Mpc control method for intelligent dust removal pipe network energy efficiency optimization facing minimum air volume constraint

Abstract

The present application relates to the technical field of industrial dust removal intelligent control, and discloses a wisdom dust removal pipe network energy efficiency optimization MPC control method for minimum air volume constraint, wherein the method collects pipe network operation data, constructs a fan model of pipe network resistance model and coupling relationship of fan frequency and air volume; identifies an active air port set based on multi-source working condition interlocking signal, analyzes the influence law of working condition change on branch flow distribution; takes the minimum air volume of dust deposition prevention as the core constraint, constructs an MPC optimization problem with the minimum fan energy consumption as the target, solves the optimal control amount for execution, combines the online correction model with measured data, and realizes rolling closed-loop optimal control. The present application can guarantee the deposition prevention safety of the dust removal system, greatly reduce the operation energy consumption, improve the working condition self-adaptive ability, and reduce the dependence on manual operation.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for industrial dust removal, and more specifically, to an intelligent dust removal pipeline network energy efficiency optimization MPC control method oriented towards minimum air volume constraints. Background Technology

[0002] In industrial production processes such as metallurgy, building materials, chemicals, and machinery processing, dust pollution control is a core aspect of ensuring production safety and meeting environmental protection requirements. Currently, centralized dust collection systems are widely used in these industries. These centralized dust collection systems typically consist of multiple dust collection points, branch pipe networks, main pipelines, dust collectors, power fans, and supporting actuators. The dust collection points at each production station are connected to the main pipeline through branch pipe networks, with the fans providing suction power to collect the dust-laden airflow to the dust collector for centralized treatment. These systems generally feature a large number of dust collection branches, strong fluid coupling between branches, frequent switching of operating conditions due to the start and stop of production stations, and time-varying characteristics of pipeline operating resistance. The core requirements for their operation and control are: to ensure that the airflow at each dust collection point meets the minimum requirements for preventing dust accumulation, avoiding safety risks such as pipeline blockage and dust explosions, while minimizing fan energy consumption to achieve economical system operation.

[0003] Currently, traditional operation and control methods for centralized dust collection pipeline systems largely rely on manual adjustments based on the on-site experience of operators, or employ a fixed-speed control mode for fans, manually adjusting the opening of valves in each branch to balance the flow distribution within the pipeline. However, given the characteristics of existing centralized dust collection pipeline systems—numerous branches, strong coupling, frequent switching of operating conditions, and time-varying pipeline resistance—existing control methods cannot simultaneously address the minimum airflow safety constraints for preventing dust accumulation in each working dust collection branch while achieving optimal overall system energy efficiency. This easily leads to safety risks from dust accumulation and blockage, as well as significant energy waste, failing to achieve a unified and coordinated approach to safety management and energy-saving optimization in the dust collection system. Therefore, there is an urgent need to provide a smart dust collection pipeline energy efficiency optimization MPC control method oriented towards minimum airflow constraints to solve these problems. Summary of the Invention

[0004] To address the aforementioned technical issues, this invention provides an intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum airflow constraints. By relying on multi-source signals to accurately identify active air outlets, the minimum airflow constraint is embedded into the control system, completely resolving the dust deposition and blockage problem caused by operating condition fluctuations and strengthening the safety baseline. Through MPC, the fan and valve are optimized in synergy, abandoning the extensive operation mode and significantly reducing energy consumption.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: This invention provides an intelligent dust collection pipeline network energy efficiency optimization MPC control method oriented towards minimum airflow constraints. It is applied to a dust collection pipeline network system including a fan, main pipeline, multiple dust collection branches, regulating valves and dust hoods for each branch, a pipeline network sensing unit, and a controller. The method includes the following steps: S1. Collect real-time operating status data of the dust removal pipeline network through the pipeline network sensing unit to obtain the structural topology and pipeline network component parameters of the dust removal pipeline network; establish a full-system pipeline network resistance model based on the structural topology and pipeline network component parameters, and establish a fan model including the coupling relationship between fan frequency and air volume based on the fan performance curve. S2. Based on the collected multi-source operating condition interlocking signals, determine the active air outlet set, set zero air volume or valve closure constraints for inactive air outlets, and perform joint scheduling of safety and energy efficiency only within the active air outlet set; construct a hierarchical resistance topology of the dust removal pipeline network based on the active air outlet set, obtain the current total resistance of the system through equivalent synthesis calculation of series and parallel pipeline network resistance, predict the total air volume of the system based on the total system resistance and the fan model, calculate the initial flow prediction value of each active air outlet according to the damping inverse proportional distribution principle, and analyze the influence law of operating condition changes on branch flow distribution based on the initial flow prediction value; S3. For each branch within the set of active air outlets, calculate the minimum air volume threshold for each branch based on the minimum conveying wind speed for preventing dust deposition and the corresponding branch pipe cross-sectional area; under a preset sampling period, construct an MPC control quantity that includes the fan operating frequency and the adjustment parameters of the regulating valves of each active branch; within a preset prediction step size, construct an energy efficiency optimal objective function with the minimum fan operating energy consumption as the core optimization objective, and set safety and actuator constraints that include at least the minimum air volume constraint, and solve to obtain the optimal combination of control quantities that satisfies all constraints; S4. The optimal control quantity combination obtained from the solution is sent to the fan frequency converter and the actuators of each branch regulating valve for execution; real-time collection of pipeline operation feedback data is performed, and the feedback data is compared and verified with the model prediction values. When the pipeline operating conditions change and the model parameters deviate, the parameters of the pipeline resistance model are corrected online based on the measured operating data; steps S3 and S4 are repeated continuously to achieve dynamic energy efficiency optimal closed-loop control under the safety and anti-deposition constraints of the dust removal pipeline network.

[0006] As a preferred embodiment of the present invention, step S1 specifically includes: S11. By using sensing units arranged throughout the dust removal pipeline network, real-time operating status data of the dust removal pipeline network is collected, and at the same time, the structural topology information of the dust removal pipeline network and the basic parameters of each component in the pipeline network are obtained. S12. Based on the obtained pipeline network topology information and basic parameters of each component, establish a full-system pipeline network resistance model that can predict pipeline network pressure and flow conditions. S13. Based on the performance curve of the fan supporting the dust removal pipe network, establish a fan model that includes the coupling relationship between the fan operating frequency and the air volume. The pipe network resistance model and the fan model together provide a calculation basis for subsequent optimization control to predict the operating conditions.

[0007] As a preferred embodiment of the present invention, the specific steps of S2 include: S21. Based on the collected multi-source operating condition interlock signals, determine the dust removal air inlets that need to be maintained for suction currently, delineate the active air inlet set, set zero air volume or valve closing constraints for the non-active air inlets, and only perform subsequent joint safety and energy efficiency scheduling within the active air inlet set; S22. Based on the pipe network structure corresponding to the active air inlet set, construct a hierarchical resistance topology relationship of the dust removal pipe network, and clarify the series-parallel logic of each branch; S23. Based on the principles of pressure balance and mass conservation in fluid mechanics, perform resistance equivalent simplification on the pipe network corresponding to the active air inlets, and calculate the current total resistance of the pipe network system; S24. Based on the total system resistance and the total system air volume predicted by the fan model, calculate the initial flow prediction values of each active air inlet, and use them as the initial state quantities for subsequent rolling optimization; S25. Based on the initial flow prediction values, analyze the influence law of different operating condition changes on the flow distribution of each branch, and identify the risk operating conditions where the branch air volume drops below the safety threshold.

[0008] As a preferred embodiment of the present invention, the specific steps of S3 include: S31. For each dust removal branch within the active air inlet set, determine the minimum air volume threshold corresponding to each branch according to the safety requirements for dust anti-deposition and the branch pipe parameters, as the safety protection red line of the dust removal system; S32. Based on the preset control sampling period, construct a MPC control quantity that includes the fan operating frequency and the adjustment parameters of the regulating valves of each active branch; S33. Within the preset prediction step range, construct an energy efficiency optimal objective function with the minimum fan operating energy consumption as the core optimization goal, while taking into account the system operation stability and constraint satisfaction degree; S34. Set multiple groups of constraint conditions for the optimization problem, including minimum air volume constraints, physical boundary constraints of the actuators, and pipe network safety pressure constraints; S35. Integrate the objective function and the constraint conditions to construct an optimization problem with constraints, and after solving, obtain the optimal control quantity combination that can be issued for execution at the current moment.

[0009] As a preferred embodiment of the present invention, the specific steps of S4 include: S41. The optimal control quantity combination obtained from the solution is sent to the wind turbine frequency converter and the actuator of each branch regulating valve to complete the corresponding regulation action; S42. Collect pipeline operation feedback data after adjustment actions in real time, compare it with the model prediction value, and complete the three verifications of safety protection effect, energy efficiency optimization effect and model prediction accuracy. S43. When changes in pipeline operating conditions cause model parameters to shift or prediction accuracy to exceed a preset threshold, the parameters of the pipeline resistance model are corrected online based on measured operating data to suppress prediction errors. S44. According to the preset sampling period, the entire process of problem solving, control execution, and model correction is repeatedly optimized to achieve dynamic closed-loop optimal control of the dust removal pipeline network under all operating conditions.

[0010] As a preferred embodiment of the present invention, the multi-source operating condition interlocking signal includes an air outlet switch signal, a valve position feedback signal, a dust collection hood process signal, and a dust removal point interlocking signal; the air outlet switch signal is a switch status signal indicating whether each dust removal branch is open; the valve position feedback signal is a feedback signal indicating the actual opening degree of the regulating valve of each branch; the dust collection hood process signal is an operating status signal of the dust-generating equipment or process station corresponding to the dust collection hood, used to indicate whether there is a need for ventilation at the corresponding point; and the dust removal point interlocking signal is a dust removal branch interlocking activation signal generated based on equipment start / stop, process conditions, or safety logic.

[0011] As a preferred embodiment of the present invention, identifying risky operating conditions in S25 includes: Identify risky operating conditions when air outlets switch: When some air outlets switch from an inactive state to an active state, the range of active air outlets expands and the total air volume demand of the system increases. If the pressure difference of the pipeline is maintained only by throttling the branch valves, the air volume of the original active branches is easily squeezed out, resulting in the risk of insufficient air volume. Identify time-varying resistance risk conditions: When the branch resistance coefficient increases with dust accumulation, filter bag pressure difference, and pipeline aging, the air volume of the corresponding branch will decrease under the same fan frequency operating conditions, and there is a risk that the air volume will fall below the minimum air volume threshold. Identify risky operating conditions for energy-saving frequency regulation: When the fan frequency is reduced to lower the operating energy consumption, the total air volume of the system decreases. The strongly coupled pipe network will cause a nonlinear redistribution of the air volume of each branch. The terminal branches with greater resistance will be the first to fall below the minimum air volume threshold. The minimum air volume constraint needs to be explicitly incorporated into the optimization control system.

[0012] As a preferred embodiment of the present invention, the minimum air volume core safety constraint for preventing dust deposition requires that the operating air volume of all active dust removal branches meet the minimum air volume threshold requirement of the corresponding branch, serving as an insurmountable safety boundary for optimized control. The physical boundary constraint on the operating frequency of the wind turbine limits the operating frequency of the wind turbine to be within the allowable operating frequency range of the equipment, so as to ensure the safe and stable operation of the wind turbine equipment. The boundary constraint of the adjustment range of the branch regulating valve limits the adjustment parameters of the regulating valves of each active branch to be within the physically executable adjustment range of the equipment, so as to avoid the valves from operating beyond the range. The safety constraint on the operating pressure of the pipeline network limits the operating pressure of the dust removal pipeline network to be within the pressure range of the pipeline network design pressure resistance level and the dust removal process requirements, so as to ensure the overall operational safety of the pipeline network system.

[0013] This invention also provides an intelligent dust removal pipeline network energy efficiency optimization MPC control system oriented towards minimum airflow constraints, applied to a dust removal pipeline network system. The dust removal pipeline network system includes a fan, main pipeline, multiple dust removal branches, regulating valves and dust collection hoods for each branch, and a pipeline network sensing unit, comprising: The data acquisition and model building module is used to collect real-time operating status data of the dust removal pipeline network through the pipeline network sensing unit, and obtain the structural topology and pipeline network component parameters of the dust removal pipeline network; it is also used to establish a full-system pipeline network resistance model based on the structural topology and pipeline network component parameters, and to establish a fan model that includes the coupling relationship between fan frequency and air volume based on the fan performance curve; The active air outlet identification and operating condition analysis module is used to determine the set of active air outlets based on the collected multi-source operating condition interlocking signals, set zero air volume or valve closure constraints for inactive air outlets, and perform joint scheduling of safety and energy efficiency only within the set of active air outlets; it is also used to construct a hierarchical resistance topology of the dust removal pipeline network based on the set of active air outlets, obtain the current total resistance of the system through equivalent synthesis calculation of series and parallel pipeline network resistance, predict the total air volume of the system based on the total system resistance and the fan model, calculate the initial flow prediction value of each active air outlet according to the damping inverse proportional distribution principle, and analyze the impact of operating condition changes on branch flow distribution based on the initial flow prediction value; The MPC optimization solution module is used to calculate the minimum air volume threshold for each branch within the set of active air outlets, based on the minimum delivery velocity for preventing dust deposition and the corresponding branch cross-sectional area. It is also used to construct MPC control quantities including the fan operating frequency and the adjustment parameters of the regulating valves of each active branch within a preset sampling period. Furthermore, it is used to construct an energy efficiency optimal objective function with the minimum fan operating energy consumption as the core optimization objective within a preset prediction step size, while setting safety and actuator constraints that include at least the minimum air volume constraint, and solving for the optimal combination of control quantities that satisfies all constraints. The closed-loop execution and rolling optimization module is used to send the optimal control quantity combination obtained from the solution to the fan frequency converter and the actuators of each branch regulating valve for execution; it is also used to collect pipeline operation feedback data in real time, compare and verify the feedback data with the model prediction values, and when the pipeline operating conditions change and the model parameters deviate, it corrects the parameters of the pipeline resistance model online based on the measured operating data; it is also used to continuously call the corresponding functions of the MPC optimization solution module and the closed-loop execution and rolling optimization module to realize the dynamic energy efficiency optimal closed-loop control under the safety and anti-deposition constraints of the dust removal pipeline network.

[0014] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the above-described intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum airflow constraints.

[0015] The beneficial technical effects of this invention are: This invention accurately identifies active air outlet sets through multi-source interlocking signals, and performs safety control only on branches with ventilation needs. At the same time, it explicitly incorporates the minimum air volume constraint for preventing dust accumulation in each branch into the control system, and predicts the risk of insufficient air volume caused by fluctuations in operating conditions by combining the flow distribution law of the pipeline network. This ensures that the air volume of all working branches always meets the safety threshold requirements, completely solving the problems of insufficient air volume and dust accumulation blockage that are prone to occur when operating conditions fluctuate in traditional control methods, and ensuring the long-term safe and stable operation of the dust removal system.

[0016] This invention takes minimizing the energy consumption of the fan as its core optimization objective. It achieves coordinated optimization of the frequency conversion regulation of the fan and the opening of the valves in each branch through model predictive control. It abandons the traditional crude mode of excessive fan operation and valve throttling to offset the problem. Under the premise of strictly meeting safety constraints, it minimizes the operating power of the fan and realizes the overall energy efficiency optimization of the dust removal pipeline system, which significantly improves the economic efficiency of the system operation.

[0017] This invention achieves accurate prediction of pipeline network operating conditions by constructing a pipeline network resistance model and a fan model. At the same time, it corrects the model parameters online based on real-time operation feedback data. It can adaptively match the time-varying characteristics of pipeline network resistance caused by factors such as dust deposition, filter bag pressure difference changes, and pipeline aging. It solves the defect of traditional fixed-value control that cannot adapt to dynamic operating conditions, and ensures that rolling optimization is always based on the latest actual operating state of the pipeline network, which greatly improves the system's adaptability to operating conditions and control accuracy.

[0018] By automatically identifying active air vents, analyzing the changing patterns of operating conditions, solving for the optimal control scheme online, and executing adjustment commands in a closed loop, the entire process of automated management and control is achieved. It eliminates the need for manual adjustments based on the experience of operators, significantly reducing the intensity of manual operations. At the same time, it can quickly respond to frequent changes in production conditions, realizing dynamic optimal control under changing operating conditions, and adapting to the multi-station and changing process characteristics of industrial production. Attached Figure Description

[0019] Figure 1 This is a technical roadmap of the present invention. Detailed Implementation

[0020] In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0021] Combination Figure 1 The present invention provides the following embodiments: Example 1: A smart dust collection pipeline network energy efficiency optimization MPC control method for minimum airflow constraints, applied to a dust collection pipeline network system including a fan, main pipeline, multiple dust collection branches, regulating valves and dust hoods for each branch, pipeline network sensing unit and controller, including the following steps: S1. Collect real-time operating status data of the dust removal pipeline network through the pipeline network sensing unit to obtain the structural topology and pipeline network component parameters of the dust removal pipeline network; establish a full-system pipeline network resistance model based on the structural topology and pipeline network component parameters, and establish a fan model including the coupling relationship between fan frequency and air volume based on the fan performance curve. S2. Based on the collected multi-source operating condition interlocking signals, determine the active air outlet set, set zero air volume or valve closure constraints for inactive air outlets, and perform joint scheduling of safety and energy efficiency only within the active air outlet set; construct a hierarchical resistance topology of the dust removal pipeline network based on the active air outlet set, obtain the current total resistance of the system through equivalent synthesis calculation of series and parallel pipeline network resistance, predict the total air volume of the system based on the total system resistance and the fan model, calculate the initial flow prediction value of each active air outlet according to the damping inverse proportional distribution principle, and analyze the influence law of operating condition changes on branch flow distribution based on the initial flow prediction value; S3. For each branch within the set of active air outlets, calculate the minimum air volume threshold for each branch based on the minimum conveying wind speed for preventing dust deposition and the corresponding branch pipe cross-sectional area; under a preset sampling period, construct an MPC control quantity that includes the fan operating frequency and the adjustment parameters of the regulating valves of each active branch; within a preset prediction step size, construct an energy efficiency optimal objective function with the minimum fan operating energy consumption as the core optimization objective, and set safety and actuator constraints that include at least the minimum air volume constraint, and solve to obtain the optimal combination of control quantities that satisfies all constraints; S4. Transmit the obtained optimal control quantity combination to the fan frequency converter and the actuator of each branch regulating valve for execution; collect the operation feedback data of the pipe network in real time, compare and verify the feedback data with the model prediction value. When the change of the pipe network working condition causes the deviation of the model parameters, online correct the parameters of the pipe network resistance model based on the measured operation data; repeat steps S3 and S4 in a rolling manner to achieve the dynamic energy efficiency optimal closed-loop control under the safety and anti-deposition constraints of the dust removal pipe network.

[0022] Further, the specific steps of S1 include: S11. Through the sensing units arranged at various places of the dust removal pipe network, collect the real-time operation state data of the dust removal pipe network, and at the same time obtain the structural topology information of the dust removal pipe network and the basic parameters of each component in the pipe network; S12. Based on the obtained pipe network structural topology information and the basic parameters of each component, establish a full-system pipe network resistance model that can predict the pipe network pressure and flow conditions; S13. Based on the performance curve of the fan supporting the dust removal pipe network, establish a fan model that includes the coupling relationship between the fan operation frequency and the air volume. The pipe network resistance model and the fan model jointly provide the calculation basis for the subsequent optimization control to predict the working conditions.

[0023] By collecting the pipe network operation data and basic parameters in all dimensions, a real and comprehensive data source is provided for model construction. The pipe network resistance model established based on the fluid mechanics law can accurately map the coupling relationship between the pipe network pressure and the flow rate. The fan model constructed in combination with the fan performance curve can match the frequency conversion regulation characteristics. The two work together to form the prediction ability of the full working conditions of the pipe network, transforming the subsequent optimization control from empirical judgment to quantitative calculation based on a mathematical model, laying a data and model foundation for the coordinated optimization of safety and energy efficiency.

[0024] Further, the specific steps of S2 include: S21. Based on the collected multi-source working condition interlock signals, determine the dust removal air inlets that need to be maintained for suction currently, delimit the active air inlet set, set zero air volume or valve closing constraints for the non-active air inlets, and only perform the subsequent safety and energy efficiency joint scheduling within the active air inlet set; S22. Based on the pipe network structure corresponding to the active air inlet set, construct the hierarchical resistance topology relationship of the dust removal pipe network, and clarify the series-parallel logic of each branch; S23. Based on the pressure balance and mass conservation principles of fluid mechanics, perform resistance equivalent simplification on the pipe network corresponding to the active air inlets, and calculate the current total resistance of the pipe network system; S24. Based on the total resistance of the system and the total air volume of the system predicted by the fan model, calculate the initial flow prediction value of each active air inlet, and use it as the initial state quantity for subsequent rolling optimization; S25. Based on the initial flow forecast, analyze the impact of different operating conditions on the flow distribution of each branch, and identify the risk conditions in which the branch air volume falls below the safety threshold.

[0025] By cross-validating multiple signals to identify the active air outlet set, ineffective scheduling of branches without ventilation needs can be avoided, significantly reducing the amount of optimization calculations. Abstracting the pipeline network into a hierarchical resistance topology and simplifying it through series and parallel equivalents can transform complex multi-branch pipeline networks into quantitatively calculable equivalent models that accurately reflect the actual resistance characteristics of the pipeline network. Calculating the initial flow rate according to the inverse proportion of damping can quickly predict the flow rate distribution trend under changing operating conditions, identify risky operating conditions in advance, and make subsequent constraint settings more targeted, thus avoiding safety hazards caused by insufficient air volume from the source.

[0026] Furthermore, the specific steps of S3 include: S31. For each dust removal branch within the active air outlet cluster, determine the minimum air volume threshold corresponding to each branch based on the safety requirements for dust anti-deposition and the branch pipeline parameters, as the safety protection red line of the dust removal system. S32. Based on the preset control sampling period, construct the MPC control quantity, which includes the fan operating frequency and the adjustment parameters of each active branch regulating valve; S33. Within the preset prediction step size, construct an energy efficiency optimal objective function with the minimum energy consumption of wind turbine operation as the core optimization objective, while also taking into account the system operation stability and constraint satisfaction. S34. To optimize the problem, set up multiple sets of constraints, including minimum air volume constraints, actuator physical boundary constraints, and pipeline safety pressure constraints. S35. Integrate the objective function and constraints to construct a constrained optimization problem. After solving the problem, obtain the optimal combination of control variables that can be issued and executed at the current moment.

[0027] The minimum airflow threshold determined based on dust anti-deposition requirements sets an inviolable safety baseline for the dust removal system. Integrating fan frequency and valve adjustment parameters into MPC control variables enables coordinated adjustment of the fan and branch actuators, avoiding flow imbalance caused by single adjustment. Constructing a multi-objective energy-efficient optimal objective function can minimize energy consumption while avoiding system fluctuations caused by large actuator movements. By setting multi-dimensional constraints, the physical limits of equipment, pipeline operation safety, and dust anti-deposition requirements can all be incorporated into the optimization system, ensuring that the solved control variable combination meets safety requirements, achieves optimal energy efficiency, and is adapted to the actual operating capacity of the equipment.

[0028] Furthermore, the specific steps of S4 include: S41. The optimal control quantity combination obtained from the solution is sent to the wind turbine frequency converter and the actuator of each branch regulating valve to complete the corresponding regulation action; S42. Collect pipeline operation feedback data after adjustment actions in real time, compare it with the model prediction value, and complete the three verifications of safety protection effect, energy efficiency optimization effect and model prediction accuracy. S43. When changes in pipeline operating conditions cause model parameters to shift or prediction accuracy to exceed a preset threshold, the parameters of the pipeline resistance model are corrected online based on measured operating data to suppress prediction errors. S44. According to the preset sampling period, the entire process of problem solving, control execution, and model correction is repeatedly optimized to achieve dynamic closed-loop optimal control of the dust removal pipeline network under all operating conditions.

[0029] The optimal control quantity is directly sent to the actuator, enabling rapid implementation of the optimization plan and ensuring accurate execution of control commands. By comparing and verifying measured data with predicted values, the achievement of system safety and energy efficiency targets can be monitored in real time, while the prediction accuracy of the model can be judged. Online correction of model parameters based on measured data allows the model to dynamically adapt to the time-varying characteristics of pipeline resistance caused by dust deposition, filter bag pressure difference changes, pipeline aging, etc., effectively suppressing prediction errors. Through rolling and repeated optimization, execution, and correction processes, a dynamic closed-loop control is formed, allowing the system to continuously adapt to real-time changes in operating conditions, ensuring that the optimal balance between safety and energy efficiency is always maintained under all operating conditions.

[0030] Furthermore, the multi-source interlocking signals include vent switch signals, valve position feedback signals, dust hood process signals, and dust removal point interlocking signals. The vent switch signals are switch status signals indicating whether each dust removal branch is open. The valve position feedback signals are feedback signals indicating the actual opening degree of the regulating valves in each branch. The dust hood process signals are operating status signals of the dust-generating equipment or process station corresponding to the dust hood, used to indicate whether there is a need for ventilation at the corresponding point. The dust removal point interlocking signals are dust removal branch interlock activation signals generated based on equipment start / stop, process conditions, or safety logic.

[0031] By employing multi-source interlocking signals for active air outlet identification, the limitations of single-signal identification are overcome. Through cross-verification of multi-dimensional signals such as air outlet opening / closing, valve position status, process equipment operation, and safety interlocking logic, the actual ventilation demand of dust removal points can be accurately and in real time. This effectively avoids deviations in active air outlet identification caused by single signal failure or misjudgment, ensuring a high degree of matching between the scheduling range and actual production needs. At the same time, it makes the constraint control of inactive air outlets more reasonable, laying the foundation for subsequent precise scheduling.

[0032] Furthermore, the risk identification process in S25 includes: Identify risky operating conditions when air outlets switch: When some air outlets switch from an inactive state to an active state, the range of active air outlets expands and the total air volume demand of the system increases. If the pressure difference of the pipeline is maintained only by throttling the branch valves, the air volume of the original active branches is easily squeezed out, resulting in the risk of insufficient air volume. Identify time-varying resistance risk conditions: When the branch resistance coefficient increases with dust accumulation, filter bag pressure difference, and pipeline aging, the air volume of the corresponding branch will decrease under the same fan frequency operating conditions, and there is a risk that the air volume will fall below the minimum air volume threshold. Identify risky operating conditions for energy-saving frequency regulation: When the fan frequency is reduced to lower the operating energy consumption, the total air volume of the system decreases. The strongly coupled pipe network will cause a nonlinear redistribution of the air volume of each branch. The terminal branches with greater resistance will be the first to fall below the minimum air volume threshold. The minimum air volume constraint needs to be explicitly incorporated into the optimization control system.

[0033] Risk identification was performed for three typical scenarios of changes in the operating conditions of dust removal pipeline networks. The core rules of flow distribution in strongly coupled multi-branch pipeline networks were accurately captured. The triggering causes and key influencing points of insufficient air volume risk under different operating conditions were clarified. This ensures that subsequent optimization constraint settings are no longer generalized safety requirements, but precise prevention and control for specific risk scenarios. By explicitly incorporating the minimum air volume constraints corresponding to these risk points into the optimization system, risks can be avoided in advance from the optimization design level, ensuring that the air volume of each branch is always above the safe threshold when the operating conditions change.

[0034] Furthermore, the minimum airflow core safety constraint for preventing dust deposition requires that the operating airflow of all active dust removal branches meet the minimum airflow threshold requirement of the corresponding branch, serving as an insurmountable safety boundary for optimized control. The physical boundary constraint on the operating frequency of the wind turbine limits the operating frequency of the wind turbine to be within the allowable operating frequency range of the equipment, so as to ensure the safe and stable operation of the wind turbine equipment. The boundary constraint of the adjustment range of the branch regulating valve limits the adjustment parameters of the regulating valves of each active branch to be within the physically executable adjustment range of the equipment, so as to avoid the valves from operating beyond the range. The safety constraint on the operating pressure of the pipeline network limits the operating pressure of the dust removal pipeline network to be within the pressure range of the pipeline network design pressure resistance level and the dust removal process requirements, so as to ensure the overall operational safety of the pipeline network system.

[0035] Setting the minimum airflow constraint as the core safety boundary tightly addresses the core safety requirements of dust removal systems, such as preventing dust accumulation and blockage, thereby fundamentally avoiding safety accidents. Setting physical boundary constraints for fan frequency and valve regulation prevents actuators from operating beyond their range, avoiding mechanical failures or shortened service life due to equipment overload or over-range operation. Setting pipeline pressure safety constraints matches the design pressure resistance characteristics of the pipeline network with the pressure requirements of the dust removal process, preventing problems such as leakage, damage, or decreased dust removal efficiency due to overpressure or underpressure. These multi-dimensional constraints work together to form a complete constraint system covering the safety of dust removal system equipment, pipeline network, and process, ensuring that optimized control is always carried out within a safe framework.

[0036] Example 2: This invention also provides an intelligent dust removal pipeline network energy efficiency optimization MPC control system oriented towards minimum airflow constraints, applied to a dust removal pipeline network system. The dust removal pipeline network system includes a fan, main pipeline, multiple dust removal branches, regulating valves and dust collection hoods for each branch, and a pipeline network sensing unit, comprising: The data acquisition and model building module is used to collect real-time operating status data of the dust removal pipeline network through the pipeline network sensing unit, and obtain the structural topology and pipeline network component parameters of the dust removal pipeline network; it is also used to establish a full-system pipeline network resistance model based on the structural topology and pipeline network component parameters, and to establish a fan model that includes the coupling relationship between fan frequency and air volume based on the fan performance curve; The active air outlet identification and operating condition analysis module is used to determine the set of active air outlets based on the collected multi-source operating condition interlocking signals, set zero air volume or valve closure constraints for inactive air outlets, and perform joint scheduling of safety and energy efficiency only within the set of active air outlets; it is also used to construct a hierarchical resistance topology of the dust removal pipeline network based on the set of active air outlets, obtain the current total resistance of the system through equivalent synthesis calculation of series and parallel pipeline network resistance, predict the total air volume of the system based on the total system resistance and the fan model, calculate the initial flow prediction value of each active air outlet according to the damping inverse proportional distribution principle, and analyze the impact of operating condition changes on branch flow distribution based on the initial flow prediction value; The MPC optimization solution module is used to calculate the minimum air volume threshold for each branch within the set of active air outlets, based on the minimum delivery velocity for preventing dust deposition and the corresponding branch cross-sectional area. It is also used to construct MPC control quantities including the fan operating frequency and the adjustment parameters of the regulating valves of each active branch within a preset sampling period. Furthermore, it is used to construct an energy efficiency optimal objective function with the minimum fan operating energy consumption as the core optimization objective within a preset prediction step size, while setting safety and actuator constraints that include at least the minimum air volume constraint, and solving for the optimal combination of control quantities that satisfies all constraints. The closed-loop execution and rolling optimization module is used to send the optimal control quantity combination obtained from the solution to the fan frequency converter and the actuators of each branch regulating valve for execution; it is also used to collect pipeline operation feedback data in real time, compare and verify the feedback data with the model prediction values, and when the pipeline operating conditions change and the model parameters deviate, it corrects the parameters of the pipeline resistance model online based on the measured operating data; it is also used to continuously call the corresponding functions of the MPC optimization solution module and the closed-loop execution and rolling optimization module to realize the dynamic energy efficiency optimal closed-loop control under the safety and anti-deposition constraints of the dust removal pipeline network.

[0037] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the above-described intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum airflow constraints.

[0038] The following is an application example of the present invention: This paper presents an intelligent dust collection network energy efficiency optimization MPC control method for minimum airflow constraints. It is applicable to multi-branch, strongly coupled industrial dust collection network systems. The system consists of: a centrifugal fan and its associated frequency converter, main duct, N dust collection branches, regulating valves / dust hoods for each branch, pressure / flow / valve position sensing units, a PLC / DCS control system, and an MPC controller. The control method includes the following four core steps.

[0039] Step 1: Data Acquisition and System Prediction Model Construction The core of this step is to complete the full-dimensional data collection of the pipeline network and build a pipeline resistance model and a fan model that can predict pressure and flow conditions, so as to provide a computational basis for subsequent optimization.

[0040] Step 1.1: Collection of Pipeline Operation Data and Basic Parameters Sensor unit layout and real-time data acquisition Pressure sensors and flow sensors are installed at the fan outlet, the reference point of the main pipeline, and the inlet and outlet of each dust removal branch. Valve position acquisition modules are installed at the regulating valves of each branch. All data are collected at a preset sampling period. Real-time uploads are made to the MPC controller. The data acquisition items and parameter definitions are as follows: Pressure difference between fan outlet and pipeline reference point Total air volume of dust removal system , No. Branch line real-time air volume , No. Branch road pressure loss , No. Branch execution mechanism status Real-time operating frequency of the fan

[0041] Pipeline network basic parameter acquisition Collect the pipeline network topology diagram, pipe diameter / length of each pipe section, resistance coefficient of components such as dust hoods / valves / pipes / tees, and branch pipe cross-sectional area. The rated parameters and factory performance curves of the wind turbine provide basic calibration parameters for model construction.

[0042] Step 1.2: Construction of System Prediction Model Pipeline resistance model

[0043] Based on the square-square resistance characteristics of pipeline networks in fluid mechanics, resistance prediction models are established for the entire system and individual branches, respectively, clarifying the coupling relationship between pressure loss and flow rate: Overall system piping characteristic curves: ; In the formula: The overall resistance coefficient of the pipeline system is obtained by combining the resistance coefficients of all components of the pipeline system in series and parallel.

[0044] Single-branch resistance model: ; In the formula: For the first The overall resistance coefficient of a branch is the sum of the resistance coefficients of all components in that branch, including pipe sections, valves, dust hoods, tees, etc.

[0045] Wind turbine model Based on the wind turbine performance curve, the variation with operating frequency was obtained through fitting. A variable fan head-flow coupling model enables accurate pressure-flow prediction under variable frequency control. The model is as follows: In the formula: The total pressure output by the fan, and the total pressure loss of the pipeline network under steady state. balance; , , The fitting coefficients for the variation of the fan frequency are obtained by calibration using the fan's factory performance curve; For the operating frequency of the wind turbine, This represents the total air volume of the system.

[0046] Through the aforementioned dual-model approach, the control system can achieve different operating conditions. , , The accurate predictions provide the core computational foundation for subsequent MPC rolling optimization.

[0047] Step 2: Identification of Active Hot Spot Clusters and Analysis of Traffic Distribution Patterns The core of this step is to identify the target air outlet that needs to be scheduled, establish the pipeline resistance topology, calculate the flow distribution trend under changing operating conditions, and identify the safety risks of insufficient air volume in advance.

[0048] Step 2.1: Identification and Constraint Setting of Active Wind Gap Sets By using multi-source operating condition interlock signals, the dust collection vents that need to maintain suction are identified, and the set of active vents is determined. : ; The multi-source interlocking signals include: Air outlet switch signal: a switch status signal indicating whether each dust removal branch is open; Valve position feedback signal: a feedback signal that characterizes the actual opening degree of the control valve in each branch; Dust hood process signal: The operating status signal of the dust-generating equipment / process station corresponding to the dust hood, indicating whether there is a need for ventilation at the corresponding point; Dust removal point interlock signal: Dust removal branch interlock activation signal generated based on equipment start / stop, process conditions or safety logic.

[0049] For inactive hot spots, i.e. ,set up Zero airflow constraint or valve-closed constraint, the control system only aggregates at active air outlets. Internally, safety and energy efficiency are jointly managed.

[0050] Step 2.2: Construction of Hierarchical Resistance Topology of Pipeline Network The dust collection network is abstracted as a directed acyclic graph. , where the node set For tees and airflow convergence points in the pipeline network, with the dust collector inlet as the root node; edge collection For each branch pipe section, the corresponding airflow path of the pipe network.

[0051] Starting from the root node, the complete airflow path from each active branch to the main pipe is identified by traversing the reverse airflow direction, establishing a hierarchical resistance topology relationship, and clarifying the series and parallel logic of each branch.

[0052] Step 2.3, Algorithm for Equivalent Synthesis of Resistance in Series and Parallel Pipeline Networks Based on the principles of pressure balance and the law of conservation of mass in fluid mechanics, the pipeline network corresponding to the active air outlet is simplified to an equivalent form, and the current total resistance of the system is calculated using the following algorithm: Series resistance synthesis For continuous pipe sections, valves, dust hoods, and local components along the same airflow path, the equivalent resistance coefficient is the sum of the resistance coefficients of all components along the path: In the formula: This represents the equivalent drag coefficient of the series path; For the first on the path The drag coefficient of each component; This represents the total number of components in the series path.

[0053] Parallel resistance synthesis For multiple active branches converging at the same node, based on the principle that the voltage drop of each branch is equal ( Its parallel equivalent drag coefficient satisfies: In the formula: The equivalent resistance coefficient of the parallel branch group; For the first The overall resistance coefficient of an active branch; This refers to the set of active hotspots converging at this node.

[0054] By synthesizing the series and parallel resistances in a hierarchical manner, the total comprehensive resistance coefficient of the dust removal pipeline system is finally calculated. .

[0055] Step 2.4, Initial Flow Allocation Calculation At the moment a new operating condition is triggered, such as fan frequency adjustment, vent switch switching, or valve position adjustment, the synthesized total system resistance coefficient is used. Total system air volume predicted by the fan model Based on the principle of inverse damping distribution, the initial flow prediction values ​​for each active vent are calculated. : In the formula: For the first Initial traffic forecast values ​​for each active trend; For the first The overall resistance coefficient of an active branch; For the first The overall resistance coefficient of an active branch; This is a collection of active trending topics.

[0056] This initial flow prediction value serves as the initial state variable for MPC rolling optimization, used to determine which branches are at risk of their airflow falling below the minimum airflow threshold at the initial moment.

[0057] Step 2.5: Analysis of the Influence of Operating Condition Changes on Flow Distribution Based on the above topology, resistance synthesis, and flow distribution calculations, three core risk scenarios are identified, providing a basis for subsequent constraint settings: Air vent switching condition: When some air vents change from inactive to active, the collection... As the system expands, the total air volume demand increases. If the pressure difference in the pipeline network is maintained only by throttling the branch valves, the air volume of the original active branches will be squeezed out, resulting in the risk of insufficient air volume. Time-varying resistance conditions: when the branch resistance coefficient As dust accumulates, filter bag pressure differential increases, and pipeline aging occurs, the air volume of the corresponding branch increases under the same fan frequency operating conditions. It may decrease, and there is a risk that it will fall below the minimum airflow threshold; Energy-saving frequency regulation mode: when the fan frequency When adjusting to reduce energy consumption, the total system air volume As airflow decreases, a strongly coupled pipe network will cause a nonlinear redistribution of airflow in each branch. The terminal branches with high resistance will be the first to fall below the minimum airflow threshold, so the minimum airflow constraint must be explicitly incorporated into the optimization.

[0058] Step 3: Problem Construction and Solution of MPC Optimization with Minimum Airflow Constraint This step is the core of the method. It takes minimizing the energy consumption of the fan as the optimization objective and the minimum air volume for preventing deposition as the core constraint. It constructs and solves the MPC optimization problem to obtain the optimal adjustment parameters of the fan and valve.

[0059] Step 3.1: Calculation of minimum airflow threshold Targeting active trend collection For each branch within the pipe, the minimum airflow threshold is calculated based on the minimum transport velocity for dust prevention and the cross-sectional area of ​​the branch pipe. The calculation formula is as follows: In the formula: For the first Minimum airflow threshold for each active branch; The minimum conveying velocity for preventing dust deposition in the dust removal medium is determined based on the physical properties of dust particles, such as particle size and density. The conventional value for industrial dust removal is 8-12 m / s. For the first The cross-sectional area of ​​the branch pipe of each branch road.

[0060] Step 3.2: Construction of MPC Control Variables and Control Increments Set the sampling period of the MPC controller to ,exist At the sampling time, construct an MPC control vector containing the fan frequency and the parameters of each branch regulating valve: In the formula: for Control vector at any given time; for The operating frequency of the wind turbine at any given time; for Time of the first The regulating parameters of an active branch regulating valve; For active trend collection Total number of branches within; superscript This is the transpose of a vector.

[0061] Define control increment This represents the change in the control command at the current moment relative to the previous moment, used to limit the adjustment range of the actuator and avoid large system fluctuations. The specific incremental forms corresponding to each implementing agency are as follows: In the formula: For fan frequency adjustment increment; For the first The adjustment increment of each branch regulating valve.

[0062] Step 3.3: Constructing the energy efficiency optimal objective function Set the prediction step size of MPC to Typically, a value of 5-20 is used, calibrated based on the inertia of the pipeline system and the sampling period; the control step size is... ,generally Within the prediction time domain, an energy efficiency-optimal objective function is constructed, which focuses on minimizing wind turbine operating energy consumption while also considering control stability and constraint satisfaction. The optimization goal is The parameters in the formula are defined as follows: This is the weighting coefficient for wind turbine energy consumption, a core optimization item, and its value is prioritized over other weights. To control the incremental weighting coefficient, which is used to limit the adjustment range of the actuator and ensure the stable operation of the system; This is the soft constraint penalty weight coefficient, used to penalize the breach of the minimum airflow constraint. It is usually set to a large value to ensure that the constraint is satisfied first. for The constant fan shaft power is calculated using the following formula: ; In the formula: This is the full pressure of the blower; This refers to the total air volume of the system. This represents the overall efficiency of the wind turbine unit.

[0063] To control the squared incremental L2, which is used to limit the adjustment rate of the actuator and avoid large movements that could impact the equipment; For the minimum air volume soft constraint relaxation variable, This is used to handle constraint softening under extreme working conditions, avoid the optimization problem being unsolvable, and make it tend to 0 by penalizing the objective function; A collection of active trending sectors; Step 3.4: Setting Constraints To ensure the safety of the dust removal system in preventing sediment deposition and the safe operation of the actuators, all constraints are applied at all discrete moments within the prediction time domain. ( The following conditions are met: Core safety constraint: Minimum airflow constraint There are two types of constraints: hard constraints and soft constraints, which can be selected according to the working conditions. Hard constraint type, no relaxation, strictly satisfied, preferred for normal operating conditions: The soft constraint form, with slack variables, allows for limited overcoming under extreme conditions, avoiding the possibility of no solution in the optimization: physical boundary constraints of the actuator Fan frequency constraints: In the formula: This is the minimum permissible operating frequency for the wind turbine. This is the maximum permissible operating frequency for the wind turbine.

[0064] Branch regulating valve constraints: In the formula: For the first Minimum opening limit of each branch control valve, This is the maximum opening limit.

[0065] Pipeline safety pressure constraints:

[0066] In the formula: This is the minimum allowable pressure limit for the pipeline network. This is the maximum allowable pressure limit for the pipeline network, calibrated according to the pipeline network's design pressure resistance level and dust removal process requirements.

[0067] Step 3.5: Optimization Problem Solving By integrating the objective function and constraints described above, a constrained quadratic programming optimization problem is constructed:

[0068] Industrial-grade quadratic programming algorithms such as the effective set method and interior point method are used to solve the optimization problem in each sampling period to obtain the optimal control quantity sequence in the prediction time domain. The first control quantity in the sequence is taken as the optimal control quantity combination at the current moment for execution.

[0069] Step 4: Rolling optimization execution and online model correction The core of this step is to execute the optimal control command, verify the control effect in real time, correct model errors online, form a closed-loop rolling optimization, and achieve optimal safety and energy efficiency control under all operating conditions.

[0070] Step 4.1: Issuance and execution of optimal control commands The optimal wind turbine frequency at the current moment obtained by solving the MPC algorithm. Optimal adjustment parameters for each active branch control valve The data is transmitted to the fan inverter and the actuators of each branch regulating valve via industrial buses such as Profinet / Modbus, and the corresponding regulating actions are executed.

[0071] Step 4.2: Real-time verification of control effect After the adjustment action is executed, data is collected in real time through the pipeline sensing unit. , , , Operational data, such as valve position feedback, were compared with model predictions to complete three core verifications: Security verification: All active branches Real-time air volume All satisfy the minimum airflow constraint, under hard constraint Under soft constraints The penalty level approaches zero, achieving dust deposition control. Energy efficiency verification: Under the premise of meeting all constraints, the fan shaft power The coordinated optimization of the fan frequency and branch valve position significantly reduced the energy consumption, thereby improving the system's energy efficiency. Model error verification: Compare the measured operating data with the model prediction values, calculate the prediction error of the pipeline resistance coefficient, and determine whether model correction is needed.

[0072] Step 4.3: Online correction of model parameters When operating conditions change, such as dust accumulation, filter bag pressure difference, pipeline aging, and air outlet disturbance, the overall resistance coefficient of the pipeline network will increase. Branch resistance coefficient When a deviation occurs and the model prediction error exceeds a preset threshold, the recursive least squares method or Kalman filter algorithm is used to estimate and correct the drag coefficient online to suppress the prediction error.

[0073] Taking the recursive least squares method as an example, the algorithm flow for online correction of the resistance coefficient of a single branch is as follows: First, construct the parameter estimation model: for the first... Branch road, based on Let the observed value Regression vector Parameters to be estimated The linear regression model is obtained as follows: In the formula For measuring noise.

[0074] Recursive update formula: In the formula: for Time of the first Estimated value of branch resistance coefficient; Kalman gain; It is the covariance matrix; It is an identity matrix.

[0075] The above algorithm updates the resistance coefficients of the pipeline network and branches in real time, corrects the pipeline network resistance model, and ensures that the rolling optimization is always based on the latest actual operating conditions of the pipeline network.

[0076] Step 4.4: Closed-loop rolling optimization In each MPC sampling period Within the system, the optimization problem-solving process of step S3 and the execution, verification, and correction process of step S4 are repeated in a rolling manner to achieve dynamic energy efficiency optimal closed-loop control of the dust removal pipeline network, which always meets the minimum air volume constraint for preventing deposition under dynamic changes in all operating conditions.

[0077] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A smart dust collection pipeline network energy efficiency optimization MPC control method oriented towards minimum air volume constraints, applied to a dust collection pipeline network system including a fan, main pipeline, multiple dust collection branches, regulating valves and dust hoods for each branch, pipeline network sensing unit, and controller, characterized in that, It includes the following steps: S1. Collect the real-time operation status data of the dust removal pipe network through the pipe network sensing unit to obtain the structural topology and pipe network component parameters of the dust removal pipe network; establish a full-system pipe network resistance model based on the structural topology and pipe network component parameters, and establish a fan model containing the coupling relationship between the fan frequency and air volume based on the fan performance curve; S2. Determine the active tuyere set based on the collected multi-source working condition interlock signals, set zero air volume or valve closing constraints for the inactive tuyeres, and only perform joint scheduling of safety and energy efficiency within the active tuyere set; Construct a hierarchical resistance topology of the dust removal pipe network based on the active tuyere set, calculate the current total resistance of the system through equivalent synthesis of series-parallel pipe network resistances, predict the total air volume of the system based on the total system resistance and the fan model, calculate the initial flow prediction value of each active tuyere according to the damping inverse proportion distribution principle, and analyze the influence law of working condition changes on the branch flow distribution based on the initial flow prediction value; S3. For each branch within the active tuyere set, convert to obtain the minimum air volume threshold of each branch according to the minimum conveying wind speed for preventing dust deposition and the corresponding branch pipe cross-sectional area; construct an MPC control quantity including the fan operating frequency and the adjustment parameters of the regulating valves of each active branch under a preset sampling period; within a preset prediction step, construct an energy efficiency optimal objective function with minimizing the fan operating energy consumption as the core optimization objective, and at the same time set safety and actuator constraint conditions including at least the minimum air volume constraint, and solve to obtain the optimal control quantity combination that satisfies all constraint conditions; S4. Send the obtained optimal control quantity combination to the fan frequency converter and the regulating valve actuators of each branch for execution; Collect the pipe network operation feedback data in real time, compare and verify the feedback data with the model prediction values. When the pipe network working conditions change and cause the model parameters to deviate, online correct the parameters of the pipe network resistance model based on the measured operation data; Repeat steps S3 and S4 in a rolling manner to achieve dynamic energy efficiency optimal closed-loop control under the safety anti-deposition constraint of the dust removal pipe network.

2. The intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum air volume constraints as described in claim 1, characterized in that, The specific steps of S1 include: S11. Collect the real-time operation status data of the dust removal pipe network through the sensing units arranged at various places of the dust removal pipe network, and at the same time obtain the structural topology information of the dust removal pipe network and the basic parameters of each component in the pipe network; S12. Establish a full-system pipe network resistance model that can predict the pipe network pressure and flow conditions based on the obtained pipe network structural topology information and the basic parameters of each component; S13. Establish a fan model containing the coupling relationship between the fan operating frequency and air volume based on the performance curve of the fan supporting the dust removal pipe network. The pipe network resistance model and the fan model jointly provide a calculation basis for subsequent optimization control for working condition prediction.

3. The intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum air volume constraints as described in claim 1, characterized in that, The specific steps of S2 include: S21. Determine the dust removal tuyeres that need to maintain suction currently based on the collected multi-source working condition interlock signals,划定活跃风口集合,对非活跃风口设置零风量或关阀约束,仅在活跃风口集合内进行后续的安全与能效联合调度; S22. Construct a hierarchical resistance topology relationship of the dust removal pipe network based on the pipe network structure corresponding to the active tuyere set, and clarify the series-parallel logic of each branch; S23. Based on the principles of pressure balance and mass conservation in fluid mechanics, the resistance of the pipeline network corresponding to the active air outlet is simplified by equivalent calculation to obtain the current total resistance of the pipeline system. S24. Based on the total system resistance and the total system air volume predicted by the fan model, calculate the initial flow prediction value of each active air outlet, and use it as the initial state quantity for subsequent rolling optimization. S25. Based on the initial flow forecast, analyze the impact of different operating conditions on the flow distribution of each branch, and identify the risk conditions in which the branch air volume falls below the safety threshold.

4. The intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum air volume constraints as described in claim 1, characterized in that, The specific steps in S3 include: S31. For each dust removal branch within the active air outlet cluster, determine the minimum air volume threshold corresponding to each branch based on the safety requirements for dust anti-deposition and the branch pipeline parameters, as the safety protection red line of the dust removal system. S32. Based on the preset control sampling period, construct the MPC control quantity, which includes the fan operating frequency and the adjustment parameters of each active branch regulating valve; S33. Within the preset prediction step size, construct an energy efficiency optimal objective function with the minimum energy consumption of wind turbine operation as the core optimization objective, while also taking into account the system operation stability and constraint satisfaction. S34. To optimize the problem, set up multiple sets of constraints, including minimum air volume constraints, actuator physical boundary constraints, and pipeline safety pressure constraints. S35. Integrate the objective function and constraints to construct a constrained optimization problem. After solving the problem, obtain the optimal combination of control variables that can be issued and executed at the current moment.

5. The intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum air volume constraints as described in claim 1, characterized in that, The specific steps in S4 include: S41. The optimal control quantity combination obtained from the solution is sent to the wind turbine frequency converter and the actuator of each branch regulating valve to complete the corresponding regulation action; S42. Collect pipeline operation feedback data after adjustment actions in real time, compare it with the model prediction value, and complete the three verifications of safety protection effect, energy efficiency optimization effect and model prediction accuracy. S43. When changes in pipeline operating conditions cause model parameters to shift or prediction accuracy to exceed a preset threshold, the parameters of the pipeline resistance model are corrected online based on measured operating data to suppress prediction errors. S44. According to the preset sampling period, the entire process of problem solving, control execution, and model correction is repeatedly optimized to achieve dynamic closed-loop optimal control of the dust removal pipeline network under all operating conditions.

6. The intelligent dust removal pipeline energy efficiency optimization MPC control method for minimum air volume constraints according to claim 1 or 3, characterized in that, The multi-source interlocking signals include vent switch signals, valve position feedback signals, dust hood process signals, and dust removal point interlocking signals. The vent switch signals are switch status signals indicating whether each dust removal branch is open. The valve position feedback signals are feedback signals indicating the actual opening degree of the regulating valves in each branch. The dust hood process signals are operating status signals of the dust-generating equipment or process station corresponding to the dust hood, used to indicate whether there is a need for ventilation at the corresponding point. The dust removal point interlocking signals are dust removal branch interlock activation signals generated based on equipment start / stop, process conditions, or safety logic.

7. The intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum air volume constraints according to claim 3, characterized in that, The risk-identifying conditions in S25 include: Identify risky operating conditions when air outlets switch: When some air outlets switch from an inactive state to an active state, the range of active air outlets expands and the total air volume demand of the system increases. If the pressure difference of the pipeline is maintained only by throttling the branch valves, the air volume of the original active branches is easily squeezed out, resulting in the risk of insufficient air volume. Identify time-varying resistance risk conditions: When the branch resistance coefficient increases with dust accumulation, filter bag pressure difference, and pipeline aging, the air volume of the corresponding branch will decrease under the same fan frequency operating conditions, and there is a risk that the air volume will fall below the minimum air volume threshold. Identify risky operating conditions for energy-saving frequency regulation: When the fan frequency is reduced to lower the operating energy consumption, the total air volume of the system decreases. The strongly coupled pipe network will cause a nonlinear redistribution of the air volume of each branch. The terminal branches with greater resistance will be the first to fall below the minimum air volume threshold. The minimum air volume constraint needs to be explicitly incorporated into the optimization control system.

8. The intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum air volume constraints according to claim 4, characterized in that, The minimum airflow core safety constraint for preventing dust deposition requires that the operating airflow of all active dust removal branches meet the minimum airflow threshold requirement of the corresponding branch, serving as an insurmountable safety boundary for optimized control. The physical boundary constraint on the operating frequency of the wind turbine limits the operating frequency of the wind turbine to be within the allowable operating frequency range of the equipment, so as to ensure the safe and stable operation of the wind turbine equipment. The boundary constraint of the adjustment range of the branch regulating valve limits the adjustment parameters of the regulating valves of each active branch to be within the physically executable adjustment range of the equipment, so as to avoid the valves from operating beyond the range. The safety constraint on the operating pressure of the pipeline network limits the operating pressure of the dust removal pipeline network to be within the pressure range of the pipeline network design pressure resistance level and the dust removal process requirements, so as to ensure the overall operational safety of the pipeline network system.

9. A smart dust removal pipeline network energy efficiency optimization MPC control system oriented towards minimum airflow constraints, applied to a dust removal pipeline network system, wherein the dust removal pipeline network system includes a fan, a main pipeline, multiple dust removal branches, regulating valves and dust collection hoods for each branch, and a pipeline network sensing unit, characterized in that, include: The data acquisition and model building module is used to collect real-time operating status data of the dust removal pipeline network through the pipeline network sensing unit, and obtain the structural topology and pipeline network component parameters of the dust removal pipeline network; it is also used to establish a full-system pipeline network resistance model based on the structural topology and pipeline network component parameters, and to establish a fan model that includes the coupling relationship between fan frequency and air volume based on the fan performance curve; The active air outlet identification and operating condition analysis module is used to determine the set of active air outlets based on the collected multi-source operating condition interlocking signals, set zero air volume or valve closure constraints for inactive air outlets, and perform joint scheduling of safety and energy efficiency only within the set of active air outlets; it is also used to construct a hierarchical resistance topology of the dust removal pipeline network based on the set of active air outlets, obtain the current total resistance of the system through equivalent synthesis calculation of series and parallel pipeline network resistance, predict the total air volume of the system based on the total system resistance and the fan model, calculate the initial flow prediction value of each active air outlet according to the damping inverse proportional distribution principle, and analyze the impact of operating condition changes on branch flow distribution based on the initial flow prediction value; The MPC optimization solution module is used to calculate the minimum air volume threshold for each branch within the set of active air outlets, based on the minimum delivery velocity for preventing dust deposition and the corresponding branch cross-sectional area. It is also used to construct MPC control quantities including the fan operating frequency and the adjustment parameters of the regulating valves of each active branch within a preset sampling period. Furthermore, it is used to construct an energy efficiency optimal objective function with the minimum fan operating energy consumption as the core optimization objective within a preset prediction step size, while setting safety and actuator constraints that include at least the minimum air volume constraint, and solving for the optimal combination of control quantities that satisfies all constraints. The closed-loop execution and rolling optimization module is used to send the optimal combination of control quantities obtained from the solution to the wind turbine frequency converter and the actuators of each branch regulating valve for execution; It is also used to collect pipeline operation feedback data in real time, compare and verify the feedback data with the model prediction values, and correct the parameters of the pipeline resistance model online based on the measured operation data when the pipeline operating conditions change and the model parameters shift. It is also used to call the corresponding functions of the MPC optimization solution module and the closed-loop execution and rolling optimization module in a rolling manner to realize the dynamic energy efficiency optimal closed-loop control under the safety and anti-deposition constraints of the dust removal pipeline network.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed, implements the intelligent dust removal pipeline energy efficiency optimization MPC control method oriented towards minimum air volume constraints as described in any one of claims 1-8.

Images