CN-121979065-A - High-end automatic ash conveying big data energy-saving control method and system

CN121979065ACN 121979065 ACN121979065 ACN 121979065ACN-121979065-A

Abstract

The invention discloses a high-end automatic ash conveying big data energy-saving control method and system, and belongs to the technical field of energy-saving ash conveying. The method specifically comprises the steps of calculating a pipeline resistance coefficient and an ash gas ratio through visual measurement and multi-source time sequence data fusion, constructing a pressure flow equation, generating a bin pump pressure change curve and an air consumption curve which can maintain stable ash plug conveying in a future limited domain, taking the ash gas ratio as a guide, establishing a conveying cost function related to air inlet valve position increment and pressure tracking deviation, and solving the conveying cost function to obtain an optimal air inlet valve position sequence in the future limited domain for energy-saving ash conveying control.

Inventors

HAN XIAOJUN
HAN HUI
ZHANG AIMIN
SUI HONGBO
Yin Huiteng
FAN XIAOBIN
SHEN BAOJUN

Assignees

北京汇研中科科技发展有限公司

Dates

Publication Date: 20260505
Application Date: 20260204

Claims (10)

1. The high-end automatic ash conveying big data energy-saving control method is characterized by comprising the following steps of: acquiring upstream production data, ash conveying process data and downstream power data of an ash conveying pipe network through visual detection, and calculating to obtain a pipeline resistance coefficient and an ash gas ratio through soft measurement estimation, wherein the upstream production data comprises ash bucket material level change data; inputting a pipeline resistance coefficient and an ash-gas ratio into a pressure flow equation, generating a bin pump pressure change curve and an air consumption curve which maintain ash plug conveying in a future limited domain, wherein the pressure flow equation is constructed according to an air inlet valve position, bin pump pressure and air inlet flow; according to the bin pump pressure change curve and the air consumption curve, a conveying cost function related to the increment of the air inlet valve position and the pressure tracking deviation is established, and the optimal air inlet valve position sequence in a future limited domain is obtained by solving the conveying cost function; and sending the optimal air inlet valve position sequence to an air inlet valve for execution and adjusting a corresponding bin pump, and carrying out energy-saving control on the ash conveying pipe network according to a pressure deviation optimization instruction in the execution process.
2. The energy-saving control method for high-end automatic ash conveying big data according to claim 1, wherein the upstream production data, ash conveying process data and downstream power data of the ash conveying pipe network are collected through visual detection, the pipeline resistance coefficient and the ash gas ratio are calculated through soft measurement estimation, the upstream production data comprise ash bucket level change data, and the method comprises the following steps: Acquiring a surface image sequence of a material in a hopper in upstream production data by an industrial vision sensor, utilizing vision detection to identify gray level and texture characteristics of the surface image sequence to establish a three-dimensional contour of the material level, and measuring hopper material level change data of the three-dimensional contour of the material level; performing time sequence alignment on upstream production data, ash conveying process data and downstream power data, and removing outlier data points to obtain a conveying sample set; Calculating pressure stability characteristics of a feed stage of the bin pump, pressure attenuation gradient of the feed stage and valve position time sequence integral values of the air inlet valve according to a transport sample set, and constructing a pressure attenuation differential equation by combining a pipeline resistance coefficient as an unknown quantity, wherein the pipeline resistance coefficient is obtained according to pipeline geometry and state information; calculating root mean square error according to the pressure predicted value obtained by solving the pressure attenuation differential equation and the pressure change parameter in the ash conveying process data, and adopting gradient descent iteration to adjust the pipeline resistance coefficient until the root mean square error is smaller than a preset error threshold value to obtain an optimized pipeline resistance coefficient; Substituting the optimized pipeline resistance coefficient into a soft measurement estimation equation, and calculating to obtain the ash-gas ratio of the current conveying cycle, wherein the soft measurement estimation equation is deduced according to a preset material conservation equation and a gas state equation.
3. The energy-saving control method for high-end automatic ash conveying big data according to claim 2, wherein the calculating the pressure stability characteristic of the feeding stage of the bin pump, the pressure attenuation gradient of the conveying stage and the valve position time sequence integral value of the air inlet valve according to the conveying sample set comprises the following steps: calculating pressure variance of the pressure time sequence data of the conveying sample set extraction bin pump in the feeding stage as a pressure stability characteristic; Identifying that the pressure of the bin pump enters a monotonically descending conveying stage according to a conveying sample set, performing linear piecewise fitting on the descending stage, and taking the obtained slope as a pressure attenuation gradient; And extracting valve position instructions of the air inlet valve and time sequence data of feedback signals according to the conveying sample set, and obtaining a corresponding valve position time sequence integral value through numerical integration calculation.
4. The energy-saving control method for high-end automatic ash conveying big data according to claim 3, wherein the pipeline resistance coefficient and the ash-gas ratio are input into a pressure flow equation to generate a bin pump pressure change curve and an air consumption curve for maintaining ash plug conveying in a future limited domain, and the pressure flow equation is constructed according to an air inlet valve position, a bin pump pressure and an air inlet flow, and the method comprises the following steps: resampling a conveying sample set according to a preset time step by taking a conveying cycle of a dust conveying pipe network as a unit to generate an air inlet valve position sequence, a bin pump pressure sequence and an air inlet flow sequence; the method comprises the steps of taking an air inlet valve position sequence, a bin pump pressure sequence and an air inlet flow sequence as inputs of a preset discrete time difference equation containing a coefficient matrix to be determined, and determining a pressure flow equation in a time-varying parameter form by recursively updating the coefficient matrix to be determined; Substituting the pipeline resistance coefficient as a correction factor into a pressure flow equation in a time-varying parameter form, and establishing a jacobian matrix related to the pipeline resistance coefficient to update parameters in the pressure flow equation in the time-varying parameter form so as to obtain the pressure flow equation; taking the ash-gas ratio as a gain coefficient, constructing a differential equation set about the pressure change rate of the cabin pump and the air inlet flow rate by using the simultaneous pressure flow equation; Taking the initial solution states of the current bin pump pressure, the air inlet valve position and the differential equation set as initial conditions, and carrying out numerical integration on the differential equation set to generate a bin pump pressure change curve and an air consumption curve for ash plug conveying in a limited future domain.
5. The energy-saving control method for high-end automatic ash conveying big data according to claim 4, wherein the step of determining the pressure flow equation in the form of time-varying parameters by recursively updating the coefficient matrix to be determined by taking the intake valve position sequence, the bin pump pressure sequence and the intake flow sequence as inputs of a preset discrete time difference equation containing the coefficient matrix to be determined comprises: According to the air inlet valve position sequence, the bin pump pressure sequence and the air inlet flow sequence, a discrete time difference equation containing a undetermined coefficient matrix is constructed to serve as a basic expression form of a pressure flow equation; subtracting the bin pump pressure prediction output term in the discrete time differential equation from an actual measurement value of a bin pump pressure sequence, and establishing an error accumulation function related to a coefficient matrix to be determined through the obtained residual sequence; Deducing a normal equation with the gradient of zero according to the error cumulative function, and iteratively updating element values of the coefficient matrix to be determined until the coefficient matrix variation of adjacent iteration steps is smaller than a preset convergence threshold value, so as to obtain a converged coefficient matrix; Substituting the coefficient matrix after iteration convergence into a discrete time difference equation to generate a pressure flow equation in a time-varying parameter form.
6. The energy-saving control method for high-end automatic ash conveying big data according to claim 5, wherein the deriving the normal equation with the gradient zero according to the error accumulation function, and iteratively updating the element values of the coefficient matrix to be determined until the coefficient matrix variation of the adjacent iteration steps is smaller than a preset convergence threshold, and obtaining the converged coefficient matrix comprises: Obtaining a gradient vector expression formed by first-order partial derivatives through partial derivatives of elements of a coefficient matrix to be determined by an error accumulation function, enabling the gradient vector expression to be equal to a zero vector, and establishing a matrix equation with the coefficient matrix to be determined as an unknown quantity; decomposing a coefficient matrix to be determined in a matrix equation and a vector term, and incrementally updating the coefficient matrix to be determined and the vector term through a preset attenuation factor to obtain a normal equation, wherein the vector term is composed of actual measurement values of an air inlet valve position sequence, an air inlet flow sequence and a bin pump pressure sequence; substituting the undetermined coefficient matrix as an initial solution into a normal equation, and calculating to obtain a residual vector and a jacobian matrix; establishing a linear equation set by taking the jacobian matrix as a coefficient and taking the residual vector as a right-end term, obtaining the correction quantity of the coefficient matrix to be determined by solving the linear equation set, and carrying out iterative updating on the coefficient matrix to be determined; comparing the two norms of the element difference values corresponding to the coefficient matrix to be determined in the current iteration step and the coefficient matrix to be determined in the previous iteration step with a preset convergence threshold, if the two norms are larger than the convergence threshold, continuing to update the iteration, and if the two norms are smaller than or equal to the convergence threshold, taking the current coefficient matrix to be determined as the coefficient matrix after convergence.
7. The energy-saving control method for high-end automatic ash conveying big data according to claim 6, wherein the step of establishing a conveying cost function about the increment of the air inlet valve position and the pressure tracking deviation according to the bin pump pressure change curve and the air consumption curve, and solving the conveying cost function to obtain the optimal air inlet valve position sequence in a future limited time domain comprises the following steps: According to the bin pump pressure change curve and the gas consumption curve, determining pressure reference values and allowable gas consumption thresholds at different moments in the future, and calculating the lowest safe pressure curve in the limited future by combining the pipeline resistance coefficient, the ash gas ratio and the ash bucket material level change data; According to the minimum safety pressure curve and the theoretical minimum gas consumption, a pressure reference curve and an allowable gas consumption threshold curve are obtained through preset process safety margin compensation, and the theoretical minimum gas consumption is calculated according to the ash gas ratio and the ash conveying amount; The pressure reference curve and the bin pump pressure change curve are subjected to point-by-point difference to obtain a pressure tracking deviation sequence, an inlet valve position increment sequence is initialized to be a decision vector, and a conveying cost function is constructed according to the pressure tracking deviation sequence and the decision vector; taking the lowest safe pressure curve and the total gas consumption upper limit obtained by integrating the allowable gas consumption threshold curve as boundary conditions, and converting the boundary conditions into a linear inequality group to form a linear constraint set in a future limited time domain; and calculating the influence of the decision vector on the conveying cost function value in a feasible domain defined by the linear constraint set, and taking the value of the decision vector which meets the linear constraint set and minimizes the conveying cost function value as the optimal air inlet valve sequence in a future limited time domain.
8. The method according to claim 7, wherein the calculating the influence of the decision vector on the delivery cost function value in the feasible domain defined by the linear constraint set takes the value of the decision vector that satisfies the linear constraint set and minimizes the delivery cost function value as the optimal intake valve sequence in the future limited time domain, and the method comprises: determining a feasible region of a decision vector according to the linear constraint set, and taking the current time air inlet valve position as an initial point of iterative search; in the feasible region, partial derivatives of the decision vector of the current iteration point are calculated through a conveying cost function, gradient information of the decision vector is obtained, and a direction vector which points to the descending direction and is in the feasible region is constructed according to the gradient information; Updating the value of the decision vector along the direction vector at the initial point according to the preset iteration step length, and calculating a pressure tracking deviation sequence corresponding to the updated decision vector; Substituting the updated pressure tracking deviation sequence into a conveying cost function, calculating a conveying cost function value and iteratively updating the position of the decision vector in a feasible domain; and converting the decision vector value which minimizes the transmission cost function value and satisfies all linear constraints into an optimal air inlet valve position sequence in a future limited time domain.
9. The energy-saving control method for high-end automatic ash conveying big data according to claim 8, wherein the sending the optimal air inlet valve position sequence to the air inlet valve to execute and adjust the corresponding bin pump, and according to the pressure deviation optimization instruction in the execution process, performing energy-saving control on the ash conveying pipe network comprises the following steps: Taking the lowest pressure for maintaining stable conveying of the ash plug as a reference, and converting the air inlet valve position value in the optimal air inlet valve position sequence into an energy-saving control instruction for driving the air inlet valve; Sending the air consumption curve and the bin pump pressure change curve in the future limited time domain as feedforward values to the bin pump to adjust the output power of the bin pump; acquiring the actual pressure in the current bin pump, and generating an instantaneous pressure deviation signal reflecting the ash plug moving state and the pipeline resistance by combining the execution result of the energy-saving control instruction of the air inlet valve; and determining the conveying efficiency and the gas consumption state of the current ash plug according to the instantaneous pressure deviation signal, and adjusting the valve position of the air inlet valve to perform energy-saving ash conveying control.
10. The high-end automatic ash conveying big data energy-saving control system is used for realizing the high-end automatic ash conveying big data energy-saving control method according to any one of claims 1-9, and is characterized by comprising a working condition sensing module, an energy efficiency measuring module, a conveying optimization module and an execution module: The working condition sensing module is used for acquiring upstream production data, ash conveying process data and downstream power data of the ash conveying pipe network through visual detection, and calculating to obtain a pipeline resistance coefficient and an ash gas ratio through soft measurement estimation, wherein the upstream production data comprises ash bucket material level change data; The energy efficiency measurement module is used for inputting the pipeline resistance coefficient and the ash-gas ratio into a pressure flow equation to generate a bin pump pressure change curve and an air consumption curve which maintain ash plug conveying in a limited future domain, and the pressure flow equation is constructed according to an air inlet valve position, bin pump pressure and air inlet flow; the conveying optimization module is used for establishing a conveying cost function related to the increment of the air inlet valve position and the pressure tracking deviation according to the bin pump pressure change curve and the air consumption curve, and solving the conveying cost function to obtain an optimal air inlet valve position sequence in a future limited domain; the execution module is used for sending the optimal air inlet valve position sequence to the air inlet valve for execution and adjusting the corresponding bin pump, and carrying out energy-saving control on the ash conveying pipeline network according to the pressure deviation optimization instruction in the execution process.

Description

High-end automatic ash conveying big data energy-saving control method and system Technical Field The invention relates to a high-end automatic ash conveying big data energy-saving control method and system, and belongs to the technical field of energy-saving ash conveying. Background Pneumatic ash conveying is a process of conveying bulk materials from one or more sources to one or more destinations by using air flow as a conveying medium, the process is a key process for treating powder materials in electric power and chemical plants, ash conveying pipelines in thermal power plants can convey fly ash generated by boiler combustion, and ash conveying pipelines in construction material production processes can convey building materials such as cement, limestone powder and the like, so that ash conveying capability is also a direct expression of economic benefits of enterprises. In the prior art, a fuzzy control or expert system is generally adopted in the control method of pneumatic ash conveying, automatic sequential start-stop and back blowing frequency adjustment of a bin pump are realized through a preset pressure time or material level time rule, however, the prior art has the following defects that compressed air is wasted and conveying is affected due to timing or constant pressure difference back blowing, cooperation with other production links is lacking, and electric energy of the extraction amount can be wasted on throttling and heating by closing an inlet valve or bypass backflow under a light load working condition. Disclosure of Invention The invention aims to provide a high-end automatic ash conveying big data energy-saving control method and system, which are used for solving the problems of high ash conveying cost, high stability and insufficient synergy in the prior art. In order to solve the technical problems, the invention is realized by adopting the following technical scheme. The high-end automatic ash conveying big data energy-saving control method comprises the following steps: acquiring upstream production data, ash conveying process data and downstream power data of an ash conveying pipe network through visual detection, and calculating to obtain a pipeline resistance coefficient and an ash gas ratio through soft measurement estimation, wherein the upstream production data comprises ash bucket material level change data; inputting a pipeline resistance coefficient and an ash-gas ratio into a pressure flow equation, generating a bin pump pressure change curve and an air consumption curve which maintain ash plug conveying in a future limited domain, wherein the pressure flow equation is constructed according to an air inlet valve position, bin pump pressure and air inlet flow; according to the bin pump pressure change curve and the air consumption curve, a conveying cost function related to the increment of the air inlet valve position and the pressure tracking deviation is established, and the optimal air inlet valve position sequence in a future limited domain is obtained by solving the conveying cost function; and sending the optimal air inlet valve position sequence to an air inlet valve for execution and adjusting a corresponding bin pump, and carrying out energy-saving control on the ash conveying pipe network according to a pressure deviation optimization instruction in the execution process. Specifically, the upstream production data, the ash conveying process data and the downstream power data of the ash conveying pipe network are collected through visual detection, the pipeline resistance coefficient and the ash gas ratio are obtained through soft measurement and estimation in a calculating mode, the upstream production data comprise ash bucket material level change data, and the method comprises the following steps: Acquiring a surface image sequence of a material in a hopper in upstream production data by an industrial vision sensor, utilizing vision detection to identify gray level and texture characteristics of the surface image sequence to establish a three-dimensional contour of the material level, and measuring hopper material level change data of the three-dimensional contour of the material level; performing time sequence alignment on upstream production data, ash conveying process data and downstream power data, and removing outlier data points to obtain a conveying sample set; Calculating pressure stability characteristics of a feed stage of the bin pump, pressure attenuation gradient of the feed stage and valve position time sequence integral values of the air inlet valve according to a transport sample set, and constructing a pressure attenuation differential equation by combining a pipeline resistance coefficient as an unknown quantity, wherein the pipeline resistance coefficient is obtained according to pipeline geometry and state information; calculating root mean square error according to the pressure predicted value obtained by solving the pressure attenu