to calculate the flow field information, dust removal efficiency, and fan model in the dust removal area, and build a flow field simulation database; S3: Perform data fusion on the historical operation database and the flow field simulation database to build a big data model sample database; S4: training the model sample data in the big data model sample database through the machine learning model, and calculating the fan parameters in the dust removal area; S5: Based on the real-time operation data of the reaction dust removal area obtained from the real-time operation database, the real-time operation data is matched with the model sample data by similarity through the machine learning model, and the machine learning model is continuously corrected online to obtain the optimal fan parameters; when the real-time operation data on site is transmitted to the trained machine learning model, it is first matched with the big data model sample database for similarity. If the similarity is high, the machine learning model directly outputs the fan parameters; if the matching degree is low, the flow field simulation model simulates and calculates the corresponding fan parameters under the same working conditions, and transmits them to the machine learning model for output. After that, the real-time operation data with a low matching degree is transmitted to the model sample database for data fusion to obtain new model sample data. After that, the machine learning model performs offline learning training on the new model sample data to update the machine learning model; S6: The controller calls the optimal fan parameters obtained by the machine learning model, and controls the fan speed in the dust removal area according to the fan parameters to achieve precise frequency conversion of the dust removal area; at the same time, the result data obtained under the optimal parameter conditions is fed back to the real-time operation database.

2. The intelligent dust removal method for an electric furnace according to claim 1, characterized in that: In step S2, a physical model of dust removal on the roof of an electric furnace is established, specifically including: using a two-fluid model coupled with a particle model to numerically simulate the movement of flue gas and dust in the dust removal area between the furnace mouth and the smoke hood on the roof of the electric furnace, determine the relationship between flow conditions and dust removal efficiency, and obtain sample data beyond the historical operation database.

3. The intelligent dust removal method for an electric furnace according to claim 1, characterized in that: In step S3, the constructed big data model sample database includes fan speed, flue gas flow, smoke concentration, flow field information, dust removal efficiency and smelting conditions.

4. An electric furnace intelligent dust removal system based on flow field simulation and image recognition for implementing the method described in any one of claims 1 to 3, characterized in that: The system includes an acquisition device, a controller and a fan adjustment device; the acquisition device includes an image acquisition device and a flow field acquisition device; the controller includes a flow field simulation model and a machine learning model; The image acquisition device is used to acquire smoke and dust images in the roof smoke hood; The flow field acquisition device is used to acquire flow field simulation boundary parameters, namely, smoke parameters, including smoke temperature, smoke flow rate and smoke pressure; The flow field simulation model is used to determine the correlation between smoke and dust characteristics, smoke parameters and fan parameters according to smoke and dust characteristics; The machine learning model is used to obtain the last fan parameters corresponding to the last fan operation, determine the smoke characteristics according to the smoke image, and determine the rationality of the last fan parameters according to the smoke characteristics, smoke parameters and flow field simulation model; The fan adjustment device is used to use the previous fan parameters as execution adjustment parameters when the fan parameter determination device determines that the result is reasonable, and perform the fan frequency conversion operation according to the execution adjustment parameters.

5. The intelligent dust removal system for electric furnace according to claim 4 is characterized in that: The image acquisition device is installed above the roof hood, and its camera passes through the roof hood and faces downward.

6. The intelligent dust removal system for electric furnace according to claim 5, characterized in that: The image acquisition device adopts an industrial camera, and the industrial camera is provided with a fill light and a lens cleaning device.

7. The intelligent dust removal system for electric furnace according to claim 4, characterized in that: The image acquisition device is also used to determine the fan frequency conversion parameters according to the smoke form and smoke concentration.

8. The intelligent dust removal system for electric furnace according to claim 4, characterized in that: The flow field collection device is installed between the roof hood and the furnace mouth of the electric furnace, close to the roof hood, and its probe extends to the center position below the roof hood.

9. The intelligent dust removal system for electric furnace according to claim 8, characterized in that: The flow field acquisition device comprises a probe and a probe driving mechanism. The probe is driven by the probe driving mechanism to measure the airflow or airflow signals at different positions in the airflow channel.

10. The intelligent dust removal system for electric furnace according to any one of claims 4 to 9, characterized in that: The system also includes a data storage device for storing historical data, including: storing smoke characteristics acquired by the image acquisition device, smoke parameters acquired by the flow field acquisition device, and fan parameters.

Images