Intelligent control algorithm and effect monitoring method for dust removal of steel ore tank

Abstract

The invention provides an intelligent control algorithm for dust removal of a steel ore tank and an operation effect monitoring method thereof, and aims to improve the energy efficiency and emission control capability of a dust removal system of the steel ore tank. The system comprises a data acquisition module, a hierarchical control module and a multi-objective optimization module. The data acquisition module is used for acquiring dust concentration, fan rotating speed, energy consumption and equipment state signals in real time; the hierarchical control module adopts the combination of model predictive control (MPC) and fuzzy PID control to accurately adjust the rotating speed of a fan, the ash removal period and the opening degree of an air door; and the multi-objective optimization module dynamically adjusts the dust-energy consumption weight ratio according to the electricity price time period, and optimizes the balance between energy consumption and emission. The invention further comprises an operation effect monitoring method, the operation effect of the system is monitored in real time, the energy-saving effect is evaluated and optimization suggestions are provided by means of data comparison and analysis, comprehensive evaluation of the energy-saving rate and the performance, visualization of a monitoring report and the like. And green transformation and low-carbon emission targets of a dust removal system of the steel and iron ore tank can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial dust removal control, and particularly relates to an intelligent control algorithm for steel ore bin dust removal and an effect monitoring method. Background Art

[0002] As a typical high-energy-consuming and high-emission industry, the energy consumption of the sintering process in the iron and steel industry accounts for about 10%-15% of the total energy consumption of the whole process. As the core link of environmental protection treatment, the supporting dust removal system has significant shortcomings in energy efficiency optimization and monitoring. With the promotion of the "dual carbon" goal and the tightening of ultra-low emission transformation requirements, the deficiencies of the existing dust removal system in terms of refined energy consumption control, intelligent equipment coordination, and in-depth data application have gradually emerged, becoming the key bottleneck restricting the green transformation of the industry.

[0003] Currently, there are mainly three major technical defects in the energy efficiency monitoring of the sintering machine dust removal system: First, the monitoring system is fragmented. The operation parameters (such as wind pressure, current, and dust concentration) of key equipment such as dust removal fans and electric bag composite dust collectors have not achieved full-process data connection. Approximately 30% of enterprises still use manual meter reading and empirical adjustment models, resulting in difficulty in timely warning of abnormal energy consumption. For example, a test of a steel plant showed that the average operating efficiency of the dust removal fan was only 65%, but due to the lack of interlock control between the frequency converter and the air volume, the annual redundant power consumption was as high as 8 million kWh. Second, the energy efficiency evaluation dimension is single. Most enterprises only focus on macro indicators such as the dust removal power consumption per ton of sintered ore, and have not established a dynamic energy efficiency model that matches the raw material characteristics and working condition fluctuations, resulting in the dust removal system being in a state of "using a big horse to pull a small cart" for a long time. A case study showed that due to the failure to adjust the fan speed in real time according to the change of flue gas volume, the proportion of invalid power consumption in the system exceeded 22%. Third, the intelligent control ability is weak. The existing PLC control system is mostly limited to equipment start-stop and threshold alarm, lacking a multi-objective optimization algorithm based on machine learning. For example, the pulse cleaning cycle still uses fixed parameters and fails to achieve adaptive adjustment in combination with the change of dust load, resulting in excessive consumption of compressed air. The transformation practice of an enterprise showed that after introducing an intelligent cleaning strategy, the compressed air consumption decreased by 18%, but the overall intelligent transformation rate of the industry was less than 15%. Summary of the Invention

[0004] The purpose of the present invention is to solve the deficiencies existing in the prior art, and to propose an intelligent control algorithm for steel ore bin dust removal and an effect monitoring method.

[0005] To achieve the above purpose, the present invention adopts the following technical solution: An intelligent control algorithm for steel ore bin dust removal, including: A data acquisition module for real-time acquisition of dust concentration, fan speed, energy consumption, and equipment status signals; A hierarchical control module, including an upper optimization layer and a lower execution layer, where: The upper - layer optimization layer generates the reference fan speed and the dust - cleaning cycle through model predictive control (MPC), and conducts dynamic optimization based on the production plan and historical data; The lower - layer execution layer adopts fuzzy PID control to adjust the damper opening according to the real - time dust concentration and the wind pressure difference, so as to achieve precise control of the dust - removal fan; The multi - objective optimization module dynamically adjusts the dust - energy consumption weight ratio (α / β) according to the real - time working conditions, and coordinately controls the working states of the fan, dust - cleaning and ash - conveying equipment.

[0006] Preferably, in the hierarchical control module: The model predictive control (MPC) of the upper - layer optimization layer takes the production plan for the next 10 - 30 minutes as the input, with the objective function of minimizing the total energy consumption, and outputs the reference fan speed and the dust - cleaning cycle; The fuzzy PID controller of the lower - layer execution layer, whose rule base is generated by training according to historical data, controls the dust concentration error and the damper opening adjustment amount.

[0007] Preferably, the multi - objective optimization module dynamically adjusts the dust - energy consumption weight ratio according to different electricity price periods: During the peak electricity price period, by increasing α (prioritizing energy consumption reduction), the dust concentration is allowed to approach the threshold; During the off - peak electricity price period, by increasing β (prioritizing emission reduction), a safety margin for the dust concentration is reserved to ensure compliance with the emission standards.

[0008] A method for monitoring the operation effect of an intelligent control algorithm for steel ore bin dust removal, the method includes: The data comparison and analysis module is used to construct a reference energy consumption curve, record the average energy consumption and emission data in the 30 days before the algorithm is enabled, and dynamically compare with the real - time data; The energy - saving rate and performance comprehensive evaluation module regularly evaluates the optimization effect of the control system by calculating the energy - saving rate η and the comprehensive performance index Q; The visual monitoring report generation module generates a real - time data dashboard, displays the energy consumption trend, the heat map of dust concentration distribution and the change of Q value, marks abnormal events and provides optimization suggestions.

[0009] Preferably, the calculation formula of the comprehensive performance index Q is: 。

[0010] Preferably, the data acquisition module collects data in real - time through the following devices: The dust sensor, using laser scattering technology, has a measurement range of 0 - 100mg / m³ and an accuracy of ±1%; The variable - frequency fan, with a power of 55kW and a speed range of 200 - 3000rpm; Edge computing device, equipped with an Intel i7 processor, for running algorithm kernels.

[0011] Preferably, the system includes: a hardware configuration module that configures key hardware components such as dust sensors, variable-frequency fans, and edge computing devices; An intelligent control module, including a hierarchical control module, a multi-objective optimization module, and an operation effect monitoring module, coordinates the control of the dust removal equipment and performs energy conservation and emission optimization.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: By introducing intelligent control algorithms and effect monitoring methods, the present invention significantly improves the energy efficiency and emission control capabilities of the iron and steel ore bin dust removal system. Adopting a hierarchical control strategy that combines model predictive control (MPC) and fuzzy PID control can accurately adjust the fan speed, ash cleaning cycle, and damper opening according to real-time data, optimizing the operation of the dust removal system. The multi-objective optimization module adjusts the dust-energy consumption weight ratio according to different electricity price periods to achieve a balance between energy consumption and emissions. Combining real-time data monitoring, energy conservation rate evaluation, and visualization reports improves the intelligent level of the dust removal system and promotes the realization of green transformation and low-carbon emission goals. This method effectively solves the problems of fragmented monitoring, single energy efficiency evaluation, and insufficient intelligent regulation in the prior art, improving the overall efficiency and stability of the dust removal system. Description of the Drawings

[0013] Figure 1 is the system architecture diagram; Figure 2 is the flowchart of the hierarchical control algorithm. Detailed Embodiments

[0014] To further understand the purpose, structure, features, and functions of the present invention, the following is a detailed description in conjunction with embodiments.

[0015] Please refer to Figure 1 , the present invention provides an intelligent control algorithm for iron and steel ore bin dust removal, including: A data acquisition module for real-time acquisition of dust concentration, fan speed, energy consumption, and equipment status signals. The acquisition of real-time data ensures that the entire dust removal system can reflect the real working state and provides accurate input information. This helps the subsequent control module to make accurate decisions and adjustments. It improves the response speed and accuracy of the system. It provides necessary real-time data support for subsequent optimization and control.

[0016] A hierarchical control module, including an upper optimization layer and a lower execution layer, where: The upper optimization layer generates a reference fan speed and ash cleaning cycle through model predictive control (MPC) and performs dynamic optimization through production plans and historical data; The underlying execution layer adopts fuzzy PID control to adjust the damper opening according to the real-time dust concentration and wind pressure difference, so as to achieve precise control of the dust removal fan.

[0017] The upper-layer optimization layer uses model predictive control, which not only realizes the adaptability of the dust removal system in complex and changing environments, but also can achieve long-term energy conservation and efficient control through the combination of historical data and production plans.

[0018] The fuzzy PID control of the underlying execution layer can precisely control the damper opening, eliminate the over-regulation or lag problems common in traditional PID control, and ensure the precise operation of the fan.

[0019] The multi-objective optimization module dynamically adjusts the dust-energy consumption weight ratio (α / β) according to the real-time working conditions, and collaboratively controls the working states of the fan, dust cleaning, and ash conveying equipment.

[0020] By adjusting the weight ratio of dust and energy consumption, the best balance point can be found between dust removal effect and energy consumption. This balance point may be dynamically adjusted with the change of external conditions (such as environmental changes, production load, etc.).

[0021] The multi-objective optimization module effectively optimizes the collaborative work of each device (fan, dust cleaning device, ash conveying equipment, etc.), reduces unnecessary energy consumption, and improves the energy efficiency of the overall system.

[0022] The multi-objective optimization algorithm makes the system have stronger adaptability and flexibility, and can be adjusted according to the needs of different production stages to maintain high-efficiency production and environmental protection levels.

[0023] Preferably, in the hierarchical control module: The model predictive control (MPC) of the upper-layer optimization layer takes the production plan for the next 10 - 30 minutes as the input, with the objective function of minimizing the total energy consumption, and outputs the reference speed of the fan and the dust cleaning cycle.

[0024] The model predictive control (MPC) of the upper-layer optimization layer uses the production plan data for the next 10 - 30 minutes. These data may include production load, equipment status, external environmental changes, etc.

[0025] Objective function: The objective function of this layer is to minimize the total energy consumption, which means that the MPC controller needs to optimize the operation of the system so that the energy consumption reaches the lowest level while maintaining the production requirements.

[0026] Output: The output of MPC is the reference speed of the fan and the dust cleaning cycle. The speed control of the fan directly affects the power consumption of the fan, while the dust cleaning cycle is related to the cleaning efficiency and energy consumption of the equipment.

[0027] By minimizing energy consumption, MPC optimizes the operation of fans and related equipment, reducing energy consumption and operating costs. Based on future production plans, MPC can make decisions in advance, rather than just reacting to the current state, which enables it to optimize the system more effectively. Since MPC can predict and optimize according to changes in production plans, it can adapt to load fluctuations and external environmental changes that may occur during the production process.

[0028] The fuzzy PID controller in the underlying execution layer, whose rule base is generated by training with historical data, controls the dust concentration error and the adjustment amount of the damper opening.

[0029] The underlying execution layer uses a fuzzy PID controller to control based on the dust concentration error and the adjustment amount of the damper opening. The dust concentration error reflects the gap between the current system state and the target, and the adjustment amount of the damper opening is the adjustment amount for controlling the opening degree of the damper, which affects the air flow rate and the dust concentration. The rule base of the fuzzy PID controller is generated by training with historical data, which means that the controller learns and adjusts control rules through historical data and can better adapt to the uncertainties and changes in the actual production process.

[0030] The fuzzy PID controller can make precise adjustments according to the dust concentration error, ensuring that the system meets the cleaning requirements while also optimizing the damper opening, thus achieving fine control.

[0031] The fuzzy PID controller can handle uncertainties and noises. Especially when facing non-linear and time-varying systems, the fuzzy control algorithm shows good robustness. The training of the rule base can help it better adapt to different operating conditions.

[0032] Through the training of historical data, the fuzzy PID controller can continuously adjust the control strategy according to the data changes in actual production, ensuring the stable operation of the system.

[0033] The upper-layer MPC and the lower-layer fuzzy PID controller are used in combination to achieve multi-level control optimization. The upper layer conducts global optimization to determine the globally optimal operation strategy, while the lower layer makes fine adjustments to local operations through real-time control to ensure that the system can maintain the best performance at different times and conditions.

[0034] The upper-layer MPC can reduce energy consumption and optimize the production process by adjusting the fan speed and the dust cleaning cycle. The lower-layer fuzzy PID controller can precisely control the dust concentration, avoiding energy waste caused by excessive or insufficient adjustment of the damper opening.

[0035] Through the forward-looking adjustment of the production plan by MPC, combined with the real-time precise control of the fuzzy PID at the bottom layer, the overall system is ensured to operate smoothly and efficiently.

[0036] Preferably, the multi-objective optimization module dynamically adjusts the dust-energy consumption weight ratio according to different electricity price periods: During peak electricity price periods, by increasing α (prioritizing energy consumption reduction), the dust concentration is allowed to approach the threshold; During off-peak electricity price periods, by increasing β (prioritizing emissions reduction), a safety margin for the dust concentration is reserved to ensure compliance with emission standards.

[0037] A method for monitoring the operation effect of an intelligent control algorithm for steel ore bin dust removal, the method comprising: A data comparison and analysis module for constructing a baseline energy consumption curve, recording the average energy consumption and emission data for 30 days before the algorithm is enabled, and dynamically comparing with real-time data.

[0038] By comparing historical data with real-time data, the improvement effect of the intelligent control algorithm can be effectively evaluated. In this way, operators can timely identify whether the control strategy meets the goals of energy conservation and emission reduction.

[0039] Real-time monitoring can immediately detect abnormal energy consumption and emissions deviating from the baseline, ensuring that the dust removal system always operates at the best state. Anomaly detection can help the system quickly take measures when accidents occur or efficiency decreases, reducing unnecessary energy waste.

[0040] An energy conservation rate and comprehensive performance evaluation module regularly evaluates the optimization effect of the control system by calculating the energy conservation rate η and the comprehensive performance index Q.

[0041] The energy conservation rate η directly measures the effectiveness of the intelligent control system in saving energy and can quantitatively reflect whether the algorithm meets the expected energy conservation goals. The higher the energy conservation rate, the better the optimization effect.

[0042] The comprehensive performance index Q provides a multi-dimensional comprehensive performance evaluation, considering not only energy efficiency but also factors such as the operating stability of the system and the dust removal effect. Through regular evaluation of the Q value, the comprehensive performance of the control system can be comprehensively understood, providing a scientific basis for subsequent optimization.

[0043] A visual monitoring report generation module generates a real-time data dashboard, displays the energy consumption trend, a heat map of the dust concentration distribution, and the change in the Q value, marks abnormal events, and provides optimization suggestions.

[0044] Through the energy consumption trend and the heat map of the dust concentration distribution, operators can intuitively see the current operating state of the system. This real-time feedback mechanism can effectively improve the response speed of the staff and timely adjust the operation strategy.

[0045] If the system malfunctions, such as abnormal energy consumption, excessive dust concentration, etc., this module will immediately mark and alert relevant personnel to facilitate the adoption of corrective measures to ensure that the system always maintains an efficient and safe operating state.

[0046] Based on real-time data analysis, the system can automatically propose optimization suggestions to help operators formulate more precise adjustment strategies. This intelligent feedback mechanism can not only reduce the errors of manual operations but also improve the efficiency of system optimization.

[0047] Preferably, the calculation formula for the comprehensive performance index Q is: 。

[0048] Preferably, the data acquisition module collects data in real time through the following devices: Dust sensor, using laser scattering technology, with a measuring range of 0 - 100 mg / m³ and an accuracy of ±1%; Variable frequency fan, with a power of 55 kW and a speed range of 200 - 3000 rpm; Edge computing device, equipped with an Intel i7 processor, for running algorithm kernels.

[0049] Preferably, the system includes: a hardware configuration module that configures key hardware components such as dust sensors, variable frequency fans, and edge computing devices; An intelligent control module, including a hierarchical control module, a multi-objective optimization module, and an operation effect monitoring module, which coordinates the control of dust removal equipment and performs energy conservation and emission optimization.

[0050] The present invention has been described by the above related embodiments. However, the above embodiments are only examples for implementing the present invention. It must be pointed out that the disclosed embodiments do not limit the scope of the present invention. On the contrary, modifications and refinements made without departing from the spirit and scope of the present invention fall within the scope of patent protection of the present invention.

Claims

1. An intelligent control algorithm for dust removal in iron and steel ore bins, characterized in that: It includes: A data acquisition module for real-time acquisition of dust concentration, fan speed, energy consumption, and equipment status signals; A hierarchical control module, including an upper-layer optimization layer and a lower-layer execution layer, where: The upper-layer optimization layer generates a reference fan speed and a dust cleaning cycle through model predictive control (MPC), and performs dynamic optimization based on production plans and historical data; The lower-layer execution layer adopts fuzzy PID control to adjust the damper opening according to the real-time dust concentration and air pressure difference to achieve precise control of the dust removal fan; A multi-objective optimization module dynamically adjusts the dust-energy consumption weight ratio (α / β) according to the real-time working conditions, and collaboratively controls the working states of the fan, dust cleaning, and ash conveying equipment.

2. The intelligent control algorithm for dust removal in a steel ore bin according to claim 1, wherein: In the hierarchical control module: The model predictive control (MPC) of the upper-layer optimization layer takes the production plan for the next 10 - 30 minutes as the input, with the objective function of minimizing the total energy consumption, and outputs the reference fan speed and the dust cleaning cycle; The fuzzy PID controller of the lower-layer execution layer, whose rule base is generated by training based on historical data, controls the dust concentration error and the damper opening adjustment amount.

3. The intelligent control algorithm for dedusting of iron and steel ore bins as claimed in claim 1, wherein: The multi-objective optimization module dynamically adjusts the dust-energy consumption weight ratio according to different electricity price periods: During peak electricity price periods, by increasing α (prioritizing energy consumption reduction), the dust concentration is allowed to approach the threshold; During off-peak electricity price periods, by increasing β (prioritizing emission reduction), a safety margin for dust concentration is reserved to ensure compliance with emission standards.

4. A monitoring method for the operation effect of the intelligent control algorithm for dedusting in a steel ore bin as described in claim 1, characterized in that, It includes: A data comparison and analysis module for constructing a reference energy consumption curve, recording the average energy consumption and emission data for 30 days before the algorithm is enabled, and dynamically comparing with real-time data; An energy saving rate and performance comprehensive evaluation module for regularly evaluating the optimization effect of the control system by calculating the energy saving rate η and the comprehensive performance index Q; A visualization monitoring report generation module for generating a real-time data dashboard to display the energy consumption trend, the heat map of dust concentration distribution, and the change of Q value, marking abnormal events and providing optimization suggestions.

5. The method for monitoring the operation effect of the intelligent control algorithm for steel ore bin dust removal according to claim 4, characterized in that: The calculation formula for the comprehensive performance index Q is: 。 6. The intelligent control algorithm for dedusting of iron and steel ore bins according to claim 1, wherein: The data acquisition module acquires data in real time through the following devices: A dust sensor using laser scattering technology, with a measurement range of 0 - 100 mg / m³ and an accuracy of ±1%; A variable-frequency fan with a power of 55 kW and a speed range of 200 - 3000 rpm; An edge computing device equipped with an Intel i7 processor for running the algorithm kernel.

7. A system of an intelligent control algorithm for dedusting of a steel ore bin as described in claim 1, characterized in that: The system includes: A hardware configuration module for configuring key hardware components such as dust sensors, variable-frequency fans, and edge computing devices; An intelligent control module, including a hierarchical control module, a multi-objective optimization module, and an operation effect monitoring module, for coordinating the control of dust removal equipment and performing energy saving and emission optimization.

Images