Intelligent control method for dust removal and consumption reduction extreme energy efficiency of iron and steel enterprises

Abstract

The invention discloses an intelligent control method for dust removal, consumption reduction and extreme energy efficiency of iron and steel enterprises. Comprising the following steps: 1) analyzing and predicting variable-working-condition dust removal air volume based on image processing; 2) establishing an electric valve deep learning model; 3) establishing an electric valve opening optimization model; and 4) establishing a regulation and control strategy based on the optimization model. According to the method, the technological process and historical data are combined, the air volume needed by each integration point is predicted in advance, the valve action is guided, the dust removal effect is judged according to machine vision, and the system is corrected and optimized. The timeliness is improved, the operation safety degree is also improved, and the operation energy consumption of the system is reduced.

Description

Technical Field

[0001] The present invention relates to an intelligent control method based on the extreme energy efficiency of the dust removal system of a steel plant. The present invention uses big data prediction, machine vision and other means to realize intelligent control of the dust removal fan, thereby achieving the purpose of energy saving and consumption reduction. Background Art

[0002] In 2019, the Ministry of Ecology and Environment issued the "Opinions on Promoting the Implementation of Ultra-low Emissions in the Steel Industry", which proposed that by 2050, the ultra-low emission transformation of steel enterprises in key areas will be basically completed, and the country will strive to complete the transformation of more than 80% of production capacity. Steel enterprises that implement ultra-low emission transformation should comprehensively strengthen the construction of automatic monitoring, process monitoring and video monitoring facilities. In order to meet the environmental protection assessment requirements for steel enterprises in the "Opinions", during the normal production process of enterprises, due to the complexity of the pipe network system and the difficulty of manual adjustment, many existing ventilation systems directly adopt full-load operation to meet the emission requirements. The dust removal system of steel enterprises is often operated in series and parallel with multiple dust collection points in multiple production units, and some dust collection points are in intermittent operation due to operational needs. The full-load operation makes the actual air volume of some ventilation points far greater than the required air volume, increasing the additional air volume, resulting in increased energy consumption of fan operation.

[0003] At present, the intelligent control of industrial ventilation is mainly focused on automatic monitoring, and few intelligent adjustment models are introduced into the control system. In addition, most of the control methods are based on PID for air volume adjustment, and the adjustment speed is relatively slow. Moreover, the overall system operation is not taken into consideration, which can easily cause a large total pressure difference in the system and high energy consumption of the fan. At the same time, there are many high-temperature working areas in steel enterprises, and it is dangerous for personnel to make on-site adjustments. Summary of the invention

[0004] In order to overcome the above-mentioned defects, the purpose of the present invention is to provide an intelligent control method for dust removal and energy reduction in steel enterprises with ultimate energy efficiency.

[0005] To achieve the above-mentioned purpose, the present invention provides an intelligent control method for dust removal and energy reduction in a steel enterprise, comprising the following steps:

[0006] 1) Analysis and prediction of dust removal air volume under variable working conditions based on image processing;

[0007] 2) Establish a deep learning model for electric valves;

[0008] 3) Establish an optimization model for the opening of electric valves;

[0009] 4) Establish a control strategy based on the optimization model.

[0010] Furthermore, the image is an image of dust concentration at the dust removal point taken by a camera at each dust removal unit.

[0011] Furthermore, the step 1) of analyzing and predicting the variable working condition dust removal air volume based on image processing is specifically as follows: utilizing the videos, pictures and actual dust removal air volume data taken by the cameras at each dust removal unit, combined with the working condition characteristics of the above-mentioned process, to determine the reasonable dust removal air volume under different working conditions of the dust generating point.

[0012] Furthermore, the step 2) establishing a deep learning model for electric valves is specifically as follows: collecting historical data on existing valve opening and air volume, establishing a theoretical relationship between electric valve opening, flow, impedance, and pressure, establishing a deep learning model for electric valves, and predicting and learning the relationship between valve opening and air volume under different pressures.

[0013] Furthermore, the step 3) of establishing an electric valve opening optimization model is specifically as follows: combining the pipe network wind model and the regulating valve deep learning model, taking the minimum most unfavorable loop pressure difference as the optimization objective function, establishing a ventilation system regulating valve optimization model, and obtaining the regulating valve opening of each branch pipeline after optimization.

[0014] Furthermore, the step 4) establishes a control strategy based on the optimization model as follows: based on centralized management and control of the production process, the collection of process data, combined with the system's analysis of the air volume required at the dust collection point, through the electric valve deep learning model, the electric valve opening adjustment model makes advance decisions and controls on the valve adjustment of the dust removal system; at the same time, the control results are analyzed based on machine vision technology, and the results of the analysis are used to iteratively optimize the control model again.

[0015] Furthermore, the step 3) of establishing the electric valve opening optimization model comprises the following steps:

[0016] 71) Use Kirchhoff’s voltage law and current law to establish a wind model for the pipe network;

[0017] 72) Write the pipeline impedance information in the pipeline network, set the total flow equal to the total demand flow, the branch flow equal to the branch demand flow, and the regulating valve opening fully open;

[0018] 73) Solve the flow rate at the current opening according to the pipe network wind model, solve the current valve impedance according to the regulating valve model, and record the absolute value d of the difference between the current flow rate and the required flow rate;

[0019] 74) If the difference is less than a fixed value d, no adjustment is performed;

[0020] If the difference is greater than the fixed value d, adjustment is required; find the most unfavorable loop, that is, the loop with the largest pressure difference, and take its minimum pressure difference as the optimization goal. After optimizing the platform control valve model and calculating the prediction results of the electric valve deep learning model, the opening of each valve, flow rate and pressure drop of the most unfavorable loop are obtained.

[0021] Furthermore, the method further comprises the following steps:

[0022] 81) Adjust valve opening. Use the network hydraulic model and valve model to calculate the valve opening, and adjust the valve according to the calculation result.

[0023] 82) Adjust the fan frequency. Adjust the fan frequency according to the calculated most unfavorable loop pressure drop and the required total flow.

[0024] 83) The camera analyzes the control results based on machine vision technology, updates and iterates the electric valve opening optimization model, and optimizes the electric valve opening optimization model again.

[0025] The present invention combines the process and historical data to predict the air volume required for each integration point and guide the valve action in advance, and judges the dust removal effect based on machine vision to correct and optimize the system. While improving timeliness, it also improves the safety of operations and reduces system operation energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a control process diagram of the present invention.

[0027] Figure 2 Kirchhoff's voltage law and current law are used to establish the wind (water) power model diagram of the pipeline network.

[0028] Figure 3 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0029] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0030] In the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship are based on the orientation or position relationship shown in the drawings, and are 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 operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0031] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0032] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0033] The present invention aims to provide an intelligent control system capable of feedforward prediction, combining process and historical data to predict the air volume required for each integration point and guide valve action in advance, and to judge the dust removal effect based on machine vision to correct and optimize the system. While improving timeliness, it also improves the safety of operations and reduces system operation energy consumption.

[0034] The method of the present invention comprises the steps of:

[0035] 1) Analysis and prediction of dust removal air volume under variable working conditions based on image processing.

[0036] By using the recorded videos, pictures and actual dust removal air volume data, combined with the working conditions of the above-mentioned process, the reasonable dust removal air volume under different working conditions at the dust generating point is determined.

[0037] 2) Establish a deep learning model for electric valves

[0038] Collect historical data on existing valve opening and air volume, and use a deep learning model to predict and learn the relationship between valve opening and air volume under different pressures.

[0039] 3) Establishment of electric valve opening optimization model

[0040] Combining the pipe network wind model and the deep learning model of the regulating valve, an optimization model of the ventilation system regulating valve is established with the minimum of the most unfavorable loop pressure difference as the optimization objective function, and the regulating valve opening of each branch pipeline after optimization is obtained.

[0041] 4) Establishment of control strategy based on optimization model

[0042] Based on the centralized management and control of the production process, the collection of process data, combined with the system's analysis of the air volume required for the dust collection point, the electric valve deep learning model and the electric valve opening adjustment model make advance decisions and control the valve adjustment of the dust removal system. At the same time, the control results are analyzed based on machine vision technology, and the results of the analysis are used to update and iterate the control model again. The control process is as follows: Figure 1 shown.

[0043] Example

[0044] A) Analyze the main factors that affect the dust removal air volume when the process state of the intermittent dust removal point changes. Analyze the main factors that affect the dust removal air volume at the continuously running dust removal point. Install a camera at each dust removal unit to take photos of the dust concentration at the dust removal point. Combined with the above influencing factors, conduct on-site tests under different process conditions to compare and analyze the dust removal effects under different air volumes. Determine the optimal dust removal air volume that meets the dust removal requirements at each dust removal point under different process conditions.

[0045] B) According to the characteristics of the on-site regulating valve, the theoretical relationship between the electric valve opening, flow, impedance and pressure is established, and a deep learning model for the electric valve is established to predict and learn the relationship between the valve opening and air volume under different pressures.

[0046] C) Use Kirchhoff's voltage law and current law to establish a wind model for the pipe network, such as Figure 2 shown.

[0047] According to the pipe network wind model and the performance parameters of the fan and other input parameters, the flow and pressure drop of each branch in the pipe network system are solved. The pipe network mainly consists of the exhaust point C, the regulating valve V and the pipeline. Among them, the valve impedance is S fi , branch pipe impedance is S pi , Main pipeline impedance S gi .

[0048]

[0049] D) Initialization. Write the pipeline impedance information in the pipeline network, set the total flow equal to the total demand flow, the branch flow equal to the branch demand flow, and the regulating valve opening fully open.

[0050] E) Solve the current branch flow. Solve the flow at the current opening according to the pipe network wind model, solve the current valve impedance according to the regulating valve model, and record the absolute value d of the difference between the current flow and the required flow.

[0051] F) If the difference is less than a fixed value d, no adjustment is made. If the difference is greater than the fixed value d, adjustment is required. Find the most unfavorable loop, that is, the loop with the largest pressure difference, and take its minimum pressure difference as the optimization goal. After optimizing the platform control valve model and calculating the prediction results of step B, the valve opening, flow rate and pressure drop of the most unfavorable loop are obtained.

[0052] G) Adjust valve opening. Use the pipe network wind model and valve model to calculate the valve opening, and adjust the valve according to the calculation result.

[0053] H) Adjust the fan frequency. Adjust the fan frequency according to the calculated most unfavorable loop pressure drop and the required total flow.

[0054] The camera analyzes the control results based on machine vision technology, updates and iterates the electric valve opening optimization model, and optimizes the electric valve opening optimization model again.

[0055] The present invention is described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of ordinary technicians in the field without departing from the purpose of the present invention. Many other changes and modifications that do not depart from the concept and scope of the present invention should be regarded as the protection scope of the present invention.

[0056] In the description of this specification, specific features, structures, materials or characteristics may be combined in an appropriate manner in any one or more embodiments or examples.

[0057] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An intelligent control method for dust removal and energy reduction in steel enterprises, characterized in that: The method comprises the following steps: 1) Analysis and prediction of dust removal air volume under variable working conditions based on image processing; 2) Establish a deep learning model for electric valves; 3) Establish an optimization model for the opening of electric valves; 4) Establish a control strategy based on the optimization model.

2. The intelligent control method for dust removal and energy reduction in a steel enterprise as claimed in claim 1 is characterized in that: The image is an image of dust concentration at the dust removal point taken by a camera at each dust removal unit.

3. The intelligent control method for dust removal and energy reduction in a steel enterprise as claimed in claim 1 is characterized in that: The step 1) of analyzing and predicting the variable working condition dust removal air volume based on image processing is specifically as follows: using the videos, pictures and actual dust removal air volume data taken by the cameras at each dust removal unit, combined with the working condition characteristics of the above-mentioned process, to determine the reasonable dust removal air volume under different working conditions of the dust generating point.

4. The intelligent control method for dust removal and energy reduction in a steel enterprise as claimed in claim 1 is characterized in that: The step 2) establishing a deep learning model for electric valves is specifically as follows: collecting historical data on existing valve opening and air volume, establishing a theoretical relationship between electric valve opening, flow, impedance, and pressure, establishing a deep learning model for electric valves, and predicting and learning the relationship between valve opening and air volume under different pressures.

5. The intelligent control method for dust removal and energy reduction in a steel enterprise as claimed in claim 1 is characterized in that: The step 3) of establishing an electric valve opening optimization model is specifically as follows: combining the pipe network wind model and the regulating valve deep learning model, taking the minimum most unfavorable loop pressure difference as the optimization objective function, establishing a ventilation system regulating valve optimization model, and obtaining the regulating valve opening of each branch pipeline after optimization.

6. The intelligent control method for dust removal and energy reduction in a steel enterprise as claimed in claim 1 is characterized in that: The step 4) establishes a control strategy based on the optimization model, specifically: based on the centralized management and control of the production process, the collection of process data, combined with the system's analysis of the air volume required at the dust collection point, through the electric valve deep learning model, the electric valve opening adjustment model makes advance decisions and controls on the valve adjustment of the dust removal system; at the same time, the control results are analyzed based on machine vision technology, and the results of the analysis are used to iteratively optimize the control model again.

7. The intelligent control method for dust removal and energy reduction in a steel enterprise as claimed in claim 5 is characterized in that: The step 3) of establishing the electric valve opening optimization model comprises the following steps: 71) Use Kirchhoff’s voltage law and current law to establish a wind model for the pipe network; 72) Write the pipeline impedance information in the pipeline network, set the total flow equal to the total demand flow, the branch flow equal to the branch demand flow, and the regulating valve opening fully open; 73) Solve the flow rate at the current opening according to the pipe network wind model, solve the current valve impedance according to the regulating valve model, and record the absolute value d of the difference between the current flow rate and the required flow rate; 74) If the difference is less than a fixed value d, no adjustment is performed; If the difference is greater than the fixed value d, adjustment is required; find the most unfavorable loop, that is, the loop with the largest pressure difference, and take its minimum pressure difference as the optimization goal. After optimizing the platform control valve model and calculating the prediction results of the electric valve deep learning model, the opening of each valve, flow rate and pressure drop of the most unfavorable loop are obtained.

8. The intelligent control method for dust removal and energy reduction in a steel enterprise as claimed in claim 7, characterized in that: The following steps are also included: 81) Adjust valve opening. Use the network hydraulic model and valve model to calculate the valve opening, and adjust the valve according to the calculation result. 82) Adjust the fan frequency. Adjust the fan frequency according to the calculated most unfavorable loop pressure drop and the required total flow. 83) The camera analyzes the control results based on machine vision technology, updates and iterates the electric valve opening optimization model, and optimizes the electric valve opening optimization model again.

Images