CN116187150A - Air pollution source inversion method based on improved ant colony algorithm - Google Patents
Air pollution source inversion method based on improved ant colony algorithm Download PDFInfo
- Publication number
- CN116187150A CN116187150A CN202111419495.8A CN202111419495A CN116187150A CN 116187150 A CN116187150 A CN 116187150A CN 202111419495 A CN202111419495 A CN 202111419495A CN 116187150 A CN116187150 A CN 116187150A
- Authority
- CN
- China
- Prior art keywords
- ant
- air pollution
- source
- pollution source
- colony algorithm
- 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.)
- Pending
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 62
- 238000003915 air pollution Methods 0.000 title claims abstract description 61
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000009792 diffusion process Methods 0.000 claims abstract description 38
- 239000003016 pheromone Substances 0.000 claims abstract description 36
- 238000012544 monitoring process Methods 0.000 claims description 34
- 241000257303 Hymenoptera Species 0.000 claims description 13
- 238000011835 investigation Methods 0.000 claims description 13
- 239000003344 environmental pollutant Substances 0.000 claims description 11
- 231100000719 pollutant Toxicity 0.000 claims description 11
- 230000007704 transition Effects 0.000 claims description 9
- 230000001965 increasing effect Effects 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 8
- 230000003247 decreasing effect Effects 0.000 claims description 6
- 230000010076 replication Effects 0.000 claims description 4
- 238000004088 simulation Methods 0.000 claims description 4
- 239000000356 contaminant Substances 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 239000013598 vector Substances 0.000 claims description 2
- 230000002068 genetic effect Effects 0.000 abstract description 6
- 230000007246 mechanism Effects 0.000 abstract description 4
- 230000002708 enhancing effect Effects 0.000 abstract description 2
- 238000003900 soil pollution Methods 0.000 abstract description 2
- 238000003911 water pollution Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 4
- 239000000809 air pollutant Substances 0.000 description 3
- 231100001243 air pollutant Toxicity 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000011109 contamination Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000002431 foraging effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Tourism & Hospitality (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Genetics & Genomics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physiology (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Geometry (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses an air pollution source inversion method based on an improved ant colony algorithm, which mainly comprises the following steps: establishing a gas diffusion model according to a gas diffusion rule, and establishing an air pollution source inversion model based on the gas diffusion model; and secondly, solving the model by adopting an improved ant colony algorithm, improving the model from two aspects of easy sinking local extremum and convergence speed of the ant colony algorithm, and combining the crossing thought of a genetic algorithm and updating a pheromone updating mechanism of the ant colony algorithm, thereby enhancing the global searching capability of the ant colony algorithm and accelerating the convergence speed, further achieving the purpose of quickly and accurately inverting the relevant parameter information of an air pollution source, achieving the quick positioning of the pollution source and carrying out targeted treatment. Meanwhile, the air pollution source inversion method based on the improved ant colony algorithm provided by the invention can be also suitable for the source inversion problems of water pollution, soil pollution and the like.
Description
Technical Field
The invention discloses a pollution source inversion method based on an intelligent swarm algorithm, in particular to an air pollution source inversion method based on an improved ant colony algorithm.
Background
With the development of chemical industry, air pollution becomes a non-negligible problem. Only the position and the emission of the pollution source are clear, a solid foundation can be laid for further pollution treatment, and the whole air monitoring treatment chain can play roles of accurate monitoring, real-time source analysis, timely management, targeted treatment and the like.
The traditional method for positioning the air pollution sources is mainly to conduct investigation through an enterprise emission source list, but in the process of acquiring the source list and creating a database, the acquisition difficulty of the early emission list is too large and the data is not complete due to the openness of air pollutants in a monitoring area. And the source list cannot be completely collected due to the existence of some unknown enterprises. In addition, the emission factor is difficult to determine due to the incomplete profile of activity levels. However, even if a complete database can be established, emission data of all pollution sources in the monitoring area can be obtained, and the data volume of the whole pollution source list is too large due to rapid development of industries in China, so that the investigation by manpower alone is very difficult and low-efficiency.
Disclosure of Invention
In order to improve the positioning accuracy and the investigation efficiency of the air pollution sources and meet the requirements of accurate positioning and targeted treatment of the pollution sources, the invention provides an air pollution source inversion method based on an improved ant colony algorithm, which can effectively position and invert parameters of the pollution sources causing air pollution exceeding accidents and can improve the efficiency and the accuracy of the existing method.
The global optimization searching method is that an artificial intelligence optimization method is adopted, a proper gas diffusion model is selected according to factors such as gas diffusion characteristics, wind speed and wind direction, a loss function is established as a model, and the loss function is repeatedly solved by adopting an artificial intelligence algorithm, so that the position and the emission amount of an air pollution emission source are obtained, the labor cost can be reduced, and the efficiency is improved. Therefore, the research of inverting the air pollution source by the swarm intelligence algorithm has certain theoretical significance and application value.
The technical scheme adopted by the invention for achieving the purpose is as follows: an air pollution source inversion method based on an improved ant colony algorithm comprises the following steps:
step 1: collecting monitoring data of all monitoring stations with air pollution exceeding standards;
step 2: constructing and optimizing a gas diffusion model, and forming an air pollution source inversion model together with a target optimization function;
step 3: and carrying out iterative solution on the air pollution source inversion model through an ant colony algorithm to obtain pollution source information so as to determine the air pollution source.
The step 2 is specifically as follows:
and (3) constructing a gas diffusion model:
wherein, C (x, y, z) represents the gas simulation concentration of any point (x, y, z) in downwind direction; q (Q) 0 Indicating the intensity of the emission source; u represents the average wind speed; x is x 0 、y 0 Representing the horizontal plane coordinates of the emission source; x and y represent horizontal plane coordinates of the monitoring station; h 0 Representing the height of the emission source; z represents the vertical coordinate of any point (x, y, z) downwind of the discharge port; sigma (sigma) y and σz Represents the horizontal and vertical diffusion coefficients, expressed in the form:
in the formula ,ω1 、ω 2 、θ 1 、θ 2 Is a coefficient;
the gas diffusion model is optimized as follows: let z=0, the optimized gas diffusion model is as follows:
the standard loss function is constructed according to the optimized gas diffusion model as follows:
wherein ,
indicating the concentration of the contaminant actually detected by the monitoring station,/-, etc.>
Represents the concentration of the pollutant, L, obtained by the optimized gas diffusion model tar The loss function value is expressed to evaluate the accuracy of the current result, the closer it is to 0, the more accurate the current result is, and T represents the number of monitoring stations.
The step 3 is specifically as follows:
step 3.1: initializing various parameters: the importance factors alpha and beta of the pheromone and heuristic function, the ant colony scale m, the iteration number N, the pheromone volatilization factor rho, the upper limit and the lower limit of each source parameter of the emission source, wherein each source parameter of the emission source comprises emission source intensity Q 0 Position (x) 0 、y 0 ) Release height H 0 ；
Step 3.2: randomly generating ant individuals in a feasible domain formed by various source parameters, and initializing the pheromone tau of each ant i (i＝1,2,3,…,m)；
Step 3.3: each ant selects the next position according to the state transition rule and the search strategy;
step 3.4: after all ant searches are completed, updating the current global optimal solution, and then carrying out selection and exchange operation;
step 3.5: updating the ants according to the magnitude relation between the loss value of the nth generation and the loss value of the nth generation of the ith ants; when the iteration number reaches N, if the loss function L tar And converging, namely performing pollution source investigation according to each source parameter corresponding to the ant individual with the current iteration times serving as an optimal ant individual.
The step 3.3 is realized by the following steps:
step 3.3.1: the state transition probabilities are as follows:
wherein, represent heuristic factor P i (n) represents the probability that the ith ant selects the next position, when P i (n) when larger, the current solution is closer to the various source parameters of the pollution source; τ i Pheromone, eta representing the ith ant i Heuristic functions representing the ith ant; n is the current iterated times;
step 3.3.2: setting a state transition constant P 0 When P i (n) is greater than P 0 When a local search strategy is adopted, otherwise a global search strategy is adopted, which can be expressed as follows:
wherein next is the next position of the current ant; now is the current position of the current ant; rand represents a distance between (0, 1)Random numbers of (a); step represents the local search step; lambda represents the step size coefficient, which takes the value of the inverse of the number of iterations
range represents the feasible region range.
The step 3.4 is specifically as follows:
after each iteration is finished, the whole ant colony is sequenced from small to large according to the loss value, and the loss value is the largest
Individual is removed from the remainder +.>
The individuals whose loss values are in the middle position are randomly selected for replication and filled in to the truncate +.>
Forming a new population;
randomly selecting two individuals P in a new population 1 and P2 Crossing to obtain P 1 、P 2 To form new individuals by swapping a portion of them in the following manner:
wherein ,P1 、P 2 Representing the individual ants selected, comprising vectors of various source parameters of the emission source,
a and BETA represent crossing points, the range is (0, 1) and A < BETA. 6. The method for inverting air pollution sources based on improved ant colony algorithm according to claim 3, wherein the updating of the ant individuals is achieved by the following formula:
τ i (n+1)＝(1-ρ)τ i (n)+Δτ i (n)
wherein n is the current iteration number; ρ represents a pheromone volatilization factor; τ i Pheromone representing the ith ant; Δτ i Pheromone delta representing the ith ant; k represents a constant coefficient; l (L) i Loss values representing individual i ants; mu represents a coefficient for which the pheromone delta tau is increased if the loss value of the n-1 generation is greater than that of the n generation for the i-th ant i And increasing, and otherwise decreasing.
The ant individual according to the current iteration times is used as an optimal ant individual, and each source parameter corresponding to the ant individual is used for pollution source investigation, specifically:
and setting the length R as a radius by taking the coordinates (x, y) of the optimal ant individual as the center, and sequencing emission sources for emission of certain pollutants in the circular area according to the distance from the inversion pollution sources to obtain the predicted pollution contribution rate of each emission source so as to determine the air pollution sources.
The contribution rate formula is as follows:
wherein ,Pi Indicating the pollution probability of the ith emission source; d (D) i Representing the distance from the ith emission source to the plane coordinates (x, y) corresponding to the optimal solution; m is the number of emissions sources within the investigation range.
An air pollution source inversion system based on an improved ant colony algorithm, comprising:
the data acquisition module is used for acquiring monitoring data of all monitoring stations with air pollution exceeding standards;
the model construction module is used for constructing and optimizing a gas diffusion model and forming an air pollution source inversion model together with the target optimization function;
and the inversion module is used for carrying out iterative solution on the air pollution source inversion model through an ant colony algorithm to obtain pollution source information so as to determine the air pollution source.
The invention has the beneficial effects and advantages that:
the invention uses the improved ant colony algorithm, uses the cross operation of the genetic algorithm, solves the defect that the traditional ant colony algorithm is easy to fall into local optimum, improves the updating mechanism of pheromone, solves the problem that the traditional group algorithm has slower convergence speed, further strengthens the searching capability, can invert the source parameter information of pollutants causing air pollution alarm more quickly and accurately, verifies the effectiveness of the method through relevant experiments, and provides practical significance for air pollution control.
Establishing a gas diffusion model according to a gas diffusion rule, and establishing an air pollution source inversion model based on the gas diffusion model; and secondly, solving the model by adopting an improved ant colony algorithm, improving the model from two aspects of easy sinking local extremum and convergence speed of the ant colony algorithm, and combining the crossing thought of a genetic algorithm and updating a pheromone updating mechanism of the ant colony algorithm, thereby enhancing the global searching capability of the ant colony algorithm and accelerating the convergence speed, further achieving the purpose of quickly and accurately inverting the relevant parameter information of an air pollution source, achieving the quick positioning of the pollution source and carrying out targeted treatment. Meanwhile, the air pollution source inversion method based on the improved ant colony algorithm provided by the invention can be also suitable for the source inversion problems of water pollution, soil pollution and the like.
Drawings
FIG. 1 is a flow diagram of an air pollution source inversion method based on an improved ant colony algorithm;
fig. 2 is a flow chart of the modified ant colony algorithm.
Detailed Description
In order to make the above objects, features and advantages of the present invention more comprehensible, the following description refers to the embodiments accompanied with examples. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit or scope of the invention, which is therefore not limited to the specific embodiments disclosed below.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
An air pollution source inversion method based on an improved ant colony algorithm comprises the following steps:
step 1: and observing the monitoring data of the deployed monitoring stations, and collecting and recording the monitoring data of all the monitoring stations when the monitoring stations send out air pollution exceeding alarms.
Step 2: and establishing a proper gas diffusion model according to a turbulent diffusion theory, modifying the gas diffusion model according to actual requirements and conditions, and establishing a target optimization function based on the modified gas diffusion model, namely an air pollution source inversion model.
Step 3: the traditional ant colony algorithm is improved, and the defects that the traditional ant colony algorithm is easy to fall into a local optimal value and the initial convergence rate is low are corrected.
Step 4: and carrying out iterative solution by using an improved ant colony algorithm air pollution source inversion model, and finally obtaining relevant parameter information of the pollution source.
Step 5: and (3) based on the solved pollution source related information, checking enterprises in a small peripheral range, and finally determining enterprises with excessive pollution emission.
The monitoring data of the monitoring station is the real-time concentration of a series of pollutants monitored by the monitoring station and related meteorological parameters.
The air pollution exceeding alarm is an air pollution alarm sent when a certain air pollutant monitored by the monitoring station exceeds an alarm threshold value.
The monitoring data of all monitoring stations at the moment are collected and recorded as the real-time concentration of pollutants causing the air pollution exceeding alarm.
The suitable gas diffusion model is an overhead continuous point source diffusion model, and the expression form is as follows:
wherein, C (x, y, z) represents the gas simulation concentration at any point in downwind direction; q (Q) 0 Indicating the intensity of the emission source; u represents the average wind speed; x is x 0 and y0 Respectively representing the plane coordinates of the emission sources; x, y and plane coordinates respectively representing the monitoring station; h 0 Representing the height of the emission source; z represents the vertical coordinate of any point downwind of the discharge port; sigma (sigma) y and σz The diffusion coefficients representing the in-plane vertical and vertical downwind directions can be determined according to the GIFFORD model diffusion coefficient equation, which generally behaves in the form of:
in the formula ,ω1 、ω 2 、θ 1 、θ 2 As a coefficient, the value of which is determined by the current atmospheric stability level; x and x 0 The abscissa of the plane of any point in the wind direction under the pollution source and the abscissa of the plane of the position of the pollution source are respectively shown.
The modified gas diffusion model only needs to pay attention to the near-ground pollutant concentration distribution, and z=0, and the expression form is as follows:
the air pollution source inversion model is a target loss function established based on the modified gas diffusion model, and adopts a square loss function, and the expression form is as follows:
wherein ,
indicating the concentration of the contaminant actually detected by the monitoring station,/-, etc.>
The pollutant concentration obtained by the simulation of the gas diffusion model is shown. From the above equation, the problem of inversion of the source parameters can be translated into a pair L tar Problem of performing optimization such that L tar The smallest solution is the relevant parameter of the pollution source needing inversion.
The traditional ant colony algorithm is an intelligent algorithm for iteratively finding the optimal foraging path by utilizing pheromones remained by ants in the foraging process to continuously enable the whole population to adjust the next moving direction.
The improved ant colony algorithm is an ant colony algorithm which is improved for the defects of easy sinking into local optimum and slow convergence speed in the traditional ant colony algorithm. The method comprises the following steps:
step 1: the cross operation in the analog genetic algorithm is selected and exchanged, so that different ants are constructed, the diversity of the whole ant colony is maintained, and the algorithm can jump out a local extremum.
(1) After each iteration is finished, the whole ant colony is sequenced from small to large according to the loss, and the loss degree is maximized
Individual is removed from the remainder +.>
The individuals with the loss degree in the middle position are randomly selected for replication and filled to be deleted +.>
Where a new population is formed.
(2) Randomly selecting two individuals P in the new population 1 and P2 Crossover is performed, and P is mapped in a crossover form in an analog genetic algorithm 1 、P 2 To form new two individuals, and if the two new individuals are more optimal than the original two individuals. The specific crossing mode is as follows:
wherein P1 、P 2 A matrix representing the hypothetical contamination source related parameters contained by the selected ant,
a and BETA represent crossing points, the range is (0, 1) and A < BETA.
Step 2: and the pheromone updating mechanism is improved, so that the searching efficiency of the algorithm is improved. And determining whether the current available solution should be increased or decreased according to the magnitude relation between the loss value of the nth generation and the loss value of the nth generation of the ith ant. The whole updating process is as follows:
τ i (n+1)＝(1-ρ)τ i (n)+Δτ i (n)
wherein n is the current iteration number; ρ represents a pheromone volatilization factor; τ i Pheromone representing the ith ant; Δτ i Pheromone delta representing the ith ant; k represents a constant coefficient; l (L) i Loss values representing individual i ants; mu represents a coefficient, and for the ith ant, if the loss value of the n-1 generation is larger than that of the nth generation, the pheromone increment is increased, and otherwise, the pheromone increment is decreased.
The method for carrying out iterative solution by using the improved ant colony algorithm air pollution source inversion model comprises the following steps:
step 1: the collected main pollutant actual measurement concentration causing the air pollution alarm
The position information (x, y) corresponding to the ith monitoring station is substituted into the objective loss function L tar Is a kind of medium.
Step 2: determining a current atmospheric stability level according to the relevant meteorological parameters of the monitoring station, and determining sigma by referring to a GIFFORD model diffusion coefficient equation and the current atmospheric stability level y and σz Is substituted into the target loss function L tar Is a kind of medium.
Step 3: parameters required by the algorithm are set, including importance factors alpha and beta of information and heuristic functions, ant colony scale m, iteration times N, pheromone volatilization factor rho, pheromone matrix X and upper and lower limits of various parameters.
Step 4: and starting an algorithm, carrying out iterative solution on the target loss function until the iteration is carried out for the nth time, and recording the target loss value and the corresponding optimal solution at the moment.
And the pollution source enterprise investigation is to examine enterprises related to the emission of related pollutants in an area with R as a radius according to the related pollution source parameters of the reverse performance, and the forecast pollution contribution rate ranking of each enterprise is given according to the distance from the inversion pollution source.
As shown in fig. 1, an air pollution source inversion method based on an improved ant colony algorithm takes a PM2.5 pollution exceeding alarm as an example, and includes the following steps:
step 1: and (5) collecting, recording and preprocessing the out-of-standard alarm data. And observing the monitoring condition of the deployment network of the whole monitoring device, setting a PM2.5 sensitivity threshold to be an alarm for exceeding PM2.5 pollution when one or a plurality of monitoring devices appear, observing for 5 minutes, and then recording PM2.5 real-time concentration data of all devices initiating the alarm for exceeding the pollution. The wind direction and wind speed information of these devices are recorded simultaneously. At the same time, abnormal pollutant data or meteorological data caused by equipment problems (equipment damage, maintenance and the like), manual operations (deletion, modification and the like) and other factors are removed. The remaining data is taken as the dataset for this secondary source inversion.
Step 2: and establishing a target loss function. Establishing target loss functions (such as square loss function, mean square error function, R) by using a near-ground concentration distribution model (namely z=0) of the wind direction under the pollution source modified based on an overhead point source continuous diffusion model (namely a Gaussian plume model) 2 Loss function, etc.), taking here as an example a square loss function:
wherein Ltar For the value of the loss to be the target,
and />
The actual concentration and the predicted concentration of the air pollutant corresponding to the ith detection device are respectively. When L tar Approaching 0, the closer the current solution is to the actual pollution source.
Step 3: sigma determination with reference to the GIFFORD model diffusion coefficient equation y and σz Is taken from the real-time atmosphereStability is determined, and its general expression is:
wherein x and x 0 The horizontal plane coordinates of any point of the wind direction under the pollution source and the horizontal plane coordinates of the position of the pollution source are respectively; omega 1 、ω 2 、θ 1 、θ 2 As a coefficient, its value is determined by the current atmospheric stability level.
Step 4: and (3) adopting M-ACO (improved ant colony algorithm) to carry out iterative solution. As shown in fig. 2, the method comprises the following steps:
step 4.1: initializing various parameters of an algorithm, importance factors alpha and beta of pheromone and heuristic functions, ant colony scale m, iteration number N, pheromone volatilization factor rho, and various source parameters (pollution source intensity Q and position x 0 ，y 0 And upper and lower limits of the release height H).
Step 4.2: the individual ants are randomly generated in the feasible domain composed of the source parameters, and the pheromone concentration of each ant is initialized.
Step 4.3: each ant selects its own next location according to the state transition rules and the search strategy.
Step 4.3.1: the state transition probabilities are as follows:
wherein Pi (n) represents the probability that the ith ant selects the next position, when P i The larger (n) is, the greater the probability that the ant individual is selected next time.
Step 4.3.2: setting a state transition constant P 0 When P i (n) is greater than P 0 When a local search strategy is adopted, otherwise a global search strategy is adopted, which can be expressed as follows:
where next is the next position of the current ant; now is the current position of the current ant; rand represents a random number between (0, 1); step represents the local search step; lambda represents the step size coefficient, which takes the value of the inverse of the number of iterations
range represents the feasible region range.
Step 4.4: and after all ant searches are completed, updating the current global optimal solution, and then performing selection and exchange operations. After each iteration is finished, the whole ant colony is sequenced from small to large according to the loss, and the loss degree is maximized
Individual is removed from the remainder +.>
The individuals with the loss degree in the middle position are randomly selected for replication and filled to be deleted +.>
Where a new population is formed.
Randomly selecting two individuals P in the new population 1 and P2 Crossover is performed, and P is mapped in a crossover form in an analog genetic algorithm 1 、P 2 To form new individuals, in the following way:
wherein P1 、P 2 A matrix representing the hypothetical contamination source related parameters contained by the selected ant,
a and BETA represent crossing points, the range is (0, 1) and A < BETA.
Step 4.5: and determining whether the current available solution should be increased or decreased according to the magnitude relation between the loss value of the nth generation and the loss value of the nth generation of the ith ant. The whole updating process is as follows:
τ i (n+1)＝(1-ρ)τ i (n)+Δτ i (n)
wherein n is the current iteration number; ρ represents a pheromone volatilization factor; τ i Pheromone concentration representing the ith ant; Δτ i Pheromone delta representing the ith ant; k represents a constant coefficient; l (L) i Loss values representing individual i ants; mu represents a coefficient, and for the ith ant, if the loss value of the n-1 generation is larger than that of the nth generation, the pheromone increment is increased, and otherwise, the pheromone increment is decreased.
Step 5: after the iteration number reaches N in the step 4, observing a loss function L tar Whether or not to converge. If the pollution source is converged, performing pollution source investigation according to pollution source parameters corresponding to the optimal ant individuals, and listing related enterprises related to PM2.5 emission in a circular area by taking coordinates (x, y) of an optimal solution as a center and a certain length R as a radius, so as to provide a pollution probability investigation list of each enterprise, wherein the probability calculation formula is as follows:
wherein ,Pi Indicating the pollution probability of the ith enterprise; d (D) i Representing the distance from the ith enterprise to the plane coordinates (x, y) corresponding to the optimal solution; m is the number of related enterprises in the investigation range
And by checking the probability list, an enterprise causing the PM2.5 pollution exceeding accident can be found, and meanwhile, the accuracy of inversion can be evaluated according to the comparison between the checked result and the inversion result.
The above detailed description is intended to illustrate the present invention by way of example only and not to limit the invention to the particular embodiments disclosed, but to limit the invention to the precise embodiments disclosed, and any modifications, equivalents, improvements, etc. that fall within the spirit and scope of the invention as defined by the appended claims.
Claims (9)
1. An air pollution source inversion method based on an improved ant colony algorithm is characterized by comprising the following steps of:
step 1: collecting monitoring data of all monitoring stations with air pollution exceeding standards;
step 2: constructing and optimizing a gas diffusion model, and forming an air pollution source inversion model together with a target optimization function;
step 3: and carrying out iterative solution on the air pollution source inversion model through an ant colony algorithm to obtain pollution source information so as to determine the air pollution source.
2. The method for inverting an air pollution source based on an improved ant colony algorithm according to claim 1, wherein the step 2 is specifically as follows:
and (3) constructing a gas diffusion model:
wherein, C (x, y, z) represents the gas simulation concentration of any point (x, y, z) in downwind direction; q (Q) 0 Indicating the intensity of the emission source; u represents the average wind speed; x is x 0 、y 0 Representing the horizontal plane coordinates of the emission source; x and y represent horizontal plane coordinates of the monitoring station; h 0 Representing the height of the emission source; z represents the vertical coordinate of any point (x, y, z) downwind of the discharge port; sigma (sigma) y and σz Represents the horizontal and vertical diffusion coefficients, expressed in the form:
in the formula ,ω1 、ω 2 、θ 1 、θ 2 Is a coefficient;
the gas diffusion model is optimized as follows: let z=0, the optimized gas diffusion model is as follows:
the standard loss function is constructed according to the optimized gas diffusion model as follows:
wherein ,
indicating the concentration of the contaminant actually detected by the monitoring station,/-, etc.>
Representing the result after the optimizationPollutant concentration, L, obtained by gas diffusion model tar The loss function value is expressed to evaluate the accuracy of the current result, the closer it is to 0, the more accurate the current result is, and T represents the number of monitoring stations.
3. The method for inverting an air pollution source based on an improved ant colony algorithm according to claim 1, wherein the step 3 is specifically as follows:
step 3.1: initializing various parameters: the importance factors alpha and beta of the pheromone and heuristic function, the ant colony scale m, the iteration number N, the pheromone volatilization factor rho, the upper limit and the lower limit of each source parameter of the emission source, wherein each source parameter of the emission source comprises emission source intensity Q 0 Position (x) 0 、y 0 ) Release height H 0 ；
Step 3.2: randomly generating ant individuals in a feasible domain formed by various source parameters, and initializing the pheromone tau of each ant i (i＝1,2,3,…,m)；
Step 3.3: each ant selects the next position according to the state transition rule and the search strategy;
step 3.4: after all ant searches are completed, updating the current global optimal solution, and then carrying out selection and exchange operation;
step 3.5: updating the ants according to the magnitude relation between the loss value of the nth generation and the loss value of the nth generation of the ith ants; when the iteration number reaches N, if the loss function L tar And converging, namely performing pollution source investigation according to each source parameter corresponding to the ant individual with the current iteration times serving as an optimal ant individual.
4. An air pollution source inversion method based on an improved ant colony algorithm according to claim 3, wherein said step 3.3 is achieved by:
step 3.3.1: the state transition probabilities are as follows:
wherein, represent heuristic factor P i (n) represents the probability that the ith ant selects the next position, when P i (n) when larger, the current solution is closer to the various source parameters of the pollution source; τ i Pheromone, eta representing the ith ant i Heuristic functions representing the ith ant; n is the current iterated times;
step 3.3.2: setting a state transition constant P 0 When P i (n) is greater than P 0 When a local search strategy is adopted, otherwise a global search strategy is adopted, which can be expressed as follows:
wherein next is the next position of the current ant; now is the current position of the current ant; rand represents a random number between (0, 1); step represents the local search step; lambda represents the step size coefficient, which takes the value of the inverse of the number of iterations
range represents the feasible region range.
5. An air pollution source inversion method based on an improved ant colony algorithm according to claim 3, wherein said step 3.4 is specifically as follows:
after each iteration is finished, the whole ant colony is sequenced from small to large according to the loss value, and the loss value is the largest
Individual is removed from the remainder +.>
The individuals whose loss values are in the middle position are randomly selected for replication and filled in to the truncate +.>
Forming a new population;
randomly selecting two individuals P in a new population 1 and P2 Crossing to obtain P 1 、P 2 To form new individuals by swapping a portion of them in the following manner:
wherein ,P1 、P 2 Representing the individual ants selected, comprising vectors of various source parameters of the emission source,
a and BETA represent crossing points, the range is (0, 1) and A < BETA.
6. The method for inverting air pollution sources based on improved ant colony algorithm according to claim 3, wherein the updating of the ant individuals is achieved by the following formula:
τ i (n+1)＝(1-ρ)τ i (n)+Δτ i (n)
wherein n is the current iteration number; ρ represents a pheromone volatilization factor; τ i Pheromone representing the ith ant; Δτ i Pheromone delta representing the ith ant; k represents a constant coefficient; l (L) i Loss values representing individual i ants; mu represents a coefficient of the sum of the coefficients,for the ith ant, if the loss value of the n-1 generation is larger than that of the nth generation, the pheromone delta tau is increased for the ith ant i And increasing, and otherwise decreasing.
7. The method for inverting the air pollution source based on the improved ant colony algorithm according to claim 1, wherein the ant individual according to the current iteration number is used as an optimal ant individual, and each corresponding source parameter is used for pollution source investigation, specifically:
and setting the length R as a radius by taking the coordinates (x, y) of the optimal ant individual as the center, and sequencing emission sources for emission of certain pollutants in the circular area according to the distance from the inversion pollution sources to obtain the predicted pollution contribution rate of each emission source so as to determine the air pollution sources.
8. The method for inverting an air pollution source based on an improved ant colony algorithm of claim 7, wherein the contribution rate formula is as follows:
wherein ,Pi Indicating the pollution probability of the ith emission source; d (D) i Representing the distance from the ith emission source to the plane coordinates (x, y) corresponding to the optimal solution; m is the number of emissions sources within the investigation range.
9. An air pollution source inversion system based on an improved ant colony algorithm, comprising:
the data acquisition module is used for acquiring monitoring data of all monitoring stations with air pollution exceeding standards;
the model construction module is used for constructing and optimizing a gas diffusion model and forming an air pollution source inversion model together with the target optimization function;
and the inversion module is used for carrying out iterative solution on the air pollution source inversion model through an ant colony algorithm to obtain pollution source information so as to determine the air pollution source.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111419495.8A CN116187150A (en) | 2021-11-26 | 2021-11-26 | Air pollution source inversion method based on improved ant colony algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111419495.8A CN116187150A (en) | 2021-11-26 | 2021-11-26 | Air pollution source inversion method based on improved ant colony algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116187150A true CN116187150A (en) | 2023-05-30 |
Family
ID=86435009
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111419495.8A Pending CN116187150A (en) | 2021-11-26 | 2021-11-26 | Air pollution source inversion method based on improved ant colony algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116187150A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116541944A (en) * | 2023-07-06 | 2023-08-04 | 国网浙江省电力有限公司湖州供电公司 | Carbon emission calculation method based on comprehensive oblique photography modeling model of transformer substation |
-
2021
- 2021-11-26 CN CN202111419495.8A patent/CN116187150A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116541944A (en) * | 2023-07-06 | 2023-08-04 | 国网浙江省电力有限公司湖州供电公司 | Carbon emission calculation method based on comprehensive oblique photography modeling model of transformer substation |
| CN116541944B (en) * | 2023-07-06 | 2023-10-20 | 国网浙江省电力有限公司湖州供电公司 | Carbon emission calculation method based on comprehensive oblique photography modeling model of transformer substation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110363347B (en) | Method for predicting air quality based on neural network of decision tree index | |
| Qiu et al. | Atmospheric dispersion prediction and source estimation of hazardous gas using artificial neural network, particle swarm optimization and expectation maximization | |
| CN110598953A (en) | Space-time correlation air quality prediction method | |
| CN109063938B (en) | Air quality prediction method based on PSODE-BP neural network | |
| CN115358332A (en) | Atmospheric pollution tracing method for multi-source data | |
| CN107346459B (en) | Multi-mode pollutant integrated forecasting method based on genetic algorithm improvement | |
| CN110766222B (en) | PM2.5 concentration prediction method based on particle swarm parameter optimization and random forest | |
| CN113610243B (en) | Atmospheric pollutant tracing method based on coupled machine learning and correlation analysis | |
| CN112163367B (en) | Firefly algorithm and cellular automaton fused city expansion simulation prediction method | |
| CN111414974B (en) | Microwave link rain measurement network topological structure optimization method based on communication base station | |
| CN111259522B (en) | Multi-watershed parallel calibration method of hydrologic model in geographic space | |
| Huang et al. | A new air quality forecasting model using data mining and artificial neural network | |
| CN110826244B (en) | Conjugated gradient cellular automaton method for simulating influence of rail transit on urban growth | |
| CN111489015A (en) | Atmosphere O based on multiple model comparison and optimization3Concentration prediction method | |
| CN112883632B (en) | Lithium battery equivalent circuit model parameter identification method based on improved ant colony algorithm | |
| CN111797572A (en) | Urban accident disaster evolution simulation and risk prediction early warning method | |
| CN116187150A (en) | Air pollution source inversion method based on improved ant colony algorithm | |
| Zhao et al. | Application of data mining to the analysis of meteorological data for air quality prediction: A case study in Shenyang | |
| CN116796805A (en) | PM2.5 concentration prediction method based on Gaussian process regression and deep learning | |
| Teng et al. | Prediction of particulate matter concentration in Chengdu based on improved differential evolution algorithm and BP neural network model | |
| CN116662935A (en) | Atmospheric pollutant spatial distribution prediction method based on air quality monitoring network | |
| CN116433017A (en) | Soil heavy metal pollution evaluation method, device, equipment and storage medium | |
| CN112529233A (en) | Method for predicting evapotranspiration amount of lawn reference crops | |
| CN114444796A (en) | Cross-regional air pollution prediction method and system based on graph neural network | |
| Hu et al. | Neural network-assisted expensive optimisation algorithm for pollution source rapid positioning of drinking water |
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 |