CN110824107A - Air quality forecasting method based on dynamic inversion emission source data - Google Patents
Air quality forecasting method based on dynamic inversion emission source data Download PDFInfo
- Publication number
- CN110824107A CN110824107A CN201910935248.XA CN201910935248A CN110824107A CN 110824107 A CN110824107 A CN 110824107A CN 201910935248 A CN201910935248 A CN 201910935248A CN 110824107 A CN110824107 A CN 110824107A
- Authority
- CN
- China
- Prior art keywords
- data
- emission source
- air quality
- source data
- emission
- 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.)
- Granted
Links
- 238000013277 forecasting method Methods 0.000 title claims abstract description 8
- 238000012544 monitoring process Methods 0.000 claims abstract description 52
- 238000000034 method Methods 0.000 claims abstract description 30
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 claims abstract description 11
- 239000003546 flue gas Substances 0.000 claims abstract description 11
- 238000005457 optimization Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 15
- 239000003344 environmental pollutant Substances 0.000 claims description 12
- 231100000719 pollutant Toxicity 0.000 claims description 12
- 239000000779 smoke Substances 0.000 claims description 10
- 230000010354 integration Effects 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 239000000470 constituent Substances 0.000 claims description 3
- 238000009792 diffusion process Methods 0.000 claims description 3
- 238000006467 substitution reaction Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000012806 monitoring device Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 41
- 238000004422 calculation algorithm Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 238000011835 investigation Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000012804 iterative process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 239000000356 contaminant Substances 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0004—Gaseous mixtures, e.g. polluted air
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
- Y02A50/20—Air quality improvement or preservation, e.g. vehicle emission control or emission reduction by using catalytic converters
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/84—Greenhouse gas [GHG] management systems
- Y02P90/845—Inventory and reporting systems for greenhouse gases [GHG]
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Environmental & Geological Engineering (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Biochemistry (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Combustion & Propulsion (AREA)
- Primary Health Care (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Analytical Chemistry (AREA)
- Operations Research (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Educational Administration (AREA)
- Atmospheric Sciences (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an air quality forecasting method based on dynamic inversion emission source data, which utilizes multiple means to dynamically update the emission source data and improves the accuracy of air quality forecasting. According to the method, emission source data Sa of an emission source provided with flue gas online monitoring equipment is calculated through emission parameters, and the data of the Sa are not adjusted during inversion; replacing the content in the emission source data Sb generated based on the emission source manifest with the data in Sa to generate emission source data Sc; after air quality forecast data is obtained by using emission source data Sc, emission source data S with minimum difference between the air quality forecast data and monitoring data is inverted1(ii) a Then on-line monitoring device for non-installed smokePreparing an emission source, and combining emission source data before and after inversion to obtain final emission source data SfinalAnd generating a final air quality forecasting result by utilizing an air quality forecasting technology.
Description
Technical Field
The invention belongs to the technical field of air quality prediction, and particularly relates to an air quality prediction method based on dynamic inversion emission source data.
Background
The air quality numerical prediction is a method for predicting air quality by applying an air quality prediction numerical mode to simulate the distribution and evolution of future atmospheric pollutants based on an emission source list, meteorological field prediction data, environment monitoring data and the like. Current air quality forecasting techniques employ emissions source inventory data based on pollution source surveys and data processing. The emission source list establishment is to carry out preliminary investigation on the emission sources in the list establishment area according to corresponding regulations, specifications and technical methods, to clarify the main composition of the local emission sources, and to select a proper emission source classification level so as to determine activity level data investigation and collection objects in the source list establishment process. The investigation and collection process of the data should be combined with the existing data statistics system, and the related information is preferentially obtained from databases such as environmental statistics, pollution source census and the like. According to the information of the combustion type of the emission source, the fuel type, the emission coefficient, the environmental protection measure and the like, a mathematical model is adopted for direct and indirect estimation.
Obviously, a large amount of investigation work is needed in the process of compiling the emission source list, so that time and labor are wasted, and the compiling period is long. When the method is actually applied to air quality prediction, pollution source emission changes caused by industrial adjustment cannot be reflected in emission source list data in time, and the prediction accuracy is influenced.
Disclosure of Invention
In view of the above, the present invention provides an air quality prediction method based on dynamic inversion of emission source data, which utilizes mass data of an existing atmospheric environment monitoring network, and on the basis of an emission source list, applies an air quality prediction technology and an emission source inversion technology to dynamically update emission source data, and performs air quality prediction on the basis of the air quality prediction technology, so as to improve the accuracy of air quality prediction.
In order to solve the technical problem, the invention is realized as follows:
an air quality forecasting method based on dynamic inversion emission source data comprises the following steps:
acquiring emission parameters from an emission source provided with the online flue gas monitoring equipment, and calculating first emission source data Sa based on the emission parameters; generating second emission source data Sb according to the original emission source list;
replacing corresponding content in the second emission source data Sb by the first emission source data Sa to generate third emission source data Sc;
inputting third emission source data Sc, meteorological field forecast data and air quality monitoring data Co into an air quality forecasting model, and calculating to obtain air quality forecast data C;
fourthly, according to an objective function J for measuring the difference between the air quality forecast data C and the air quality monitoring data Co, inverting emission source data S enabling the objective function J to be minimum1(ii) a During inversion, data from the first discharge source data Sa are not adjusted;
step five, for the emission source without the flue gas on-line monitoring equipment, combining the emission source data before and after inversion to obtain new emission source data SnewSubstitution of inverted emission Source data S1To obtain final emission source data Sfinal；
Step six, emission source data SfinalAnd inputting the air quality forecasting model to obtain a final air quality forecasting result.
Wherein the emission source data S1The optimization solution of (a) is expressed as:
wherein, formula (I) is an objective function optimizing expression, and formula (II) is an air quality forecasting model; t is the time period of the adopted monitoring data, and T is the time; Ω represents integration over the full spatial range; lambda is a weight coefficient, reflects the reliability of observation data at different spatial positions and is a function of the spatial positions; the emission distribution S is S (x, y, z, t), x and y are the positions of the discharge ports, z is the smoke lifting height, and t represents time; f (C, P)1,P2...Pn) Is a mathematical description representing a series of processes, P, of transport, diffusion, chemical transformation undergone by a pollutant in the atmosphere1,P2...PnIs n constituent elements in the meteorological field forecast data;
introducing a companion variable C into formula (I)*Constructing a Lagrangian function L and deriving a adjoint model (III):
the fourth step specifically comprises the following steps:
step 4a, substituting air quality forecast data C and air quality monitoring data Co into an expression of an objective function to obtain an objective function value J; running the adjoint model (III), the gradient vector g, i.e. the adjoint variable C, is calculated*；
Step 4b, accompanying variable C*Target function value J and emission source data vector to be solved of required dimension
Inputting the data into a software package for gradient information optimization, and automatically resolving and outputting an emission source data optimization result S of the iteration by the software packagek(ii) a The upper corner mark k represents the kth iteration;
step 4c, optimizing the emission source data SkSubstituting the air quality prediction model with the air quality prediction model to obtain new air quality prediction data;
step 4d, the new air quality forecast data is utilized to execute the steps 4a to 4c again, and the iteration is carried out for a plurality of times in a circulating mode until the difference between the air quality forecast data and the air quality monitoring data meets the set condition, and the inversion result S is obtained1。
Preferably, the manner of determining whether the difference between the forecast data and the air quality monitoring data satisfies the set condition in step 4d is as follows:
at the k-th iteration, the adjoint variable C obtained at the k-th and k-1-th iterations are comparedk *And Ck-1 *Judging Ck *-Ck-1 *If the absolute value is smaller than the set threshold value, if so, judging that the difference between the air quality forecast data and the air quality monitoring data meets the set condition, exiting the circulation, and obtaining an optimization result S of the emission source datak-1As the emission source data S1。
Has the advantages that:
1. the method defines an objective function for measuring the difference between air quality monitoring data Co and air quality forecast data C, and inverts emission source data enabling the objective function J to be minimum; emission source data S obtained by inversion1Performing air quality forecast calculation again to improve the accuracy of air quality forecast; since the inversion process is based on air quality monitoring data.
2. The invention directly obtains the emission parameters from the online smoke monitoring system for the emission sources provided with the online smoke monitoring equipment, calculates the emission source data by using the emission parameters, and does not adjust the emission sources in an inversion algorithm, thereby fully utilizing reliable monitoring data to improve the accuracy of air quality prediction.
3. For the emission source without the flue gas on-line monitoring equipment, analyzing the relative error sigma of data acquisition according to the acquisition means of the emission source list, and then inverting the emission source data S before inversion and inverting the adjusted emission source data S1Combining to obtain new emission source data SnewThe method not only considers the source error of the emission source data, but also utilizes the information of air quality monitoring, and more objectively reflects the current situation of the emission source.
4. The inversion process of the emission source data employs an optimization scheme based on gradient information. And because of the nonlinearity of the target function, the target function is reduced in the process of adjusting the data of the pollution source, and the gradient direction of the target function is changed, so that the invention designs a gradual iterative and progressive calculation process, and the line number gradient is recalculated once per iteration, thereby ensuring the calculation accuracy.
Drawings
FIG. 1 is an air quality forecasting method based on dynamic inversion of emission source data.
FIG. 2 is a detailed flow chart of the inversion step of FIG. 1.
Detailed Description
The invention is described in detail below by way of example with reference to the accompanying drawings.
The invention provides an air quality forecasting method based on dynamic inversion emission source data, which has the basic idea that when air quality forecasting is carried out by utilizing emission source data, air quality monitoring data Co and meteorological field forecasting data, an objective function for measuring the difference between the air quality monitoring data Co and the air quality forecasting data C is further defined, and emission source data enabling the objective function J to be minimum is inverted; emission source data S obtained by inversion1And performing air quality forecast calculation again to obtain an accurate air quality forecast result.
Because the air quality monitoring data Co for inverting the emission source data has errors and the monitoring sites are not uniformly distributed, the air quality monitoring data only partially reflects the space-time component of the concentration of the atmospheric pollutantsAnd (4) distributing information. If inversion is completely adopted, the obtained emission source data S1In order to make the predicted concentration and the monitored concentration as close as possible, the inversion algorithm may make excessive adjustments to some emission sources, which makes it difficult to actually reflect the current situation of the emission sources. Therefore, in view of this fact, the present invention proposes the following new method to improve the accuracy of inverting the emission source data:
1. for the emission source provided with the smoke online monitoring equipment, directly acquiring emission parameters from a smoke online monitoring system, calculating emission source data by using the emission parameters, and not adjusting the emission sources in an inversion algorithm;
2. for the emission source without the flue gas on-line monitoring equipment, analyzing the relative error sigma of data acquisition according to the acquisition means of the emission source list, and then inverting the emission source data S before inversion and inverting the adjusted emission source data S1Combining to obtain new emission source data SnewThe method not only considers the source error of the emission source data, but also utilizes the information of air quality monitoring, and more objectively reflects the current situation of the emission source. When the relative error sigma is smaller in designing the combined algorithm, the original data S is shown0The more reliable, the closer the inverted emission source data is to S0(ii) a The larger the relative error σ is, the more the original data S is indicated0The less reliable the inverted emission source data is, the closer to S1(ii) a One combination algorithm is:
Snew＝S1+(S0-S1)e-σ
wherein S is0For inverting the preceding data, S1Representing the data after inversion, the relative error sigma can be determined according to the acquisition means of the emission source list, and the data can be obtained in advance through statistics and the like.
Based on the above analysis, the air quality forecasting method of the present invention specifically includes the steps of:
step 1, obtaining emission parameters from an emission source provided with the online flue gas monitoring equipment, and calculating emission source data Sa based on the emission parameters. Only data of emission sources of the monitoring devices are included in the Sa.
The emission parameters include the concentration of various pollutants in the flue gas and the flue gas parameters (temperature, pressure, flow rate or quantity, humidity, oxygen content, etc.). The emission source data records the emission distribution S, the contents of which include the emission of the pollution source, the spatial coordinates (x, y), and the smoke lifting height z.
Calculating emission source data based on the emission parameters as: calculating the pollutant discharge amount in unit time according to the pollutant concentration and the flue gas flow in the discharge parameters; according to the smoke temperature in the emission parameters and meteorological elements provided by meteorological field forecast data, the smoke lifting height z can be calculated.
And 2, generating emission source data Sb according to the original emission source list. In the step, the list of the emission sources is arranged into a form which can be identified by an air quality forecasting model. The emission source data Sb includes data of all emission sources.
The step 1 and the step 2 are not in sequence.
And 3, replacing corresponding content in the emission source data Sb by using the emission source data Sa obtained based on the monitoring data to obtain emission source data Sc. Since the emission source data Sa obtained based on the monitored data is reliable data, it needs to be retained.
And 4, inputting the emission source data Sc, the meteorological field forecast data and the air quality monitoring data Co (namely pollutant concentration monitoring data) into an air quality forecasting model, and calculating to obtain forecast data C of pollutant concentration space-time distribution, namely air quality forecast data. The air quality monitoring data Co provides background concentration data on the boundary for the air quality forecasting model mode.
Step 5, according to a defined target function J for measuring the difference between the forecast data C and the air quality monitoring data Co, inverting the emission source data S enabling the target function J to be minimum1. During inversion, no adjustment is made to the data from the emission source data Sa. Since the air quality monitoring data Co is discrete data and the forecast data C is continuous data, it is necessary to obtain the emission source data by inversion in the case of minimizing the air quality monitoring data Co and the forecast data C.
Step 6, on-line monitoring device for non-installed smokePreparing an emission source, combining emission source data before and after inversion to obtain new emission source data Snew(ii) a By using SnewSubstitution of inverted emission source data S1To obtain final emission source data Sfinal. The combination method is as described above, or the combination is performed in a weighted manner.
Step 7, emission source data SfinalAnd inputting the air quality forecasting model to obtain an accurate air quality forecasting result.
Due to the fact that the calculated amount is large, dynamic inversion control of the emission source data is suitable and better at a certain frequency, and monthly updating is achieved to reflect the influence of seasonal changes on the emission source data. Because the concentration monitoring data Co is added in the inversion process, the change of the emission source caused by the change of the enterprise operation state, such as enterprise shutdown, plant address relocation and the like, can also be reflected in the dynamically inverted emission source data, so that timely information is provided for the management and control of the emission source, and the air quality prediction accuracy is improved.
The mathematical theory and iterative algorithm for inversion of emission sources are as follows
Let the contaminant concentration C satisfy the following basic equation in the form of an operator:
wherein, F (C, P)1,P2...Pn) Is a mathematical description representing a series of processes, P, of transport, diffusion, chemical transformation undergone by a pollutant in the atmosphere1,P2...PnIs n constituent elements in the meteorological field forecast data, such as flow rate, temperature, humidity, oxygen content, etc. S is emission source data, i.e., emission amount distribution, S ═ S (x, y, z, t) is pollution source emission amount, x, y are positions of discharge ports, z is smoke lifting height, and t represents time.
Due to the fact that there are many pollutant concentration measurement data C at different time and different positionsOThe invention is to deduce the value and distribution of the emissions of the pollutant from these data, so that from the emissions dataAnd the pollutant concentration value obtained by starting simulation at each moment is closest to the observed value. This closest approach may be to define an objective function in terms of least squares:
wherein T is the time period of the adopted monitoring data, and T is the time; Ω represents integration over the full spatial range; lambda is a weight coefficient, reflects the reliability of observation data at different spatial positions and is a function of the spatial positions; for points without observed data, λ is 0.
The emission amount distribution S (x, y, z, t) that minimizes the objective function value is obtained. The objective function includes the concentration C, which is in turn related to the pollution source emission distribution S by equation (1).
Due to the initial field inversion problem, it is solved by defining a nonlinear optimization control problem. The objective function J of the optimization control problem depends on a series of state variables and control variables u (in this problem u is the pollution source S), and the objective function value is changed to be optimal by adjusting the control variables to cause the state variables to change. The mathematical expression of the optimization problem is as follows:
the control variable in the above equation is the emission source emission profile: s ═ S (x, y, z, t).
Introducing Lagrange multiplier C*(accompanying variables), the Lagrange function is constructed:
the Lagrange function (4) has the requirement of obtaining an extremum
When equation of state (1) is satisfied, it is apparent that
This is true.
By
The adjoint equation can be derived.
The derivation is preceded by a modification of the second term in the Lagrange function (4) using fractional integration.
Using the above relationship, the adjoint equation is obtained as
By
To obtain
C*＝0 (7)
Theoretically, solving equations (1), (6), and (7) can yield pollution source data that satisfies the optimization problem (3). In practice, the differential equations (1) and (6) are discretized into a spatial grid to be numerically solved, wherein the equation (1) corresponds to the air quality prediction model, the equation (6) corresponds to the adjoint model, and finally a set of equations with high dimension is obtained, so that simultaneous solution is difficult.
In order to solve the problem of difficult solution, the invention utilizes gradient information to carry out iterative optimization, and only needs to solve the state equation (1) and the adjoint equation (6), and under the premise that the state equation (1) and the adjoint equation (6) are satisfied, the derivative of the Lagrange function (4) with respect to the pollution source S can be obtained from the equation (7), and the derivative is also equal to the derivative of the objective function with respect to S. Namely, it is
Since the contamination sources S are distributed discretely on the grid, a vector can be obtained by arranging in a certain order, and the derivative values of the objective function are also distributed discretely, so that a gradient vector can be obtained by arranging.
Gradient:
based on the solution thought, the concrete implementation process of the step 5 is as follows:
and 51, substituting the air quality forecast data C and the air quality monitoring data Co into an expression (2) of the objective function to obtain an objective function value J.
Step 52, substituting the air quality forecast data C and the air quality monitoring data Co into the running adjoint model (6) to calculate a gradient vector g, namely an adjoint variable C*。
Step 53, accompany variable C*Target function value J and emission source data vector to be solved of required dimension
Inputting the data into a software package for gradient information optimization, and automatically resolving and outputting an emission source data optimization result S of the iteration by the software packagek(ii) a The upper corner mark k indicates the kth iteration. Wherein the accompanying variable C*Is also in accordance withThe same order vectorizes the results. Since the invention does not process emission sources in Sa at inversion, C*Andin the vector of (2), the elements of the corresponding emission source are removed.
The algorithm for finding the optimal solution by using gradient information is well-established and can be called by a general software package, which is not described herein, please refer to Byrd et al 1995.A limited memory algorithm for bound connectivity, journal of Scientific Computing 16(5): 1190-. A functional description of a software package that is currently available is found in The module M1QN3 Version 3.3(October 2009). The software package for finding the optimal solution by using the gradient information can call a target function calculation program and a target function gradient calculation operator program, and appoint the dimension of a variable to be optimized in advance, which is the emission source data S in the invention. The emission source data S is three-dimensional spatial distribution data, and thus it is necessary to vectorize it in a certain order and input it into a software package. The software package outputs the emission source data optimizing result of the corresponding dimensionality through resolving, and the emission source data optimizing result is rearranged to obtain the emission space distribution.
Step 54, optimizing the emission source data SkSubstituting the air quality prediction model with the air quality prediction model to obtain new air quality prediction data.
The emission source data is adjusted by the running adjoint model by using the gradient value calculated by the adjoint model, so that the objective function (3) is reduced, and theoretically, the closer the objective function is to 0, the closer the observed value and the simulated value are. Since the gradient direction is the direction in which the objective function increases the fastest, the opposite direction of the gradient may make the objective function decrease the fastest. However, due to the nonlinearity of the objective function, in the process of adjusting the pollution source data, the objective function is lowered, and the direction of the gradient of the objective function is changed, so that the process is a gradual iterative process. An iterative process of step 55 needs to be performed.
Step 55, executing steps 52-54 again by using the new air quality forecast data, and performing loop iteration for multiple times until the difference between the air quality forecast data C and the air quality monitoring data Co meets the set condition, so as to obtain a final inversion result S1。
The method for judging whether the difference between the forecast data C and the air quality monitoring data Co meets the set conditions or not in the step is as follows: at the k-th iteration, the accompanying variable C obtained at the k-1 st iteration is comparedk-1 *And the accompanying variable C of the kth iterationkJudgment of||Ck*-Ck-1Whether the value of the integral is smaller than a set threshold value or not, and if so, exiting the circulation; otherwise, the loop continues.
When exiting iteration, the emission source data optimizing result Sk-1As the emission source data S1。
In summary, the above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (3)
1. An air quality forecasting method based on dynamic inversion emission source data is characterized by comprising the following steps:
acquiring emission parameters from an emission source provided with the online flue gas monitoring equipment, and calculating first emission source data Sa based on the emission parameters; generating second emission source data Sb according to the original emission source list;
replacing corresponding content in the second emission source data Sb by the first emission source data Sa to generate third emission source data Sc;
inputting third emission source data Sc, meteorological field forecast data and air quality monitoring data Co into an air quality forecasting model, and calculating to obtain air quality forecast data C;
fourthly, according to an objective function J for measuring the difference between the air quality forecast data C and the air quality monitoring data Co, inverting emission source data S enabling the objective function J to be minimum1(ii) a During inversion, data from the first discharge source data Sa are not adjusted;
step five, for the emission source without the flue gas on-line monitoring equipment, combining the emission source data before and after inversion to obtain new emission source data SnewSubstitution of inverted emission Source data S1To obtain final emission source data Sfinal；
Step six, emission source data SfinalInputting an air quality forecasting model to obtain final airAnd (5) quality forecasting results.
2. The method of claim 1, wherein the emission source data S is1The optimization solution of (a) is expressed as:
wherein, formula (I) is an objective function optimizing expression, and formula (II) is an air quality forecasting model; t is the time period of the adopted monitoring data, and T is the time; Ω represents integration over the full spatial range; lambda is a weight coefficient, reflects the reliability of observation data at different spatial positions and is a function of the spatial positions; the emission distribution S is S (x, y, z, t), x and y are the positions of the discharge ports, z is the smoke lifting height, and t represents time; f (C, P)1,P2...Pn) Is a mathematical description representing a series of processes, P, of transport, diffusion, chemical transformation undergone by a pollutant in the atmosphere1,P2...PnIs n constituent elements in the meteorological field forecast data;
introducing a adjoint variable C in the formula (I) to construct a Lagrangian function L, and deriving an adjoint model (III):
the fourth step specifically comprises the following steps:
step 4a, substituting air quality forecast data C and air quality monitoring data Co into an expression of an objective function to obtain an objective function value J; running the adjoint model (III) and calculating a gradient vector g, namely an adjoint variable C;
4b, carrying out adjoint variable C, objective function value J and emission source data vector to be solved of required dimension
Inputting the data into a software package for gradient information optimization, and automatically resolving and outputting an emission source data optimization result S of the iteration by the software packagek(ii) a The upper corner mark k represents the kth iteration;
step 4c, optimizing the emission source data SkSubstituting the air quality prediction model with the air quality prediction model to obtain new air quality prediction data;
step 4d, the new air quality forecast data is utilized to execute the steps 4a to 4c again, and the iteration is carried out for a plurality of times in a circulating mode until the difference between the air quality forecast data and the air quality monitoring data meets the set condition, and the inversion result S is obtained1。
3. The method of claim 2, wherein the step 4d of determining whether the difference between the forecast data and the air quality monitoring data satisfies the predetermined condition is performed by:
at the k-th iteration, the adjoint variable C obtained at the k-th and k-1-th iterations are comparedk *And Ck-1 *Judging Ck *-Ck-1 *If the absolute value is smaller than the set threshold value, if so, judging that the difference between the air quality forecast data and the air quality monitoring data meets the set condition, exiting the circulation, and obtaining an optimization result S of the emission source datak-1As the emission source data S1。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910935248.XA CN110824107B (en) | 2019-09-29 | 2019-09-29 | Air quality forecasting method based on dynamic inversion emission source data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910935248.XA CN110824107B (en) | 2019-09-29 | 2019-09-29 | Air quality forecasting method based on dynamic inversion emission source data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110824107A true CN110824107A (en) | 2020-02-21 |
| CN110824107B CN110824107B (en) | 2023-04-07 |
Family
ID=69548482
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910935248.XA Active CN110824107B (en) | 2019-09-29 | 2019-09-29 | Air quality forecasting method based on dynamic inversion emission source data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110824107B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170124461A1 (en) * | 2015-10-28 | 2017-05-04 | International Business Machines Corporation | Air quality forecast by adapting pollutant emission inventory |
| CN106651036A (en) * | 2016-12-26 | 2017-05-10 | 东莞理工学院 | Air quality forecasting system |
| CN109685281A (en) * | 2018-12-29 | 2019-04-26 | 中科三清科技有限公司 | Pollution source prediction technique, device and electronic equipment |
| CN110334438A (en) * | 2019-07-04 | 2019-10-15 | 北京思路创新科技有限公司 | A kind of air pollutant emission inventory inversion method and equipment |
-
2019
- 2019-09-29 CN CN201910935248.XA patent/CN110824107B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170124461A1 (en) * | 2015-10-28 | 2017-05-04 | International Business Machines Corporation | Air quality forecast by adapting pollutant emission inventory |
| CN106651036A (en) * | 2016-12-26 | 2017-05-10 | 东莞理工学院 | Air quality forecasting system |
| CN109685281A (en) * | 2018-12-29 | 2019-04-26 | 中科三清科技有限公司 | Pollution source prediction technique, device and electronic equipment |
| CN110334438A (en) * | 2019-07-04 | 2019-10-15 | 北京思路创新科技有限公司 | A kind of air pollutant emission inventory inversion method and equipment |
Non-Patent Citations (9)
| Title |
|---|
| ROBIN L. DENNIS等: "The next generation of integrated air quality modeling: epa"s models-3", 《ATMOSPHERIC ENVIRONMENT》 * |
| SANDE HROUWELING等: "Inverse modelingof methanesourcesand sinksusingthe adjoint of a global transport model", 《JOURNAL OF GEOPHYSICAL RESEARCH》 * |
| 刘峰，黄顺祥著: "《大气环境风险控制的优化理论与应用》", 30 June 2011 * |
| 刘峰等: "大气化学传输模式 CAMx 的伴随模式: 构建及应用", 《北京大学学报(自然科学版) 》 * |
| 刘峰等: "用伴随方法对毒气泄漏事件进行危害评估", 《中国安全科学学报》 * |
| 李禾: "新技术提高空气污染预报精度", 《科技日报》 * |
| 程兴宏等: "SO2、NO2加密观测对排放源反演及空气质量预报影响的数值模拟研究", 《2007年中国气象学会年会论文集》 * |
| 郭少冬等: "基于伴随方程和MCMC方法的室内污染源反演模型研究", 《应用基础与工程科学学报》 * |
| 黄顺祥等: "基于伴随方法的大气污染溯源", 《中国科学》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110824107B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104049538B (en) | Method and apparatus for compensating the simulation model based on First Principle | |
| CN110795780B (en) | XGboost algorithm-based cable-stayed bridge finite element correction method | |
| CN110889079A (en) | Air quality forecasting system based on dynamic inversion emission source data | |
| CN109446696B (en) | CMAQ model-based rapid atmospheric environment capacity measuring and calculating method, storage medium and terminal | |
| US20060224534A1 (en) | Method and apparatus for training a system model with gain constraints using a non-linear programming optimizer | |
| CN110716512A (en) | Environmental protection equipment performance prediction method based on coal-fired power plant operation data | |
| Yang et al. | Simulation-based transfer function modeling for transient analysis of general queueing systems | |
| KR20140024541A (en) | System for providing atmospheric modeling system service | |
| Akpinar et al. | Naive forecasting of household natural gas consumption with sliding window approach | |
| US20060100721A1 (en) | Non-linear model with disturbance rejection | |
| CN109240090A (en) | A kind of adaptive soft-sensor modeling method of the incremental learning XGBOOST model based on time difference | |
| CN113281229B (en) | Multi-model self-adaptive atmosphere PM based on small samples 2.5 Concentration prediction method | |
| CN103150581B (en) | Based on the boiler optimization method and apparatus of least square method supporting vector machine combustion model | |
| CN114112995A (en) | Aerosol optical characteristic data assimilation method and device based on three-dimensional variational technology | |
| CN114611280A (en) | Ozone comprehensive source analysis method based on OBM and EBM multi-mode | |
| CN109871642B (en) | Temperature field modeling method for internal structure of steam generator | |
| CN111160667A (en) | Method and device for improving robustness of food safety prediction model | |
| CN110824107B (en) | Air quality forecasting method based on dynamic inversion emission source data | |
| CN101206727A (en) | Data processing apparatus, data processing method data processing program and computer readable medium | |
| Bell et al. | Accounting for observation uncertainty and bias due to unresolved scales with the Schmidt-Kalman filter | |
| CN116151469A (en) | Model for forecasting air quality | |
| CN115705510A (en) | Factory gas pollution distribution prediction method and system, electronic equipment and storage medium | |
| CN115081200B (en) | Acceleration factor and failure boundary domain analysis method for complex equipment | |
| Zhai | Dynamic duration of load models | |
| KR101525544B1 (en) | Development of an anthropogenic emissions spatial processing system on windows |
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 |