CN-121997758-A - Dynamic dust concentration and wind net balance optimization method and system in dust removal equipment pipeline

Abstract

The invention relates to the technical field of industrial dust removal and dust monitoring, in particular to a method and a system for optimizing dynamic dust concentration and wind net balance in a dust removal device pipeline. The state estimation method is used for constructing a state vector containing parameters such as dust concentration, wind speed, wind pressure and the like, combining Kalman filtering physical constraint with pre-training GRU network time sequence analysis, obtaining optimal estimation through correction, weighting and fusion of a predicted value, determining a risk level through three-level threshold value, concentration trend quantification and material blocking confidence calculation based on the estimation by the operation risk judgment method, establishing a relation model of branch pipe wind speed, valve opening and total wind pressure by the opening prediction method, and optimizing the valve opening by taking wind network balance and energy conservation as targets. The technical problems of inaccurate state estimation, risk prejudging hysteresis, unbalanced wind net and high energy consumption of the dust removing system are effectively solved.

Inventors

• LIU QIANBO
• WEI YUYI
• YU SHUANGLIN
• Xi Feiruo
• WANG JIANHUA
• YU GANG
• YANG MINGYUAN
• Liang Weifei
• He haining
• LI WEI
• MAO LEI
• WEI CAIWANG
• Hao Yuncong
• LI YI
• LU SHAOXIN
• HUANG MINJIE

Assignees

• 广州港股份有限公司南沙粮食通用码头分公司
• 合肥弘恩机电科技有限公司

Dates

Publication Date

20260508

Application Date

20260129

Claims (10)

1. 1. The state estimation method of the dust removing equipment pipeline is characterized by comprising the following estimation steps: Constructing a state vector containing the dust concentration, the wind speed, the wind pressure and the change rate of the dust concentration, the wind pressure and the change rate of the wind pressure, and loading an observation noise covariance matrix; based on the state optimal estimated vector at the previous moment, carrying out state prediction under physical rule constraint through a state transition matrix to obtain a predicted state vector at the current moment and a prediction error covariance matrix thereof; Performing time sequence feature analysis on the historical observation sequence and the real-time observation state vector at the current moment by utilizing a pre-trained gating circulation unit network so as to distinguish real state change from abnormal interference and output a corrected observation state vector; And carrying out weighted fusion on the predicted state vector and the corrected observed state vector by utilizing the Kalman gain to obtain a state optimal estimated vector at the current moment.
2. 2. The method for estimating the state of a dust removing apparatus pipe according to claim 1, wherein the state prediction is implemented by the following state prediction equation and covariance prediction equation: state prediction equation: ; Covariance prediction equation: ; in the formula, A predicted state vector based on the state optimal estimation vector at the previous time t-1 for the current time t, wherein A is a state transition matrix, and A T is a transposition of A; The method comprises the steps of setting a state optimal estimation vector of a previous time t-1, setting B as a control input matrix, setting u t-1 as a control quantity of the previous time t-1, setting P t|t-1 as a prediction error covariance matrix of the state optimal estimation vector of the current time t based on the previous time t-1, setting P t-1|t-1 as a prediction error covariance matrix of the state optimal estimation vector of the previous time t-1, and setting Q as a process noise covariance matrix.
3. 3. The method for estimating the state of a dust removing apparatus pipe according to claim 2, wherein the state optimum estimated vector at the present time is expressed as follows: ; ; in the formula, The method comprises the steps of estimating a vector for the optimal state of a current time t, K t is Kalman gain of the current time t, Z t is a corrected observation state vector of the current time t, H is an observation matrix, H T is a transpose of H, and R is an observation noise covariance matrix.
4. 4. The operation risk judging method of the dust removing equipment pipeline is characterized by comprising the following judging steps of: Based on the state estimation method of the dust removing equipment pipeline of any one of claims 1-3, rolling prediction is carried out so as to output dust concentration prediction sequences of a plurality of time steps in the future; Comparing the real-time dust concentration of each monitoring point with a preset three-level concentration threshold value to generate a first alarm level signal; Calculating the instantaneous slope of concentration change by adopting a linear regression method as a quantification index of concentration accumulation trend according to the real-time dust concentration data of a plurality of historical moments of each monitoring point, and calculating the average predicted concentration change rate of a plurality of future moments according to a dust concentration prediction sequence according to each monitoring point; The instantaneous slope and the average predicted concentration change rate are fused, the real-time wind speed attenuation rate of the corresponding monitoring point is combined, and the comprehensive confidence coefficient of the blocking risk of the monitoring point is calculated through a preset confidence coefficient calculation model; and determining a final risk level according to the first alarm level signal and the aggregate risk confidence level through a predefined risk mapping rule.
5. 5. The method for determining the running risk of a dust removing equipment pipeline according to claim 4, wherein the calculation formula of the instantaneous slope is as follows: ; Wherein k i is the instantaneous slope of the dust concentration change of the ith monitoring point, M is the total number of historical time, t t-M+m represents the historical time of M-M time steps from the current time t; C i,t-M+m is the dust concentration of the ith monitoring point at t t-M+m ; the average dust concentration at the M historical moments of the ith monitoring point.
6. 6. The method for determining the running risk of a dust removing equipment pipeline according to claim 5, wherein the confidence coefficient calculation model is represented as follows: ; ; Wherein Conf i,block is the comprehensive confidence coefficient of the blocking risk of the ith monitoring point, and min (&) is the minimum value; c i,t is the dust concentration of the ith monitoring point at the moment t, delta t is the time variation; The real-time wind speed attenuation rate of the ith monitoring point; 、 the judgment threshold values are respectively the predicted concentration change and the wind speed decay.
7. 7. The method for determining the running risk of a dust removing equipment pipeline according to claim 6, wherein the risk mapping rule is as follows: if the first alarm level signal of the current monitoring point is that the comprehensive confidence coefficient of the emergency alarm or the putty risk is larger than a preset high confidence coefficient threshold value, judging that the current monitoring point is high risk; If the first alarm level signal of the current monitoring point is that the comprehensive confidence of the early warning or blocking risk is between a preset high confidence threshold and a preset low confidence threshold, judging that the current monitoring point is a middle risk; And if the first alarm level signal of the current monitoring point is normal or the comprehensive confidence coefficient of the putty risk is lower than a preset low confidence coefficient threshold value, judging that the current monitoring point is risk-free.
8. 8. The method for predicting the opening of the dust removing equipment pipeline is characterized by comprising the following prediction steps: Acquiring real-time wind speed data, real-time wind pressure data and valve real-time opening data of each monitoring point, and obtaining final risk level and blocking risk comprehensive confidence level output by the operation risk judging method of the dust removing equipment pipeline according to any one of claims 4-7; Based on the real-time wind speed, the real-time wind pressure and the real-time opening data of the valve, a quantitative relation model of the wind speed of each branch pipe, the opening of the valve and the total wind pressure is established; taking the minimum wind speed unbalance among the branch pipes as a first optimization target, and taking the minimum total energy consumption as a second optimization target, so as to construct a multi-target optimization problem; converting the final risk level and the comprehensive confidence coefficient of the blocking risk into the minimum wind speed constraint of the corresponding branch pipe, and using the minimum wind speed constraint as the wind speed basic constraint to jointly form the constraint condition of the multi-objective optimization problem; solving the multi-objective optimization problem to obtain an optimal opening instruction of each branch pipe valve in the next control period, and controlling the opening of the valve according to the optimal opening instruction.
9. 9. The method for predicting the opening degree of a dust removing apparatus pipe according to claim 8, wherein the quantitative relation model is expressed as follows: ; Wherein v j is the real-time wind speed of the jth branch pipe, o j is the real-time opening of the valve of the jth branch pipe, eta j is the real-time resistance coefficient of the jth branch pipe, and P is the total wind pressure; and/or updating the real-time resistance coefficient by a recursive least square method with forgetting factors: ; in the formula, 、 Respectively the real-time resistance coefficients of the jth branch pipe in the n-1 th iteration and the n-1 th iteration; 、 Respectively the real-time wind speed and the real-time opening of the valve of the jth branch pipe in the nth round of iteration; the total wind pressure is the total wind pressure during the nth iteration; and/or, the first optimization target is to minimize a wind speed unbalance index U, and the calculation formula is as follows: ; Wherein J represents the total number of branch pipes; Representing the average wind speed of all the branches; And/or, the minimum wind speed constraint is expressed as follows: ; Wherein v j,min is the minimum wind speed of the jth branch pipe, v min,base is the basic process required wind speed, L j,final is the final risk level of the jth branch pipe, the final risk level of all monitoring points in the jth branch pipe is taken as the average value, conf j,block is the comprehensive confidence coefficient of the blocking risk of the jth branch pipe, the final risk level of all monitoring points in the jth branch pipe is taken as the average value, and beta L 、β C is the corresponding calibration coefficient respectively.
10. 10. A control system for a dust removing apparatus duct, comprising: A state estimation unit configured to perform a state estimation method of a dust removing equipment pipeline according to any one of claims 1 to 3, so as to output a real-time state optimal estimation vector and a dust concentration prediction sequence of each monitoring point after cleaning and fusion; An operation risk judging unit, which is in communication connection with the state estimating unit and is configured to execute the operation risk judging method of the dust removing equipment pipeline according to any one of claims 4-7, receive the real-time state optimal estimating vector and the dust concentration predicting sequence, and output a comprehensive risk judging result comprising the comprehensive confidence of the risk grade and the blocking risk; A dynamic balance unit, which is respectively connected with the state estimation unit and the operation risk determination unit in a communication way, and is configured to execute the opening prediction method of the dust removal equipment pipeline according to claim 8 or 9, receive real-time wind speed, wind pressure, valve opening data and comprehensive risk determination results, and output an optimal opening instruction of each branch pipe valve; The arbitration and execution unit is in communication connection with the running risk judging unit and the dynamic balancing unit, and is configured to arbitrate the optimal opening instruction from the dynamic balancing unit according to the risk level in the comprehensive risk judging result, generate a final control instruction and send the final control instruction to the corresponding valve executing mechanism.

Description

Dynamic dust concentration and wind net balance optimization method and system in dust removal equipment pipeline Technical Field The invention relates to the technical field of industrial dust removal and dust monitoring, in particular to a method and a system for optimizing dynamic dust concentration and wind net balance in a dust removal device pipeline. Background In the industrial scenes of port bulk grain ship unloading, grain conveying, mine metallurgy and the like, the dust removing system is core equipment for guaranteeing production safety and controlling dust pollution, and the operation state of the dust removing system is directly related to operation continuity and personnel and equipment safety. Under such a scene, the dust removal system is in a continuous running state with high dust concentration and high airflow load for a long time, key parameters such as dust concentration, wind speed, wind pressure and the like in a pipeline are easy to change in a nonlinear way along with fluctuation of production load, and if the change trend of the control parameters is not timely carried out, serious risks such as pipeline blockage, dust removal efficiency reduction and even dust explosion are extremely easy to be caused. The existing dust removing system generally adopts a single-point monitoring and alarming mechanism based on a fixed threshold to realize state monitoring by arranging sensors at key positions, namely, triggering alarming by comparing data collected by the sensors with a preset threshold, and adjusting the frequency of a fan, the opening of a valve and the like by relying on manual experience at a control layer or adopting feedback control logic based on the current/historical state. The technical means essentially belongs to a post-hoc response mode, can only prompt risks when an abnormality occurs or is close to a critical state, and the control logic of the monitoring, alarming and executing mechanism is mutually split, so that overall perception and prospective regulation and control capability of the global state are lacked. Although the prior art realizes basic state monitoring and control, the following technical problems still exist: Firstly, sensor data is easily affected by dust interference and airflow fluctuation to generate noise and drift, single monitoring data is not reliable enough, and a fusion mechanism of physical rule constraint and time sequence feature learning is lacked, so that dynamic dust concentration and system state estimation are inaccurate, and follow-up accurate decision is difficult to support. And secondly, risk judgment depends on single threshold comparison, only whether the current parameter exceeds standard is concerned, comprehensive judgment is carried out by not combining multidimensional information such as concentration change trend, wind speed attenuation characteristics and the like, the risk of blocking germination and dust explosion accumulation cannot be recognized in advance, and early warning hysteresis is outstanding. Thirdly, the wind network regulation adopts a single-point independent control mode, a global state evaluation and multi-objective optimization model is not established, the phenomena of wind robbing and wind stagnation of the branch pipes are easily caused, the balance degree of the wind network is poor, the opening degree of the pipeline valve is difficult to accurately regulate and control, and energy consumption optimization cannot be realized on the premise of ensuring safety, so that energy waste is caused. It follows that there is a need for further improvements in current dust removal systems. Disclosure of Invention The invention provides a state estimation method of a dust removing device pipeline, which aims to solve the technical problem that the state estimation of the dust removing device pipeline is inaccurate due to the fact that sensor data are interfered by dust and air flow fluctuates. In order to solve the technical problem that the risk judgment depends on single threshold comparison and only pays attention to whether the current parameter exceeds the standard or not, so that the pipeline safety early warning is lagged, the invention provides a method for judging the operation risk of a dust removing equipment pipeline on the basis of the state estimation method. The invention provides a method for predicting the opening of a pipeline of dust removal equipment based on the operation risk judging method, aiming at solving the technical problem that the opening of each valve of the pipeline is difficult to accurately regulate and control due to the fact that a single-point independent control mode is adopted for wind network regulation and control and risk balance is not considered. The invention further provides a control system of the dust removing equipment pipeline. In order to achieve the above purpose, the present invention provides the following technical solutions: The state estimation met