CN118050995B - Self-adaptive parameter adjusting method of electric dust collector based on concentration distribution of particulate matters - Google Patents
Self-adaptive parameter adjusting method of electric dust collector based on concentration distribution of particulate matters Download PDFInfo
- Publication number
- CN118050995B CN118050995B CN202410451496.8A CN202410451496A CN118050995B CN 118050995 B CN118050995 B CN 118050995B CN 202410451496 A CN202410451496 A CN 202410451496A CN 118050995 B CN118050995 B CN 118050995B
- Authority
- CN
- China
- Prior art keywords
- dust removal
- historical
- electric
- parameters
- parameter group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000000428 dust Substances 0.000 title claims abstract description 909
- 238000000034 method Methods 0.000 title claims abstract description 73
- 239000013618 particulate matter Substances 0.000 claims abstract description 130
- 239000002245 particle Substances 0.000 claims abstract description 113
- 238000005457 optimization Methods 0.000 claims abstract description 83
- 230000008569 process Effects 0.000 claims abstract description 36
- 230000003044 adaptive effect Effects 0.000 claims abstract description 15
- 238000012360 testing method Methods 0.000 claims description 76
- 230000005684 electric field Effects 0.000 claims description 34
- 238000013507 mapping Methods 0.000 claims description 31
- 230000000875 corresponding effect Effects 0.000 claims description 25
- 239000012716 precipitator Substances 0.000 claims description 12
- 238000001556 precipitation Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 6
- 230000002596 correlated effect Effects 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 17
- 230000006870 function Effects 0.000 description 30
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 10
- 239000003546 flue gas Substances 0.000 description 10
- 230000008859 change Effects 0.000 description 7
- 230000008901 benefit Effects 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000000746 purification Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B03—SEPARATION OF SOLID MATERIALS USING LIQUIDS OR USING PNEUMATIC TABLES OR JIGS; MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
- B03C—MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
- B03C3/00—Separating dispersed particles from gases or vapour, e.g. air, by electrostatic effect
- B03C3/02—Plant or installations having external electricity supply
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B03—SEPARATION OF SOLID MATERIALS USING LIQUIDS OR USING PNEUMATIC TABLES OR JIGS; MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
- B03C—MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
- B03C3/00—Separating dispersed particles from gases or vapour, e.g. air, by electrostatic effect
- B03C3/34—Constructional details or accessories or operation thereof
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B03—SEPARATION OF SOLID MATERIALS USING LIQUIDS OR USING PNEUMATIC TABLES OR JIGS; MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
- B03C—MAGNETIC OR ELECTROSTATIC SEPARATION OF SOLID MATERIALS FROM SOLID MATERIALS OR FLUIDS; SEPARATION BY HIGH-VOLTAGE ELECTRIC FIELDS
- B03C2201/00—Details of magnetic or electrostatic separation
- B03C2201/24—Details of magnetic or electrostatic separation for measuring or calculating of parameters, e.g. efficiency
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Electrostatic Separation (AREA)
Abstract
The application discloses an adaptive parameter adjusting method of an electric dust collector based on particulate matter concentration distribution, belonging to the field of intelligent control, wherein the method comprises the following steps: collecting particle distribution information in a plurality of dust removal plate areas where the electric dust remover removes dust; according to the particle distribution information, performing historical search to obtain a plurality of historical electric dust removal parameters adopted in the process of homogeneous particle distribution information; outputting M historical electric dust removal parameters as a historical dust removal parameter group; and constructing an electric dust removal function, carrying out search optimization on electric dust removal parameters by combining historical dust removal parameter groups and particulate matter distribution information to obtain optimal electric dust removal parameters, and carrying out parameter adjustment and dust removal on the electric dust remover. The application achieves the technical effect of adaptively optimizing parameters of the electric dust collector according to the concentration distribution of the particulate matters so as to remove the dust, thereby improving the dust removal efficiency.
Description
Technical Field
The invention relates to the field of intelligent control, in particular to an adaptive parameter adjusting method of an electric dust collector based on particulate matter concentration distribution.
Background
As common industrial dust removing equipment, the electric dust remover is used for charging particles in the flue gas by applying a high-voltage electric field, and the particles are adsorbed onto a dust removing polar plate under the action of electrostatic force, so that the purpose of purifying the flue gas is achieved. However, in the prior art, the control parameters of the electric dust collector are usually fixed and set according to experiments and experience, and the control mode lacks of intelligence and self-adaption and is difficult to cope with the change of the concentration distribution of the particulate matters under the actual working condition. When the flue gas condition changes, the fixed control parameters cannot guarantee the optimal dust removal efficiency.
Disclosure of Invention
The application provides a self-adaptive parameter adjusting method of an electric dust collector based on particulate matter concentration distribution, and aims to solve the technical problems that in the prior art, control parameters of the electric dust collector are fixedly set according to experimental and empirical settings, intelligent self-adaptive adjustment is lacking, and the change of the particulate matter concentration distribution is difficult to deal with, so that the dust collection efficiency is poor.
In view of the above problems, the application provides an adaptive parameter adjusting method for an electric precipitator based on particulate matter concentration distribution.
In a first aspect of the disclosure, an adaptive parameter adjustment method for an electric precipitator based on particulate matter concentration distribution is provided, the method comprising: collecting particle distribution information in a plurality of dust removing plate areas of the electric dust remover, wherein the particle concentration distribution information comprises particle concentration information in the plurality of dust removing plate areas; according to the particle distribution information, performing historical search to obtain nearest homogeneous particle distribution information, and obtaining a plurality of historical electric dust removal parameters adopted when the homogeneous particle distribution information is obtained, wherein each historical electric dust removal parameter comprises electric field intensity and dust removal power of a plurality of dust removal plates; outputting M historical electric dust removal parameters with highest dust removal efficiency of the historical electric dust removal parameters as a historical dust removal parameter group, wherein M is an integer greater than 1; constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, and carrying out search optimization on the electric dust removal parameters by combining historical dust removal parameter groups and particulate matter distribution information to obtain optimal electric dust removal parameters, wherein in the search optimization process, historical search optimization is carried out on the historical dust removal parameter groups; and carrying out parameter adjustment and dust removal on the electric dust remover based on the optimal electric dust removal parameters.
In another aspect of the disclosure, an adaptive parameter adjustment system for an electric precipitator based on particulate matter concentration distribution is provided, the system comprising: the information acquisition module is used for acquiring the particle distribution information in a plurality of dust removal plate areas of the electric dust remover for removing dust, wherein the particle concentration distribution information comprises the particle concentration information in the plurality of dust removal plate areas; the parameter acquisition module is used for carrying out historical search according to the particle distribution information, acquiring the closest homogeneous particle distribution information and acquiring a plurality of historical electric dust removal parameters adopted when the homogeneous particle distribution information is acquired, wherein each historical electric dust removal parameter comprises the electric field intensity and the dust removal power of a plurality of dust removal plates; the parameter group construction module is used for outputting M historical electric dust removal parameters with highest dust removal efficiency of a plurality of historical electric dust removal parameters, wherein M is an integer larger than 1 as a historical dust removal parameter group; the searching optimization module is used for constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, and searching and optimizing the electric dust removal parameters by combining historical dust removal parameter groups and particulate matter distribution information to obtain optimal electric dust removal parameters, wherein in the searching and optimizing process, historical searching and optimizing are carried out on the historical dust removal parameter groups; and the parameter adjustment module is used for carrying out parameter adjustment and dust removal on the electric dust remover based on the optimal electric dust removal parameters.
One or more technical schemes provided by the application have at least the following technical effects or advantages:
The data reflecting the internal working conditions of the electric dust collector are obtained by collecting the particle distribution information in the areas of the dust collecting plates, so that a basis is provided for parameter optimization; according to the particle distribution information, historical search is carried out to obtain nearest homogeneous particle distribution information, and a plurality of historical electric dust removal parameters adopted in the homogeneous particle distribution information are obtained so as to reduce the parameter search space and provide effective initial solutions for optimization; outputting M historical electric dust removal parameters with highest dust removal efficiency of a plurality of historical electric dust removal parameters as a historical dust removal parameter group for guiding subsequent parameter searching and optimizing; constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, and combining historical dust removal parameter groups and particulate matter distribution information, and searching and optimizing the electric dust removal parameters to obtain optimal electric dust removal parameters; based on the optimal electric dust removal parameter, parameter adjustment and dust removal are carried out on the electric dust remover, the optimal parameter obtained by optimizing is applied to the actual control of the electric dust remover, the technical scheme of self-adaptive adjustment of working conditions is realized, the technical effects that the control parameter of the electric dust remover in the prior art is fixedly set according to the setting of experiments and experiences, the intelligent self-adaptive adjustment is lacking, the change of the concentration distribution of particulate matters is difficult to deal with, the technical problem of poor dust removal efficiency is caused, and the self-adaptive optimization of the parameters of the electric dust remover according to the concentration distribution of the particulate matters is achieved so as to remove dust, thereby improving the dust removal efficiency are achieved.
The foregoing description is only an overview of the present application, and is intended to be implemented in accordance with the teachings of the present application in order that the same may be more clearly understood and to make the same and other objects, features and advantages of the present application more readily apparent.
Drawings
FIG. 1 is a schematic flow chart of an adaptive parameter adjusting method for an electric precipitator based on particulate matter concentration distribution according to an embodiment of the present application;
FIG. 2 is a schematic flow chart of obtaining a test dust collection parameter set in an adaptive parameter adjustment method of an electric precipitator based on particulate matter concentration distribution according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of an adaptive parameter adjusting system of an electric precipitator based on concentration distribution of particulate matters according to an embodiment of the present application.
Reference numerals illustrate: the system comprises an information acquisition module 11, a parameter acquisition module 12, a parameter group construction module 13, a search optimization module 14 and a parameter adjustment module 15.
Detailed Description
The technical scheme provided by the application has the following overall thought:
The embodiment of the application provides an adaptive parameter adjusting method of an electric dust collector based on particulate matter concentration distribution. Firstly, acquiring particle distribution information in a plurality of dust collecting plate areas of an electric dust collector to acquire particle distribution data reflecting the current working condition. Then, historical searching is carried out by utilizing the particle distribution information, a historical working condition closest to the current working condition is found out, and the excellent electric dust collection parameter group corresponding to the historical working condition is extracted and used as the historical dust collection parameter group. Then, an electric dust removal optimization function is constructed, historical dust removal parameter groups and particulate matter distribution information are introduced as constraint conditions, and the optimal electric dust removal parameters are searched and solved. And then, applying the optimized optimal electric dust removal parameter setting to the actual control of the electric dust remover, and carrying out self-adaptive adjustment on the electric dust removal parameter setting so as to improve the dust removal efficiency.
In general, the application uses the history excellent experience as a reference by sensing the distribution change of the particulate matters and combines the online parameter optimization to dynamically adjust the working state of the electric dust collector, thereby realizing the stable improvement of the dust collection efficiency under complex and changeable working conditions. Compared with the traditional fixed parameter control according to different working conditions, the method has stronger adaptability, robustness and optimizing capability, and improves the dust removal efficiency of the electric dust remover.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Embodiment one:
as shown in fig. 1, an embodiment of the present application provides an adaptive parameter adjustment method for an electric precipitator based on concentration distribution of particulate matters, the method comprising:
s1: collecting particle distribution information in a plurality of dust removing plate areas of the electric dust remover, wherein the particle concentration distribution information comprises particle concentration information in the plurality of dust removing plate areas;
Specifically, in the actual operation process of the electric dust remover, for better dust removal treatment of flue gas, the electric dust remover is provided with a plurality of electrode plates, namely a plurality of dust removal plates. Each dust removing plate has independently adjustable electric field intensity and dust removing power, and the electric field intensity and the dust removing power of each dust removing plate are adjusted to better adapt to the distribution of particulate matters in the flue gas, so that the dust removing efficiency is improved.
In order to obtain the optimal parameters of the electric dust collector, the particle distribution information in a plurality of dust collecting plate areas of the electric dust collector for dust collection is acquired. The particulate matter concentration distribution information comprises particulate matter concentration information in a plurality of dust collecting plate areas, namely the concentration condition of the particulate matters in each dust collecting plate area.
The particle distribution information is acquired in real time through particle concentration sensors arranged in the areas of each dust removing plate. The particle concentration sensor needs a plurality of according to the regional quantity of dust collecting plate, and the time interval of gathering is set up according to the cycle that obtains the optimum electric precipitation parameter to guarantee the make full use of collection data. The detailed distribution condition of the particles in the dust removing area can be obtained through the collected particle distribution information, and a data basis is provided for subsequent self-adaptive parameter adjustment.
S2: according to the particle distribution information, performing historical search to obtain nearest homogeneous particle distribution information, and obtaining a plurality of historical electric dust removal parameters adopted when the homogeneous particle distribution information is obtained, wherein each historical electric dust removal parameter comprises electric field intensity and dust removal power of a plurality of dust removal plates;
Specifically, after the particle distribution information in a plurality of dust collecting plate areas where the current electric dust collector collects dust is obtained, searching is carried out in historical operation data of the electric dust collector, and the same family particle distribution information closest to the current particle distribution information is found out. The homogeneous particulate matter distribution information refers to a historical state in which concentration distribution characteristics of particulate matters are most similar to that of the currently collected particulate matter distribution information.
In the actual searching process, historical data in the operation historical period of the electric dust collector is called, and a plurality of pieces of historical particulate matter distribution information are extracted and obtained. Wherein, each historical particulate matter distribution information comprises the electric field intensity and the setting value of dust removal power of a plurality of dust removal plates, and is the electric dust collector parameter adopted by the actual operation of the electric dust collector under the historical particulate matter distribution information. Then, a plurality of pieces of distribution deviation information are obtained by calculating the distribution deviation of each piece of historical particulate matter distribution information and the current particulate matter distribution information, namely calculating the particulate matter concentration differences in a plurality of dust removing plate areas and carrying out variance calculation on the particulate matter concentration differences. Wherein, the smaller the distribution deviation, the closer the two particulate matter distribution information is. And then, sorting the plurality of distribution deviation information according to the deviation from small to large, and outputting historical particulate matter distribution information with the distribution deviation of the first 20% as particulate matter distribution information of the same family as the current particulate matter distribution to obtain a plurality of historical electric dust removal parameters.
The parameters adopted by the actual operation of the electric dust collector under the same family of particle distribution information closest to the current particle distribution information are obtained by obtaining a plurality of historical electric dust collection parameters, so that the electric dust collector has a good reference value for dust collection parameter optimization of the current particle distribution information and is used for guiding the subsequent parameter optimization process.
S3: outputting M historical electric dust removal parameters with highest dust removal efficiency of the historical electric dust removal parameters as a historical dust removal parameter group, wherein M is an integer greater than 1;
Specifically, after a plurality of historical electric dust removal parameters which are the same as the current particulate matter distribution information are obtained, a part of historical electric dust removal parameters with highest dust removal efficiency are screened out from the historical electric dust removal parameters to serve as a historical dust removal parameter group, and the historical electric dust removal parameters are key reference objects in the follow-up optimization process. The historical electric dust removal parameters selected here are not only the historical particulate matter distribution states most similar to the current particulate matter distribution information, but also the optimal dust removal efficiency which can be achieved by the electric dust remover under the historical particulate matter distribution states. The historical dust removal parameter group is used as the reference for optimization, so that the parameter searching process can be effectively guided and inspired, and the optimizing efficiency and effect are improved.
In order to select a historical dust removal parameter group, historical dust removal efficiency information corresponding to each historical electric dust removal parameter is firstly obtained. The historical dust removal efficiency information is obtained by calling historical operation data of the electric dust remover. After the dust removal efficiency of all the historical electric dust removal parameters is obtained, sequencing is carried out according to the dust removal efficiency, and M historical electric dust removal parameters with highest dust removal efficiency are output to serve as the optimized first-generation historical population and are key reference objects in the subsequent optimization process. Wherein M is the number of groups of the selected historical electric dust removal parameters, is an integer greater than 1, is set by an expert group according to actual conditions, and needs to balance between the richness of the reference information and the optimization efficiency.
And defining the output M historical electric dust removal parameters with highest dust removal efficiency as a historical dust removal parameter group, wherein the historical electric dust removal parameter group represents the optimal parameter setting of the electric dust remover operation under the historical working condition which is most similar to the current particulate matter distribution information. Based on the historical dust removal parameter group, the optimal electric dust removal parameter is efficiently found in the parameter searching and optimizing process, and the self-adaptive optimization control of electric dust removal operation is realized.
S4: constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, and carrying out search optimization on the electric dust removal parameters by combining the historical dust removal parameter group and the particulate matter distribution information to obtain optimal electric dust removal parameters, wherein in the search optimization process, historical search optimization is carried out on the historical dust removal parameter group;
specifically, after the historical dust removal parameter group is obtained, the optimal electric dust removal parameters are obtained by further optimizing and searching by combining the current particle distribution information on the basis of the historical dust removal parameter group.
In order to obtain the optimal electric dust collection parameters, an electric dust collection function is firstly constructed and used for evaluating and guiding the self-adaptive optimization process of the electric dust collection parameters, the function comprehensively considers the particle concentration distribution characteristics of all dust collection areas and the electric dust collection operation parameters, and the dust collection effect of the electric dust collection parameters in the optimization process is evaluated.
After the electric dust removal function is constructed, the electric dust removal function can be combined with the historical dust removal parameter group and the particle distribution information to perform optimized search on the electric dust removal parameter. In the searching and optimizing process, a new candidate electric dust collection parameter is generated nearby based on and referenced by the historical dust collection parameter group, the new candidate electric dust collection parameter is substituted into an electric dust collection function to evaluate the dust collection effect, and the parameter searching direction is continuously and preferentially updated and improved according to the evaluation result, so that the optimal electric dust collection parameter under the current particulate matter distribution information is obtained.
In the process of optimizing searching, the advantage information of the historical dust removal parameter group is utilized and updated in a key way. On one hand, the historical dust removal parameter group is used as an excellent parameter under similar particle distribution information, and a preferential direction is provided for generation of new parameters; on the other hand, with the progress of search optimization, some candidate electric dust removal parameters with excellent performances are preferentially added into the historical dust removal parameter group, so that the parameters can be continuously perfected and evolved, and better reference information and direction guide are provided for parameter search optimization. Through continuous accumulation and optimization of the historical dust removal parameter groups, the quality of the historical dust removal parameter groups can be effectively improved, and therefore the efficiency and effect of adaptive optimization of the electric dust removal parameters are improved.
By constructing the electric dust removal function, combining the electric dust removal function with historical experience and current information, searching the optimal electric dust removal parameter in a parameter searching and optimizing mode, so that the self-adaptive optimization of the electric dust removal parameter is realized, the electric dust remover can always keep the optimal running state under complex and changeable working conditions, and the flue gas purification task is stably and efficiently completed.
S5: and carrying out parameter adjustment and dust removal on the electric dust remover based on the optimal electric dust removal parameters.
Specifically, after self-adaptive optimization searching, the optimal electric dust removal parameters under the working condition corresponding to the current particulate matter distribution information are obtained, and the operation control of the electric dust remover can be guided, so that the optimal dust removal effect is achieved under the current working condition.
The optimal electric dust removal parameters are sent to a control unit of the electric dust remover, such as a programmable logic controller or an industrial control computer. The control unit correspondingly adjusts the running state of each dust removing plate in the electric dust remover according to the received optimal electric dust removing parameters, wherein the running state comprises the electric field intensity, the dust removing power and the like applied by each dust removing plate. Through the optimal setting of the optimal electric dust removal parameters, the dust removal efficiency of each dust removal plate can be optimal, and the overall dust removal effect of the electric dust remover is obviously improved.
After parameter adjustment is completed according to the optimal electric dust removal parameters, the electric dust remover can carry out actual dust removal under the guidance of the optimal electric dust removal parameters. When the dust-containing flue gas passes through a plurality of dust removing plate areas of the electric dust remover, dust particles in the flue gas are efficiently separated and trapped under the action of an optimized electric field. Compared with the traditional fixed parameter control, the optimal electric dust removal parameter control can greatly enhance the dust removal efficiency and the adaptability of the electric dust remover.
The working conditions of the electric dust collector are dynamically changed, namely, the distribution information of the particulate matters in the areas of the dust collecting plates is dynamically changed, and parameters such as the air quantity, the temperature, the dust concentration and the like of the flue gas can be changed along with time. In order to ensure that the electric dust collector is always in an optimal running state, the electric dust collector parameter search optimization is triggered according to real-time particle distribution, the electric dust collector parameter is subjected to self-adaptive optimization adjustment, and the electric dust collector is enabled to respond to working condition change rapidly through dynamic closed-loop control, so that the maximum dust collection efficiency is continuously and stably exerted.
Further, the embodiment of the application further comprises:
the method comprises the steps of calling historical data in an operation historical period of an electric dust collector, and extracting and obtaining a plurality of pieces of historical particulate matter distribution information;
Calculating the distribution deviation of the plurality of historical particulate matter distribution information and the particulate matter distribution information to obtain a plurality of distribution deviation information;
Outputting historical particulate matter distribution information corresponding to the smallest distribution deviation information as peer particulate matter distribution information;
And extracting electric dust removal parameters when the distribution concentration of the particles in the dust removal area is the same family of particle distribution information, and obtaining the historical electric dust removal parameters.
In a possible implementation manner, in order to find out the same family particle distribution information which is most similar to the current particle distribution information, firstly, the historical data of the electric dust collector is called, the running state and parameters of the electric dust collector in a historical period are recorded, the particle distribution information at each moment, the electric field intensity, the dust removal power and other information of a plurality of corresponding dust removal plates are included, and a plurality of historical particle distribution information is obtained. The historical period is set according to the data quantity, for example, the data of the last complete operation period (such as one month or one quarter) of the electric dust collector is selected for analysis. After the historical particulate matter distribution information is obtained, the historical particulate matter distribution information and the currently collected particulate matter distribution information are compared one by one, and the similarity degree is calculated. Wherein, the distribution deviation is used as an index for measuring the similarity. The smaller the distribution deviation is, the closer the two pieces of particle distribution information are, and the more similar the represented electric dust removal working conditions are; the distribution deviation can be calculated by various methods, such as Euclidean distance, manhattan distance, cosine similarity, etc. And obtaining a plurality of historical particulate matter distribution information by calculating the distribution deviation of the currently acquired particulate matter distribution information and each particulate matter distribution information, and providing a quantitative basis for screening the homogeneous particulate matter distribution information.
And then judging which historical particulate matter distribution information is closest to the current particulate matter distribution information according to the deviation sizes of the plurality of pieces of distribution deviation information. Finding out the smallest distribution deviation from the plurality of distribution deviation information, and taking the corresponding historical particulate matter distribution information as particulate matter distribution information of the same family as the current particulate matter distribution information to obtain the same family particulate matter distribution information. The minimum distribution deviation means that the historical particulate matter distribution information is most similar to the current particulate matter distribution information in the spatial distribution of the particulate matter concentration, and the corresponding electric dust removal working condition is also the closest. After the same family of particle distribution information is determined, the historical data in the operation historical period of the electric dust collector is searched again, the historical records corresponding to the same family of particle distribution information are found out, and the parameter values such as the electric field intensity, the corona current and the like adopted in the historical records are extracted to serve as a plurality of historical electric dust collection parameters.
And through quantitative analysis of distribution deviation, homogeneous particulate matter distribution information which is most similar to the current working condition is quickly found out, and electric dust removal parameters matched with the homogeneous particulate matter distribution information are further extracted, so that a plurality of historical electric dust removal parameters are obtained, priori knowledge and reference basis are provided for subsequent self-adaptive optimization control, and the optimization efficiency and reliability are improved.
Further, the embodiment of the application further comprises:
acquiring dust removal efficiency of the plurality of historical electric dust removal parameters for removing dust from the same family of particulate matter distribution information, and acquiring a plurality of historical dust removal efficiency information;
outputting M historical electric dust removal parameters corresponding to the M pieces of maximum historical dust removal efficiency information as a historical dust removal parameter group.
In one possible embodiment, after obtaining a plurality of historical electric dust removal parameters corresponding to the same family of particulate matter distribution information, the actual dust removal effect of these historical electric dust removal parameters is further evaluated in order to screen out relatively better electric dust removal parameters. The dust removal efficiency is used as an evaluation index to reflect the dust removal performance of the electric dust remover under a given working condition. For each historical electric dust collection parameter in a plurality of historical electric dust collection parameters, extracting inlet and outlet particulate matter concentration data corresponding to the historical electric dust collection parameters from historical data in an operation historical period of the electric dust collector, subtracting the outlet particulate matter concentration from the inlet particulate matter concentration, dividing the inlet particulate matter concentration by the inlet particulate matter concentration, multiplying the inlet particulate matter concentration by 100%, and obtaining corresponding historical dust collection efficiency information, thereby obtaining a plurality of historical dust collection efficiency information.
According to the obtained historical dust removal efficiency information, optimizing the historical electric dust removal parameters, and selecting M historical electric dust removal parameters with good dust removal efficiency to form a historical dust removal parameter group which is used as a key reference object for optimizing search. The specific value of M is flexibly set according to actual needs and calculation conditions; the larger the M value is, the more abundant the priori knowledge is contained in the historical dust removal parameter group, and the more obvious the guiding effect on optimization is; however, too large a value of M increases the amount of computation and the time overhead.
The method has the advantages that the quantitative evaluation and optimization are carried out on the dust removal efficiency of a plurality of historical electric dust removal parameters, the parameters with excellent dust removal performance under similar working conditions are extracted and used as historical dust removal parameter groups, a high-value priori knowledge group is formed, the pertinence and the effectiveness of subsequent optimized search can be improved based on the historical dust removal parameter groups and guided, the iteration convergence is accelerated, the optimization effect is improved, and therefore the rapid self-adaptive optimization control of the electric dust removal parameters is realized.
Further, the embodiment of the application further comprises:
constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, wherein the electric dust removal function comprises the following formula:
，
wherein DRE is dust removal fitness, N is the number of a plurality of dust removal plate areas, The weight of the ith dust collecting plate area distributed according to the particle concentration information in the plurality of dust collecting plate areas in the particle concentration distribution information is positively correlated with the size of the particle concentration information,Is the dust removal efficiency of the ith dust removal plate area;
and according to the electric dust removal function, combining the historical dust removal parameter group and the particulate matter distribution information, performing historical search optimization on the electric dust removal parameter to obtain the optimal electric dust removal parameter.
In one possible embodiment, the construction of the electric dust removal function that adaptively optimizes the electric dust removal parameters is:
，
wherein DRE is the dust removal fitness and represents the integral dust removal effect of the electric dust remover; n is the number of the plurality of dust removing plate areas; the weight of the ith dust collecting plate area distributed according to the particle concentration information in the plurality of dust collecting plate areas in the particle concentration distribution information is positively correlated with the size of the particle concentration information, namely, the higher the particle concentration is, the larger the weight is; Is the dust removal efficiency of the ith dust removal plate area. In the electric dust removal function, the dust removal fitness comprehensively considers the dust removal efficiency of each dust removal plate area and the corresponding particle concentration weight, and comprehensively reflects the dust removal performance of the electric dust remover in different areas and the influence degree of the electric dust remover on the whole dust removal effect.
After the electric dust removal function is constructed, the electric dust removal function is used as an optimization target, and the electric dust removal parameter is adaptively optimized and searched by combining the historical dust removal parameter group and the particle distribution information. Taking a plurality of historical electric dust removal parameters in the historical dust removal parameter group as initial solutions, and carrying out local search nearby the initial solutions to find new solutions capable of improving dust removal fitness. In the searching process, the distribution characteristics of the dust removal parameters are used for adaptively adjusting the searching direction and the step length, so that the searching is more efficient. For example, a key search is performed toward the direction of the history optimum parameters, or a detailed search is performed in a region where the history parameter distribution is concentrated. Meanwhile, in the searching process, the current particle distribution information is utilized to dynamically optimize the searching strategy. For example, according to the change trend of the concentration of the particulate matters, the weight coefficient of each dust removing plate area is adjusted, so that the searching is focused on the key area; or according to the uniformity of the particle distribution, the searching step length and the searching range are adjusted in a self-adaptive mode, and the local fine optimization is considered while global exploration is ensured.
Through a history search optimization strategy, under the evaluation of the electric precipitation function, the optimal electric precipitation parameters are efficiently searched and approximated. Compared with the traditional large-area search, the method has the advantages that the historical experience and the real-time information are fully utilized, the optimization efficiency and the effect are improved, and the adaptability to the change of the working condition is enhanced.
Further, the embodiment of the application further comprises:
Acquiring parameter intervals of electric field intensity and dust removal power, and randomly generating the electric field intensity and the dust removal power of a plurality of dust removal plates to obtain an initial dust removal parameter group, wherein the initial dust removal parameter group comprises M initial electric dust removal parameters;
randomly constructing a mapping relation between the initial dust removal parameter group and electric dust removal parameters in the historical dust removal parameter group;
taking M historical electric dust removal parameters in the historical dust removal parameter group as a search direction, and performing historical search on the M initial electric dust removal parameters in the initial dust removal parameter group according to a search step length to obtain a search dust removal parameter group, wherein the search dust removal parameter group and the historical dust removal parameter group also have a mapping relation;
Adopting the searching dust removal parameter group and the historical dust removal parameter group, and carrying out random cross updating according to a mapping relation to obtain a test dust removal parameter group;
According to a plurality of test electric dust removal parameters in the test dust removal parameter group, combining the particulate matter distribution information, analyzing to obtain a plurality of test dust removal efficiency sets, and according to the electric dust removal function, calculating to obtain a plurality of test dust removal fitness;
Updating the initial dust removal parameter group by adopting the test dust removal parameter group according to the test dust removal fitness, and using the initial dust removal parameter group as a history dust removal parameter group optimized by a new history search;
and continuing to optimize until the convergence times are reached, outputting the electric dust removal parameter with the largest dust removal fitness in the optimization process, and obtaining the optimal electric dust removal parameter.
In a preferred embodiment, another method of history search optimization of the electrical dust removal parameters is provided. Firstly, before starting historical search optimization, determining the value range of electric field strength and dust removal power according to the design characteristics and operation conditions of the electric dust remover, and obtaining the interval range of two parameters. Then, in the two intervals, an electric field intensity value and a dust removal power value are randomly generated for each dust removal plate, and the electric field intensity value and the dust removal power value are combined to form the operation parameters of the dust removal plate. And combining random parameters of all the dust removal plates to obtain a complete electric dust removal parameter, namely an initial electric dust removal parameter. For the continuity of the subsequent iteration, a plurality of different initial electric dust removal parameters are required to be generated, M initial electric dust removal parameters are obtained, and an initial dust removal parameter group is formed. The number of M initial electric dust removal parameters is set according to the number of the historical electric dust removal parameters in the historical dust removal parameter group, so that the size of the initial dust removal parameter group is consistent with that of the historical dust removal parameter group.
After the initial dust removal parameter group and the historical dust removal parameter group are obtained, in order to perform effective optimization search between the two parameter groups, a mapping relation is established between the electric dust removal parameters in the two parameter groups. And traversing the initial dust removal parameter group, wherein in the traversing process, one initial electric dust removal parameter is selected as a current processing object each time, and a mapping relation is established between one historical electric dust removal parameter selected from the historical dust removal parameter group at random according to the current selected initial electric dust removal parameter. And repeatedly executing the mapping process for M times until each initial electric dust removal parameter in the initial dust removal parameter group and one historical electric dust removal parameter in the historical dust removal parameter group establish a mapping relation. In the mapping process, each historical electric dust collection parameter can only be mapped once and cannot be reused, and the mapped historical electric dust collection parameters are eliminated when randomly selected. Obtaining a one-to-one mapping relation of parameters from an initial dust removal parameter group to a historical dust removal parameter group through random mapping, and ensuring that each initial electric dust removal parameter in the initial parameter group can be associated with the historical electric dust removal parameter without omission; meanwhile, randomness and diversity of the optimization process are increased, subsequent cross optimization operation is facilitated, and parameters mapped with each other are conveniently matched and combined for optimization.
After the mapping relation between the initial dust removal parameter group and the historical dust removal parameter group is established, the historical electric dust removal parameter is used as a guide, and the initial dust removal parameter is searched and optimized to generate a new searching dust removal parameter group. Specifically, for each initial electric dust removal parameter in the initial dust removal parameter group, determining a corresponding historical electric dust removal parameter in the mapping relation, and carrying out local search in a surrounding parameter space by taking the historical electric dust removal parameter as a search center and a certain search step length. The size of the search step is adaptively adjusted according to the characteristics of the optimization problem, the convergence speed requirement and other factors, so that the search step is large enough to jump out of local optimum, and the optimal solution cannot be missed due to the fact that the search step is too large. The preferred search strategy is to gradually reduce the search step size with increasing iteration number, search a larger range in the early stage and perform fine local search in the later stage. And performing historical search on each initial dust removal parameter in the initial dust removal parameter group once to obtain a search dust removal parameter group with the same size as the initial dust removal parameter group. Because each searching dust removing parameter is obtained by searching from the initial dust removing parameter, the searching dust removing parameter group and the historical dust removing parameter group naturally have a mapping relation.
And then, according to the mapping relation between the searching dust removing parameter group and the historical dust removing parameter group, pairing each searching dust removing parameter with the corresponding historical dust removing parameter to form M pairs of parameter combinations. For each pair of parameter combinations, the historical dust removal parameters and the search dust removal parameters are subjected to parameter random cross, for example, the electric field intensity and the dust removal power of a plurality of dust removal plates are randomly selected from the historical dust removal parameters, and the electric field intensity and the dust removal power of the corresponding dust removal plates in the corresponding mapped search dust removal parameters are replaced. And after the replacement is finished, searching all the dust removal parameters to form a test dust removal parameter group which comprises a plurality of test electric dust removal parameters, wherein the number of the test electric dust removal parameters is M. And then, applying each test electric dust removal parameter in the test dust removal parameter group to simulation of the electric dust removal process, analyzing and calculating the dust removal efficiency of each dust removal area under the current particulate matter distribution information to obtain a test dust removal efficiency set, and substituting the dust removal efficiency in a plurality of dust removal plate areas in the test dust removal efficiency set into an electric dust removal function to obtain the test dust removal fitness. And obtaining test dust removal fitness values corresponding to each test electric dust removal parameter through calculation of the electric dust removal function, and obtaining a plurality of test dust removal fitness values.
Thereafter, for each pair of mapped initial and test electric dust removal parameters, their corresponding dust removal fitness is compared. If the test dust removal fitness of the test electric dust removal parameter is higher than the initial electric dust removal parameter, replacing the initial electric dust removal parameter with the test electric dust removal parameter; otherwise, the original initial electric dust removal parameters are reserved. Through preferential updating based on the fitness, the electric dust removal parameters with poor dust removal performance are effectively eliminated, and meanwhile, high-quality electric dust removal parameters are reserved and accumulated, so that the dust removal parameter group is continuously evolved and improved in the process of optimizing and searching. After the updating is completed, the original initial dust removal parameter group is converted into a new electric dust removal parameter group with better performance, and the new electric dust removal parameter group is used as a target of the next historical search optimization, namely the new historical dust removal parameter group.
Continuing to search and optimize the electric dust removal parameters, and when a new round of optimization is carried out, randomly distributing electric field strength and dust removal power parameters for each dust removal plate in a parameter interval of the electric field strength and the dust removal power by adopting a random generation mode again to obtain M new initial electric dust removal parameters; and performing historical search optimization on the new initial electric dust removal parameter group by taking the new historical dust removal parameter group as a search direction, and searching electric dust removal parameters with better performance. In the whole optimization process, the electric dust removal parameters of each round of iteration and the corresponding dust removal fitness are recorded and stored. When the search optimization of the electric dust removal parameters reaches the preset convergence times, selecting the electric dust removal parameter with the highest dust removal fitness from all electric dust removal parameters generated in the optimization process as the optimal electric dust removal parameter obtained by the optimization search.
Further, as shown in fig. 2, the embodiment of the present application further includes:
according to the size of the particle concentration information of the plurality of dust collecting plate areas in the particle distribution information, distributing and calculating to obtain a plurality of cross update probabilities, wherein the sum of the cross update probabilities is 1;
According to the multiple cross update probabilities and the mapping relation, randomly selecting electric field intensity and dust removal power of T dust removal plates in the electric dust removal parameters in the search dust removal parameter group and the history dust removal parameter group to carry out cross update, wherein the cross update probability of the selected dust removal plate area is reduced, the cross update probability of the unselected dust removal plate area is improved, and T is an integer greater than 1 and less than N;
and obtaining a test dust removal parameter group based on the cross updating result.
In a preferred embodiment, another method of random crossing of a search parameter set and a history parameter set is provided.
When random cross updating is carried out, in order to more pointedly optimize the dust removal performance, the cross updating probability of different areas is adaptively adjusted according to the concentration distribution condition of the particles in each dust removal plate area. Firstly, extracting the particle concentration values of all the dust removing plate areas in the particle distribution information, and distributing the cross update probability of all the dust removing plate areas based on the particle concentration values, so that the sum of the cross update probabilities of all the areas is 1, and the areas with higher particle concentration are endowed with larger cross update probability to reflect the important influence on the whole dust removing performance. For example, for each dust-collecting plate region, the ratio of the concentration of the particulate matter thereof to the sum of the concentrations of the particulate matter of all the dust-collecting plate regions is calculated as the cross update probability of that dust-collecting plate region, thereby obtaining a plurality of cross update probabilities.
After the cross update probability of each dust removing plate area is determined, for each pair of mapped searching dust removing parameters and historical dust removing parameters, whether the electric field intensity and the dust removing power of the area are subjected to cross update is randomly determined according to the cross update probability of each area. For example, by the roulette selection method, whether each area participates in the crossover is randomly determined according to the size of the crossover update probability. The whole process is repeated for T times, and each time, one area is randomly selected for parameter crossover until crossover updating of T areas is completed. Wherein T is a positive integer less than the total dust removal plate number N, representing the number of areas involved in cross updating of each pair of dust removal parameters. In order to further increase the randomness and diversity of the cross operation, after each round of cross updating is completed, the cross updating probability of each dust removing plate area is properly adjusted. For the dust removing plate area selected for crossing in the round, the crossing update probability is reduced according to a certain proportion (such as 0.1 or 0.05); and the cross probability of the unselected dust removing plate area is correspondingly improved. The adjusted crossover update probability will be used for random selection for the next round of crossover operation. The participation opportunities of all the areas can be balanced to a certain extent by the dynamic probability adjustment method, the long-term dominant crossing process of individual areas is avoided, and the expansion of the search breadth and depth is facilitated. It should be noted that, when probability adjustment is performed, the sum of the cross update probabilities of the respective dust-collecting plate regions is always kept at 1 to ensure the validity of the probability distribution. After the cross updating of all the dust removing parameter pairs is completed, a group of new electric dust removing parameters can be obtained, and part of characteristics of the searching dust removing parameters and the historical dust removing parameters are inherited. And summarizing all the cross updated electric dust removal parameters to obtain a test dust removal parameter group.
By introducing cross update probability and dynamic probability adjustment mechanism, pertinence and flexibility of cross operation are further enhanced, non-uniformity of concentration distribution of particulate matters is fully considered, key dust removal areas are adaptively focused and optimized through preferential resource allocation and sustainable strategy adjustment, and powerful support is provided for obtaining optimal electric dust removal parameters.
Further, the embodiment of the application further comprises:
Acquiring a sample particulate matter concentration information set and a sample electric dust collection parameter set according to electric dust collection historical data, and acquiring a sample dust collection efficiency set;
constructing a dust removal efficiency predictor by adopting the sample particulate matter concentration information set, the sample electric dust removal parameter set and the sample dust removal efficiency set;
based on the dust removal efficiency predictor, a plurality of test electric dust removal parameters in the test dust removal parameter group are input in combination with the particulate matter distribution information, a plurality of test dust removal efficiency sets are obtained through prediction, and a plurality of test dust removal fitness is obtained through calculation.
In a possible implementation manner, when the adaptability evaluation is performed on the experimental electric dust removal parameters, besides performing simulation calculation on the dust removal efficiency through a simulation model, the electric dust removal historical data can be used for constructing a dust removal efficiency predictor for rapidly estimating the dust removal efficiency of the experimental electric dust removal parameters.
Firstly, sample data are extracted from electric dust removal historical data, wherein the sample data comprise particulate matter concentration distribution information, electric dust removal parameter settings and corresponding actual dust removal efficiency at each historical moment. And respectively summarizing the sample data to obtain a sample particulate matter concentration information set, a sample electric dust removal parameter set and a sample dust removal efficiency set. Then, based on the obtained sample particulate matter concentration information set, sample electric dust collection parameter set, and sample dust collection efficiency set, a dust collection efficiency predictor is trained for estimating dust collection efficiency given the test electric dust collection parameters and particulate matter distribution information. The dust removal efficiency predictor can be constructed by adopting various machine learning modes, such as multiple regression, a neural network, a support vector machine and the like, and a mathematical model among electric dust removal parameters, particle concentration and dust removal efficiency is built by learning rules and mapping relations in sample data, so that the dust removal efficiency prediction of new experimental electric dust removal parameters and new working conditions is realized. When the dust removal efficiency predictor is trained, sample particulate matter concentration information and sample electric dust removal parameters are used as input features of the dust removal efficiency predictor, the sample dust removal efficiency is used as an output target of the dust removal efficiency predictor, and the dust removal efficiency predictor with high prediction accuracy is obtained by minimizing the error between the predicted dust removal efficiency and the actual dust removal efficiency, so that the parameters of the dust removal efficiency predictor are continuously optimized, the prediction performance is continuously improved.
After the test dust removal parameter group is obtained, sequentially inputting a plurality of test electric dust removal parameters and current particulate matter concentration distribution information into a dust removal efficiency predictor together, obtaining predicted dust removal efficiency of the test electric dust removal parameters in each dust removal area, and forming a plurality of test dust removal efficiency sets. And then substituting the test dust removal efficiency sets into the electric dust removal function in sequence, comprehensively considering the weight influence of each dust removal area, and calculating the dust removal fitness degree of each test electric dust removal parameter to obtain a plurality of test dust removal fitness degrees.
By the dust removal efficiency obtaining method of the dust removal efficiency predictor, massive historical operation data in the electric dust removal process are fully utilized, association rules and mapping relations contained in the data are mined and summarized through a machine learning algorithm, and the efficiency and the speed of test electric dust removal parameter adaptability assessment are remarkably improved.
In summary, the self-adaptive parameter adjusting method for the electric precipitator based on the concentration distribution of the particulate matters provided by the embodiment of the application has the following technical effects:
And collecting the particle distribution information in a plurality of dust removal plate areas of the electric dust remover for removing dust, acquiring key data reflecting the internal actual working condition of the electric dust remover, and providing a basis for parameter self-adaptive adjustment. According to the particle distribution information, historical search is carried out to obtain nearest homogeneous particle distribution information, and a plurality of historical electric dust removal parameters adopted when homogeneous particle distribution information is obtained, so that the past experience is fully consulted, the parameter search space is reduced, an effective initial solution is provided for optimization, and the convergence process is accelerated; and outputting M historical electric dust collection parameters with highest dust collection efficiency of the historical electric dust collection parameters as a historical dust collection parameter group, indicating a general direction for parameter optimization of the current working condition, and improving optimizing quality and efficiency. An electric dust removing function which adaptively optimizes electric dust removing parameters is constructed, historical dust removing parameter groups and particle distribution information are combined, electric dust removing parameters are searched and optimized, optimal electric dust removing parameters are obtained, the obtained particle distribution information and the historical dust removing parameter groups are included in constraint conditions, a multi-target and multi-constraint parameter optimizing model is formed, the optimal electric dust removing parameters are searched, and the aim of working condition adaptation is achieved. Meanwhile, the history parameter group is utilized to guide the optimized search, so that the optimizing performance is further improved. Based on the optimal electric dust removal parameters, parameter adjustment and dust removal are carried out on the electric dust remover, the optimal electric dust removal parameters obtained through optimization are practically applied to the control of the electric dust remover, and the working state of the electric dust remover is dynamically adjusted, so that the electric dust remover is suitable for the current flue gas working condition, the optimal dust removal efficiency is achieved, and the optimal dust removal effect is achieved.
Embodiment two:
Based on the same inventive concept as the adaptive parameter adjustment method of the electric precipitator based on the concentration distribution of the particulate matters in the foregoing embodiment, as shown in fig. 3, an embodiment of the present application provides an adaptive parameter adjustment system of the electric precipitator based on the concentration distribution of the particulate matters, which includes:
The information acquisition module 11 is used for acquiring the particle distribution information in a plurality of dust removing plate areas where the electric dust remover removes dust, wherein the particle concentration distribution information comprises the particle concentration information in the plurality of dust removing plate areas;
The parameter obtaining module 12 is configured to perform a historical search according to the particulate matter distribution information, obtain closest homogeneous particulate matter distribution information, and obtain a plurality of historical electric dust removal parameters used when the homogeneous particulate matter distribution information is obtained, where each historical electric dust removal parameter includes electric field intensities and dust removal powers of a plurality of dust removal plates;
the parameter group construction module 13 is configured to output M historical electric dust removal parameters with highest dust removal efficiencies of the plurality of historical electric dust removal parameters, where M is an integer greater than 1 as a historical dust removal parameter group;
the search optimization module 14 is configured to construct an electric dust removal function that adaptively optimizes electric dust removal parameters, and perform search optimization on the electric dust removal parameters in combination with the historical dust removal parameter group and the particulate matter distribution information to obtain optimal electric dust removal parameters, where in the search optimization process, historical search optimization is performed on the historical dust removal parameter group;
and the parameter adjustment module 15 is used for carrying out parameter adjustment and dust removal on the electric dust remover based on the optimal electric dust removal parameters.
Further, the parameter obtaining module 12 includes the following steps:
the method comprises the steps of calling historical data in an operation historical period of an electric dust collector, and extracting and obtaining a plurality of pieces of historical particulate matter distribution information;
Calculating the distribution deviation of the plurality of historical particulate matter distribution information and the particulate matter distribution information to obtain a plurality of distribution deviation information;
Outputting historical particulate matter distribution information corresponding to the smallest distribution deviation information as peer particulate matter distribution information;
And extracting electric dust removal parameters when the distribution concentration of the particles in the dust removal area is the same family of particle distribution information, and obtaining the historical electric dust removal parameters.
Further, the parameter group construction module 13 includes the following steps:
acquiring dust removal efficiency of the plurality of historical electric dust removal parameters for removing dust from the same family of particulate matter distribution information, and acquiring a plurality of historical dust removal efficiency information;
outputting M historical electric dust removal parameters corresponding to the M pieces of maximum historical dust removal efficiency information as a historical dust removal parameter group.
Further, the search optimization module 14 includes the following execution steps:
constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, wherein the electric dust removal function comprises the following formula:
，
wherein DRE is dust removal fitness, N is the number of a plurality of dust removal plate areas, The weight of the ith dust collecting plate area distributed according to the particle concentration information in the plurality of dust collecting plate areas in the particle concentration distribution information is positively correlated with the size of the particle concentration information,Is the dust removal efficiency of the ith dust removal plate area;
and according to the electric dust removal function, combining the historical dust removal parameter group and the particulate matter distribution information, performing historical search optimization on the electric dust removal parameter to obtain the optimal electric dust removal parameter.
Further, the search optimization module 14 further includes the following execution steps:
Acquiring parameter intervals of electric field intensity and dust removal power, and randomly generating the electric field intensity and the dust removal power of a plurality of dust removal plates to obtain an initial dust removal parameter group, wherein the initial dust removal parameter group comprises M initial electric dust removal parameters;
randomly constructing a mapping relation between the initial dust removal parameter group and electric dust removal parameters in the historical dust removal parameter group;
taking M historical electric dust removal parameters in the historical dust removal parameter group as a search direction, and performing historical search on the M initial electric dust removal parameters in the initial dust removal parameter group according to a search step length to obtain a search dust removal parameter group, wherein the search dust removal parameter group and the historical dust removal parameter group also have a mapping relation;
Adopting the searching dust removal parameter group and the historical dust removal parameter group, and carrying out random cross updating according to a mapping relation to obtain a test dust removal parameter group;
According to a plurality of test electric dust removal parameters in the test dust removal parameter group, combining the particulate matter distribution information, analyzing to obtain a plurality of test dust removal efficiency sets, and according to the electric dust removal function, calculating to obtain a plurality of test dust removal fitness;
Updating the initial dust removal parameter group by adopting the test dust removal parameter group according to the test dust removal fitness, and using the initial dust removal parameter group as a history dust removal parameter group optimized by a new history search;
and continuing to optimize until the convergence times are reached, outputting the electric dust removal parameter with the largest dust removal fitness in the optimization process, and obtaining the optimal electric dust removal parameter.
Further, the search optimization module 14 further includes the following execution steps:
according to the size of the particle concentration information of the plurality of dust collecting plate areas in the particle distribution information, distributing and calculating to obtain a plurality of cross update probabilities, wherein the sum of the cross update probabilities is 1;
According to the multiple cross update probabilities and the mapping relation, randomly selecting electric field intensity and dust removal power of T dust removal plates in the electric dust removal parameters in the search dust removal parameter group and the history dust removal parameter group to carry out cross update, wherein the cross update probability of the selected dust removal plate area is reduced, the cross update probability of the unselected dust removal plate area is improved, and T is an integer greater than 1 and less than N;
and obtaining a test dust removal parameter group based on the cross updating result.
Further, the search optimization module 14 further includes the following execution steps:
Acquiring a sample particulate matter concentration information set and a sample electric dust collection parameter set according to electric dust collection historical data, and acquiring a sample dust collection efficiency set;
constructing a dust removal efficiency predictor by adopting the sample particulate matter concentration information set, the sample electric dust removal parameter set and the sample dust removal efficiency set;
based on the dust removal efficiency predictor, a plurality of test electric dust removal parameters in the test dust removal parameter group are input in combination with the particulate matter distribution information, a plurality of test dust removal efficiency sets are obtained through prediction, and a plurality of test dust removal fitness is obtained through calculation.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any method for implementing an embodiment of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the scope of the application. Thus, the present application is intended to include such modifications and alterations insofar as they come within the scope of the application or the equivalents thereof.
Claims (4)
1. The self-adaptive parameter adjusting method of the electric dust collector based on the concentration distribution of the particulate matters is characterized by comprising the following steps of:
Collecting particle distribution information in a plurality of dust removing plate areas of the electric dust remover, wherein the particle concentration distribution information comprises particle concentration information in the plurality of dust removing plate areas;
According to the particle distribution information, performing historical search to obtain nearest homogeneous particle distribution information, and obtaining a plurality of historical electric dust removal parameters adopted when the homogeneous particle distribution information is obtained, wherein each historical electric dust removal parameter comprises electric field intensity and dust removal power of a plurality of dust removal plates;
outputting M historical electric dust removal parameters with highest dust removal efficiency of the historical electric dust removal parameters as a historical dust removal parameter group, wherein M is an integer greater than 1;
constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, and carrying out search optimization on the electric dust removal parameters by combining the historical dust removal parameter group and the particulate matter distribution information to obtain optimal electric dust removal parameters, wherein in the search optimization process, historical search optimization is carried out on the historical dust removal parameter group;
Based on the optimal electric dust removal parameters, carrying out parameter adjustment and dust removal on the electric dust remover;
the method comprises the steps of constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, searching and optimizing the electric dust removal parameters by combining the historical dust removal parameter group and the particulate matter distribution information to obtain optimal electric dust removal parameters, and comprises the following steps:
constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, wherein the electric dust removal function comprises the following formula:
，
wherein DRE is dust removal fitness, N is the number of a plurality of dust removal plate areas, The weight of the ith dust collecting plate area distributed according to the particle concentration information in the plurality of dust collecting plate areas in the particle concentration distribution information is positively correlated with the size of the particle concentration information,Is the dust removal efficiency of the ith dust removal plate area;
According to the electric dust removal function, combining the historical dust removal parameter group and the particulate matter distribution information, performing historical search optimization on the electric dust removal parameter to obtain an optimal electric dust removal parameter;
Wherein, according to the electric precipitation function, combine the historical dust removal parameter group and particulate matter distribution information, carry out historical search optimization to electric precipitation parameter, include:
Acquiring parameter intervals of electric field intensity and dust removal power, and randomly generating the electric field intensity and the dust removal power of a plurality of dust removal plates to obtain an initial dust removal parameter group, wherein the initial dust removal parameter group comprises M initial electric dust removal parameters;
randomly constructing a mapping relation between the initial dust removal parameter group and electric dust removal parameters in the historical dust removal parameter group;
taking M historical electric dust removal parameters in the historical dust removal parameter group as a search direction, and performing historical search on the M initial electric dust removal parameters in the initial dust removal parameter group according to a search step length to obtain a search dust removal parameter group, wherein the search dust removal parameter group and the historical dust removal parameter group also have a mapping relation;
Adopting the searching dust removal parameter group and the historical dust removal parameter group, and carrying out random cross updating according to a mapping relation to obtain a test dust removal parameter group;
According to a plurality of test electric dust removal parameters in the test dust removal parameter group, combining the particulate matter distribution information, analyzing to obtain a plurality of test dust removal efficiency sets, and according to the electric dust removal function, calculating to obtain a plurality of test dust removal fitness;
Updating the initial dust removal parameter group by adopting the test dust removal parameter group according to the test dust removal fitness, and using the initial dust removal parameter group as a history dust removal parameter group optimized by a new history search;
Continuing to optimize until convergence times are reached, outputting electric dust removal parameters with the largest dust removal fitness in the optimization process, and obtaining the optimal electric dust removal parameters;
the method comprises the steps of adopting the searching dust removal parameter group and the historical dust removal parameter group, carrying out random cross updating according to a mapping relation to obtain a test dust removal parameter group, and comprising the following steps:
according to the size of the particle concentration information of the plurality of dust collecting plate areas in the particle distribution information, distributing and calculating to obtain a plurality of cross update probabilities, wherein the sum of the cross update probabilities is 1;
According to the multiple cross update probabilities and the mapping relation, randomly selecting electric field intensity and dust removal power of T dust removal plates in the electric dust removal parameters in the search dust removal parameter group and the history dust removal parameter group to carry out cross update, wherein the cross update probability of the selected dust removal plate area is reduced, the cross update probability of the unselected dust removal plate area is improved, and T is an integer greater than 1 and less than N;
Based on the result of the cross updating, obtaining a test dust removal parameter group;
wherein, according to a plurality of experimental electric precipitation parameters in the experimental dust removal parameter group, combine particulate matter distribution information, analysis obtains a plurality of experimental dust removal efficiency sets, includes:
Acquiring a sample particulate matter concentration information set and a sample electric dust collection parameter set according to electric dust collection historical data, and acquiring a sample dust collection efficiency set;
constructing a dust removal efficiency predictor by adopting the sample particulate matter concentration information set, the sample electric dust removal parameter set and the sample dust removal efficiency set;
based on the dust removal efficiency predictor, a plurality of test electric dust removal parameters in the test dust removal parameter group are input in combination with the particulate matter distribution information, a plurality of test dust removal efficiency sets are obtained through prediction, and a plurality of test dust removal fitness is obtained through calculation.
2. The method of claim 1, wherein performing a historical search based on the particulate matter distribution information, obtaining closest historical particulate matter distribution information, and obtaining a plurality of historical electrical dust removal parameters employed in the historical particulate matter distribution information, comprises:
the method comprises the steps of calling historical data in an operation historical period of an electric dust collector, and extracting and obtaining a plurality of pieces of historical particulate matter distribution information;
Calculating the distribution deviation of the plurality of historical particulate matter distribution information and the particulate matter distribution information to obtain a plurality of distribution deviation information;
Outputting historical particulate matter distribution information corresponding to the smallest distribution deviation information as peer particulate matter distribution information;
And extracting electric dust removal parameters when the distribution concentration of the particles in the dust removal area is the same family of particle distribution information, and obtaining the historical electric dust removal parameters.
3. The method according to claim 1, wherein outputting M historical electric dust removal parameters having highest dust removal efficiency of the plurality of historical electric dust removal parameters as a historical dust removal parameter group comprises:
acquiring dust removal efficiency of the plurality of historical electric dust removal parameters for removing dust from the same family of particulate matter distribution information, and acquiring a plurality of historical dust removal efficiency information;
outputting M historical electric dust removal parameters corresponding to the M pieces of maximum historical dust removal efficiency information as a historical dust removal parameter group.
4. An adaptive parameter adjustment system for an electric precipitator based on particulate matter concentration distribution, which is used for implementing the adaptive parameter adjustment method for an electric precipitator based on particulate matter concentration distribution according to any one of claims 1-3, and comprises the following steps:
The information acquisition module is used for acquiring the particle distribution information in a plurality of dust removal plate areas of the electric dust remover for removing dust, wherein the particle concentration distribution information comprises the particle concentration information in the plurality of dust removal plate areas;
The parameter acquisition module is used for carrying out historical search according to the particulate matter distribution information to acquire nearest homogeneous particulate matter distribution information and acquiring a plurality of historical electric dust removal parameters adopted when the homogeneous particulate matter distribution information is acquired, wherein each historical electric dust removal parameter comprises electric field strength and dust removal power of a plurality of dust removal plates;
The parameter group construction module is used for outputting M historical electric dust removal parameters with highest dust removal efficiency of the historical electric dust removal parameters as a historical dust removal parameter group, wherein M is an integer larger than 1;
The search optimization module is used for constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, and carrying out search optimization on the electric dust removal parameters by combining the historical dust removal parameter group and the particulate matter distribution information to obtain optimal electric dust removal parameters, wherein in the search optimization process, historical search optimization is carried out on the historical dust removal parameter group;
the parameter adjustment module is used for adjusting parameters and removing dust of the electric dust collector based on the optimal electric dust collection parameters;
the method comprises the steps of constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, searching and optimizing the electric dust removal parameters by combining the historical dust removal parameter group and the particulate matter distribution information to obtain optimal electric dust removal parameters, and comprises the following steps:
constructing an electric dust removal function which adaptively optimizes electric dust removal parameters, wherein the electric dust removal function comprises the following formula:
，
wherein DRE is dust removal fitness, N is the number of a plurality of dust removal plate areas, The weight of the ith dust collecting plate area distributed according to the particle concentration information in the plurality of dust collecting plate areas in the particle concentration distribution information is positively correlated with the size of the particle concentration information,Is the dust removal efficiency of the ith dust removal plate area;
According to the electric dust removal function, combining the historical dust removal parameter group and the particulate matter distribution information, performing historical search optimization on the electric dust removal parameter to obtain an optimal electric dust removal parameter;
Wherein, according to the electric precipitation function, combine the historical dust removal parameter group and particulate matter distribution information, carry out historical search optimization to electric precipitation parameter, include:
Acquiring parameter intervals of electric field intensity and dust removal power, and randomly generating the electric field intensity and the dust removal power of a plurality of dust removal plates to obtain an initial dust removal parameter group, wherein the initial dust removal parameter group comprises M initial electric dust removal parameters;
randomly constructing a mapping relation between the initial dust removal parameter group and electric dust removal parameters in the historical dust removal parameter group;
taking M historical electric dust removal parameters in the historical dust removal parameter group as a search direction, and performing historical search on the M initial electric dust removal parameters in the initial dust removal parameter group according to a search step length to obtain a search dust removal parameter group, wherein the search dust removal parameter group and the historical dust removal parameter group also have a mapping relation;
Adopting the searching dust removal parameter group and the historical dust removal parameter group, and carrying out random cross updating according to a mapping relation to obtain a test dust removal parameter group;
According to a plurality of test electric dust removal parameters in the test dust removal parameter group, combining the particulate matter distribution information, analyzing to obtain a plurality of test dust removal efficiency sets, and according to the electric dust removal function, calculating to obtain a plurality of test dust removal fitness;
Updating the initial dust removal parameter group by adopting the test dust removal parameter group according to the test dust removal fitness, and using the initial dust removal parameter group as a history dust removal parameter group optimized by a new history search;
Continuing to optimize until convergence times are reached, outputting electric dust removal parameters with the largest dust removal fitness in the optimization process, and obtaining the optimal electric dust removal parameters;
the method comprises the steps of adopting the searching dust removal parameter group and the historical dust removal parameter group, carrying out random cross updating according to a mapping relation to obtain a test dust removal parameter group, and comprising the following steps:
according to the size of the particle concentration information of the plurality of dust collecting plate areas in the particle distribution information, distributing and calculating to obtain a plurality of cross update probabilities, wherein the sum of the cross update probabilities is 1;
According to the multiple cross update probabilities and the mapping relation, randomly selecting electric field intensity and dust removal power of T dust removal plates in the electric dust removal parameters in the search dust removal parameter group and the history dust removal parameter group to carry out cross update, wherein the cross update probability of the selected dust removal plate area is reduced, the cross update probability of the unselected dust removal plate area is improved, and T is an integer greater than 1 and less than N;
Based on the result of the cross updating, obtaining a test dust removal parameter group;
wherein, according to a plurality of experimental electric precipitation parameters in the experimental dust removal parameter group, combine particulate matter distribution information, analysis obtains a plurality of experimental dust removal efficiency sets, includes:
Acquiring a sample particulate matter concentration information set and a sample electric dust collection parameter set according to electric dust collection historical data, and acquiring a sample dust collection efficiency set;
constructing a dust removal efficiency predictor by adopting the sample particulate matter concentration information set, the sample electric dust removal parameter set and the sample dust removal efficiency set;
based on the dust removal efficiency predictor, a plurality of test electric dust removal parameters in the test dust removal parameter group are input in combination with the particulate matter distribution information, a plurality of test dust removal efficiency sets are obtained through prediction, and a plurality of test dust removal fitness is obtained through calculation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410451496.8A CN118050995B (en) | 2024-04-16 | 2024-04-16 | Self-adaptive parameter adjusting method of electric dust collector based on concentration distribution of particulate matters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410451496.8A CN118050995B (en) | 2024-04-16 | 2024-04-16 | Self-adaptive parameter adjusting method of electric dust collector based on concentration distribution of particulate matters |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118050995A CN118050995A (en) | 2024-05-17 |
| CN118050995B true CN118050995B (en) | 2024-07-05 |
Family
ID=91046902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410451496.8A Active CN118050995B (en) | 2024-04-16 | 2024-04-16 | Self-adaptive parameter adjusting method of electric dust collector based on concentration distribution of particulate matters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118050995B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117718143A (en) * | 2023-10-31 | 2024-03-19 | 华能烟台八角热电有限公司 | Intelligent energy management system for electric precipitation |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9645558B2 (en) * | 2012-09-29 | 2017-05-09 | Operation Technology, Inc. | Dynamic parameter tuning using particle swarm optimization |
| CN112099343B (en) * | 2020-07-29 | 2022-06-17 | 福建龙净环保股份有限公司 | Intelligent energy-saving optimization method and medium for electric precipitation system based on neural network |
| CN113976323A (en) * | 2021-09-10 | 2022-01-28 | 华能曲阜热电有限公司 | Multi-signal optimization energy-saving electric precipitation control method |
| CN117150677A (en) * | 2023-09-06 | 2023-12-01 | 福建龙净环保股份有限公司 | Method and device for determining type selection parameters of electric dust collector |
-
2024
- 2024-04-16 CN CN202410451496.8A patent/CN118050995B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117718143A (en) * | 2023-10-31 | 2024-03-19 | 华能烟台八角热电有限公司 | Intelligent energy management system for electric precipitation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118050995A (en) | 2024-05-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Block-wisely supervised neural architecture search with knowledge distillation | |
| CN110930198A (en) | Electric energy substitution potential prediction method and system based on random forest, storage medium and computer equipment | |
| CN108985381A (en) | The determination method, device and equipment of nitrogen oxide emission prediction model | |
| CN108154003B (en) | Blasting vibration prediction method based on Spark gene expression optimization | |
| CN112099343A (en) | Intelligent energy-saving optimization method of electric precipitation system based on neural network | |
| CN111179108A (en) | Method and device for predicting power consumption | |
| Sahoo et al. | Automated test case generation and optimization: a comparative review | |
| CN106919504B (en) | Test data evolution generation method based on GA algorithm | |
| CN118050995B (en) | Self-adaptive parameter adjusting method of electric dust collector based on concentration distribution of particulate matters | |
| CN112200000A (en) | Welding stability recognition model training method and welding stability recognition method | |
| Chaudhary et al. | HyGADE: hybrid of genetic algorithm and differential evolution algorithm | |
| CN114509679A (en) | Battery SOH prediction model construction method based on deep learning | |
| CN111199304B (en) | Multi-target combustion optimization method based on data-driven fusion strategy | |
| CN116701919B (en) | Optimization monitoring method and system for gyro-type hydrogen fuel generator | |
| CN112466416B (en) | Material data cleaning method combining nickel-based alloy priori knowledge | |
| CN111813669B (en) | Adaptive random test case generation method based on multi-target group intelligence | |
| CN117151272A (en) | Carbon check power emission factor prediction method based on machine learning | |
| Nielsen et al. | Novel efficient asynchronous cooperative co-evolutionary multi-objective algorithms | |
| CN114139619A (en) | Boiler combustion optimization control method and device based on improved K-means algorithm | |
| Lu et al. | Dynamic multi-objective evolutionary algorithm based on decomposition for test task scheduling problem | |
| Odebode et al. | KD1: A Sampling-Based Clustering Scheme for Large Data Sets | |
| Sukthanker et al. | HW-GPT-Bench: Hardware-Aware Architecture Benchmark for Language Models | |
| CN118134210B (en) | Carbon footprint management method and system for steel production | |
| Xiao et al. | Ranking-based elitist differential evolution for many-objective optimization | |
| US20210136153A1 (en) | Server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |