CN109523066A - A kind of newly-increased mobile site site selecting method of the PM2.5 based on Kriging regression - Google Patents
A kind of newly-increased mobile site site selecting method of the PM2.5 based on Kriging regression Download PDFInfo
- Publication number
- CN109523066A CN109523066A CN201811270277.0A CN201811270277A CN109523066A CN 109523066 A CN109523066 A CN 109523066A CN 201811270277 A CN201811270277 A CN 201811270277A CN 109523066 A CN109523066 A CN 109523066A
- Authority
- CN
- China
- Prior art keywords
- variance
- newly
- grid
- kriging
- monitoring station
- 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
Classifications
-
- 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"
- G06Q10/043—Optimisation of two dimensional placement, e.g. cutting of clothes or wood
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of newly-increased mobile site site selecting method of the PM2.5 based on Kriging regression, including carrying out preliminary analysis to primary data, including between the PM2.5 concentration data that monitoring station has been arranged pretreatment, website distance statistics and survey region grid division, using each grid as an alternative website；It chooses several data that monitoring station has been set at survey region a certain moment and chooses best variation function model and calculated；The predicted value for calculating the PM2.5 concentration of each grid calculates the Kriging estimation error variance of each grid, calculates the Kriging estimation error variance of all alternative mesh stations in survey region and divided rank is compared；Calculating increases the overall region variance after different grades of alternative mesh stations newly.The present invention, which utilizes, encrypts monitoring station, carries out addressing optimization from the angle of newly-increased mobile site, estimates more accurate PM2.5 concentration information using less monitoring station, more efficiently carries out prevention and control of air pollution work.
Description
Technical field
The invention belongs to geographical information space analysis technical fields, and it is newly-increased to be related to a kind of PM2.5 based on Kriging regression
Mobile site site selecting method.
Background technique
When being monitored management to city PM2.5 concentration near the ground, need to obtain PM2.5 concentration by ground monitoring station
Data are simultaneously analyzed, and are controlled management for PM2.5 concentration and are provided reliable basis.But as atmospheric environment protection is contradictory increasingly
It is prominent and to PM2.5 research there is an urgent need to, monitoring station net itself the shortcomings that also increasingly appear.Under ideal conditions, it monitors
Point is more, and distribution is wider, and the information reflected is more complete, but the laying of PM2.5 monitoring station, operation and
Maintenance needs to pay a large amount of human and material resources, financial resources, therefore can not be distributed monitoring station to each corner in city.And
Most of present PM2.5 monitoring station is artificial random addressing, and monitoring station distribution is excessively sparse to lead to that area cannot be obtained
The key message in domain, or excessively intensively lead to information redundancy in a certain piece of region website distribution, so that the sky of monitoring station net
Between layout lack it is scientific.Therefore, PM2.5 monitoring station addressing Optimization Work becomes more and more important.It at present can only be by website cloth
Office optimizes to symbolize the atmosphere quality of whole region, and the position of current stationary monitoring website is to immobilize
, to be optimized using the data of existing fixed station to monitoring station layout, can position to newly-increased website into
Row is researched and analysed, and optimal newly-increased site location is sought, to improve the interpolation precision of monitoring station net entirety.
In conclusion analysis may be tied by carrying out interpolation analysis using information acquired in current PM2.5 monitoring station
Fruit causes large error, the interpolation precision of whole monitoring station network can be improved by increasing website newly, original method cannot be sought
The optimum position of newly-increased mobile site is found, therefore proposes a kind of new website optimization method.
Relevant references are as follows:
[1]Sophocleous M,Paschetto J E,Olea R A.Ground‐Water Network Design
for Northwest Kansas,Using the Theory of Regionalized Variables[J]
.Groundwater,1982,20(1):48-58.
[2]Cesare L D,Myers D E,Posa D.Estimating and modeling space-time
correlation structures[J].Statistics&Probability Letters,2001,51(1):9-14.
[3] the coastal enterprise of Huang Zhongwei Yulin City atmosphere environment supervision spot optimization research [J] and science and technology, 2004,
(S1):42-44.
[4] Duan Yusen, Wei Haiping, Huang Rong wait the Shanghai City TSP-Pb air quality monitoring spot optimization to study [J] environment
Science and management, 2007,23 (9): 131-134.
[5] the space-time Kriging regression of the three provinces in the northeast of China Li Sha, Shu Hong, Xu Zhengquan monthly total precipitation studies [J] hydrology,
2011,31(3):31-35.
[6]Shen Y,Wu Y.Optimization of marine environmental monitoring sites
in the Yangtze River estuary and its adjacent sea,China[J].Ocean&Coastal
Management,2013,73(73):92-100.
[7] Qin Yiwen, Qian Yu, Rong Tingting are based in monitoring location addressing optimizing research [J] of Atmospheric Characteristics pollutant
State's environmental science, 2015,35 (4): 1056-1064.
[8] Liu Sha spatial data and the analysis method of space-time data and compared with [D] Chang An University, 2015.
[9] Ni Minjie is studied based on the bay monitoring water environment network optimization of Kriging regression method --- with Quanzhou, Fujian
[D] Xiamen University for gulf, 2015.
[10] survey position optimization method [J] of Xuan Teng, Li Jinhui, Li Dianqing, Song Lei based on common Kriging technique is military
Chinese college journal (engineering version), 2016,49 (05): 714-719+739.
Summary of the invention
It is an object of the invention to the Limited Number for stationary monitoring website in city and distribution lacks science, if
It is only analyzed in the data basis that these websites return, this problem of large error may be caused to analysis result, mentioned
A kind of PM2.5 based on Kriging regression increases mobile site site selecting method newly out.Technical solution of the present invention provide it is a kind of based on gram
In the PM2.5 of golden interpolation increase mobile site site selecting method newly, comprising the following steps:
Step a carries out preliminary analysis to primary data, including to the pre- of the PM2.5 concentration data that monitoring station has been arranged
The division of distance statistics and survey region grid between processing, website, using each grid as an alternative website；
Step b, calculates golden variation function in common gram, and monitoring has been arranged including choosing the several of survey region a certain moment
The data of website are simultaneously chosen best variation function model and are calculated；
Step c, calculates the predicted value of the PM2.5 concentration of each grid, and calculates the Kriging estimation error of each grid
Variance calculates the Kriging estimation error variance of all alternative mesh stations in survey region by the principle of error variance and draws
Graduation is compared；
Step d, calculating increase the overall region variance after different grades of alternative mesh stations newly, more whole variance and its
The relationship of corresponding newly-increased site level, reaches verifying effect.
Moreover, the implementation of the step b is as follows,
Assuming that regionalized variable Z (x) meets intrinsic hypothesis and second-order stationary it is assumed that its mathematic expectaion is m, covariance function
C (h) and variation function γ (h) exist, and are expressed as follows,
E [Z (x)]=m
C (h)=E [Z (x) Z (x+h)]-m2
In formula, h indicates the space length between regionalized variable；
Kriging regression method is by estimation region variable z (x) in website x to be inserted0The value of position is come approximate close to true
Value, predictor formula is as follows,
In formula, Z*(x0) it is estimated value after interpolation, Z (xi) it is sampled value, total sample number n, λiIt is each space sample point
Weight coefficient, weight coefficient λiCalculating must satisfy following two condition,
①Z*(xi) it is Z (xi) unbiased estimator；
2. estimate variance is minimum,
Estimate variance is expressed as follows with covariance function,
In formula, c (x0,x0) indicate mesh stations x to be inserted0Covariance, c (xi,xj) indicate the association for having between monitoring station
Variance, c (xi,x0) indicate existing monitoring station and mesh stations x to be inserted0Between covariance；
Keep estimate variance minimum, according to lagrange's method of multipliers, enables the Lagrangian of hypothesisIn formula, μ is Lagrange multiplier,For estimate variance, enable F to λiIt is equal to 0 with the partial derivative of μ,
Golden equation group is as follows in obtaining gram,
After arrangement formula (4) it is as follows,
Moreover, step c implementation is as follows,
Under the conditions of existing for the variation function, according to the relational expression of covariance function c (h) and variation function γ (h): c (h)
=c (0)-γ (h), golden equation group and Kriging estimation variance are as follows in common gram indicated with variation function:
Golden equation group in common gram:
Kriging estimation error variance:
In formula, γ (xi,xj) it is existing variation function between monitoring station i and j, γ (xj,x0) it is existing monitoring station
J and mesh stations x to be inserted0Between variation function, μ is Lagrange's multiplier；In being obtained gram by golden equation group in transformation gram
Golden equation matrix, matrix form such as (7) is shown,
K λ=D
Solving equations (7), find out weight coefficient λiWith Lagrange's multiplier μ, substitute into formula (5) and (6), find out estimated value with
Estimate variance.
Moreover, the implementation of the step d is as follows,
In order to verify the correctness of newly-increased mobile site, to the Kriging estimation error of the grid in the upper right corner in survey region
Variance is divided into 1,2,3,4 four grade according to absolute value from small to large, chooses website respectively in each grade and is increased newly；
It never chooses after grid is used as newly-increased website and the newly-increased grid of calculating with the region of variance grade and entirely studies respectively
The average Kriging estimation error variance of region grid increases the entire research area after the grid of different variance grades newly by comparison
The common Kriging estimation error variance in domain show that in the variance higher grade where newly-increased grid, newly-increased rear region is averaged
With regard to smaller, the Ke Lijin by primary data by grid in ordinary kriging interpolation rear region estimates the variance of Kriging estimation error
The size for counting the variance absolute value of error is directly proportional to the priority of newly-increased mobile site in survey region.
The present invention utilizes the method that encrypts to monitoring station, selective analysis on the basis of existing stationary monitoring website, from
It is excellent to carry out newly-increased bus station position to Nanchang urban district PM2.5 monitoring station net for newly-increased two angles of mobile site and newly-increased fixed station
Change, so that more accurate PM2.5 concentration information is estimated using less monitoring station, to more efficiently carry out atmosphere pollution
Preventing and controlling.The present invention proposes to be that survey region finds optimal newly-increased mobile site using ordinary kriging interpolation method, as a result
Have certain science, the precision of prediction of monitoring station net can be improved.
Detailed description of the invention
Fig. 1 is the flow chart of the embodiment of the present invention；
Fig. 2 is the flow chart of Kriging regression method of the present invention；
Fig. 3 is the variation function model curve figure of the embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawings and examples, the present invention is furture elucidated, it should be understood that embodiment be merely to illustrate the present invention and
It is not used in and limits the scope of the invention, after the present invention has been read, those skilled in the art is to various equivalences of the invention
The modification of form falls within the application range as defined in the appended claims.
It is some concepts related to the present invention below:
Space statistics are interaction and the changing rule that will have between the practical judgment of geospatial information feature
As research object, in conjunction with the computing technique of statistics and Modern Graphic, by more intuitive way, to disclose spatial data
The a series of data characteristicses such as middle implied spatial model, spatial distribution and steric interaction.In the statistics of address,
It is not mutually indepedent existing for thinking the relation on attributes between geological phenomenon in a certain range, but has certain phase
Guan Xing, not only the distance dependent between sample, the relative direction also between sample are related for this correlation.
Regionalized variable, geostatistics scientific terms.When as soon as spatial distribution is presented in variable, referred to as region
Change.This variable often reflects certain space's speciality, with regionalized variable come the phenomenon that description, referred to as compartmentalization phenomenon.Also
Claim compartmentalization stochastic variable.
Regionalized variable tool is there are two significant feature: randomness and structural.For randomness, regionalized variable is
One random function has part, random, abnormal property；And for structural, regionalized variable have it is general or
The numerical value Z (x) of average structural property, i.e. variable at point x and deviation x space length are the numerical value Z (x+h) at the point x+h of h
With autocorrelation to a certain degree, this auto-correlation depends on the distance vector h and characteristics of variables of point-to-point transmission.
Technical solution for a better understanding of the present invention, following example is with the newly-increased shifting really based on Kriging regression
Dynamic bus station position experiment is to illustrate specific embodiment.Referring to Fig. 1, method that embodiment provides the following steps are included:
1 data preliminary analysis
The preliminary analysis of data includes three portions of division of distance statistics and region grid between the pretreatment of data, website
Point.Locate in advance in embodiment including obtaining PM2.5 concentration, the PM2.5 concentration data data of the 9 state's control monitoring stations in Nanchang urban district
Reason, the distance of 9 state's control monitoring stations calculates and Nanchang urban area grid dividing.
In embodiment step 1 specific implementation the following steps are included:
1.1 data prediction
Before carrying out ordinary kriging interpolation analysis to data, it is necessary to carry out pre-processing to data, including to having set
The division for setting distance statistics and survey region grid between the pretreatment of the PM2.5 concentration data of monitoring station, website, will be each
Grid is as an alternative website.In the data for determining moment monitoring station return without less than 0 or in very big situation, select
The PM2.5 concentration data near the ground of on 01 23rd, 2015 11:00 of 9, Nanchang urban district atmosphere environment supervision website is taken, specific number
According to as shown in table 1 below.
1 monitoring station PM2.5 concentration value of table
1.2 website distance analysis
Due to Nanchang urban district altogether include between 9 atmospheric monitoring websites and website maximum distance be 25km, so to change
During different function is calculated, this 9 websites are all participated in the calculating of variation function, and are united by website distance
Meter.The distribution of this 9 websites for entire survey region have good representativeness, according to the data of this 9 websites into
Row calculates the preferable variation function of fitting effect that can be obtained.
In the calculating process to variation function, website distance is classified, taking step-length herein is that 4km classifies,
Totally 7 class, it may be assumed that
h≤4km
4km < h≤8km
8km < h≤12km
12km < h≤16km
16km≤h≤20km
20km < h≤24km
24km < h≤28km
Since the current website in Nanchang has 9, so needing to calculate distance 36 between website, concrete outcome such as the following table 2
It is shown.
2 Nanchang urban district air monitoring station point distance statistics table of table
According to results of measuring, distance is divided into 7 sections from 0-28km, specific data are as shown in table 3 below.
Apart from overview between 3 website of table
As can be seen from the table, for the urban district of Nanchang, the distance and point between 9 atmospheric monitoring websites are between number
Relationship can reflect out preferable representativeness.
In conclusion can be using the data of the 23 days 01 month Nanchang 11:00 urban district in 2015,9 atmospheric monitoring websites to south
Prosperous urban district carries out interpolation analysis.
1.3 region grid partitions
Before carrying out Nanchang urban district newly-increased bus station position, need to carry out grid to Nanchang urban district whole region, it will
Each grid is researched and analysed as an alternative website, and the centre coordinate of these grid is the coordinate of alternative website.
The longitude range of the mesh region marked off herein is 115.7 ° -116.0 °, and latitude scope is 28.5 ° -28.8 °,
31 × 31 small grid are evenly dividing into, the size of each small grid is about 1km × 1km, why by grid
It is according to the practical region area in Nanchang urban district and the suitable distance of Kriging regression algorithm that approximate size, which is set as 1km × 1km,
Etc. combined factors consider and determine.
After designing grid, grid is numbered, is convenient for subsequent searching website.Since the grid in one, the lower left corner
Number, is incremented by upwards in turn in first row, until after traversing each column, it can the number of specific grid and the longitude and latitude letter of grid
Breath is mapped.Part grid number and coordinate information are as shown in table 4 below.
4 part grid of table number and coordinate information statistical form
2 calculate golden variation function in common gram
In step 2, golden variation function in common gram is calculated, website is controlled in 9 states including choosing the Nanchang urban district a certain moment
Data are simultaneously chosen best variation function model and are calculated；
Kriging regression method in referring to fig. 2, the present invention propose step 2 implementation:
Assuming that regionalized variable Z (x) meets intrinsic hypothesis and second-order stationary it is assumed that its mathematic expectaion is m, covariance function
C (h) and variation function γ (h) exist, it may be assumed that
E [Z (x)]=m
C (h)=E [Z (x) Z (x+h)]-m2
In formula, h indicates that the space length between regionalized variable, E indicate mathematic expectaion, and m indicates constant.
Kriging regression method is the dependency structure based on stochastic variable on room and time and establishes.Its basic thought
It is: by estimation region variable z (x) in mesh stations x to be inserted0The value of position is next approximate close to true value, and predictor formula is formula
(1)。
In formula, Z*(x0) it is estimated value after interpolation, Z (xi) it is sampled value, total sample number n, λiIt is each space sample point
Weight coefficient, i=1,2 ..., n, weight coefficient λiCalculating must satisfy following two condition:
①Z*(xi) it is Z (xi) unbiased estimator, i.e.,
When, have
2. estimate variance is minimum, that is, Z*(xi) and Z (xi) difference quadratic sum it is minimum.I.e.
It is minimum.
Estimate variance is represented by formula (2) with covariance function.
In formula, c (x0,x0) indicate the covariances of mesh stations to be inserted, c (xi,xj) indicate in existing monitoring station i-th
Covariance between j-th of website, c (xi,x0) indicate the covariance for having between monitoring station and mesh stations to be inserted.
Keep estimate variance minimum, according to lagrange's method of multipliers, enablesIn formula, F is vacation
If Lagrangian, μ is Lagrange multiplier,For estimate variance.
Enable F to λiIt is equal to 0 with the partial derivative of μ, golden equation group (3) in obtaining gram
Formula (4) are obtained after arrangement
In embodiment, chooses 9 states control station data at Nanchang urban district a certain moment and choose best variation function model
It is calculated.After classifying to website distance, the rule of thumb Computing Principle of variation function model, it is necessary to calculate website
Between PM2.5 concentration data near the ground horizontal increment, horizontal increment such as the following table 5 institute of each website PM2.5 concentration near the ground
Show.
5 Nanchang urban district website PM2.5 concentration level accrual accounting list position of table: ug/m3
Tab.5 Statistical of increase of PM2.5 concentration level in
nanchang urban site unit:ug/m3
In calculating process, determined according to the difference degree between calculated variation function value and the function model of fitting
The fixed variation function model finally used, by the calculated variation function cloud atlas of calculation formula institute of experimental variations functional value with
Spherical model coincide the most, so choosing spherical model using primary data to be more to close to survey region progress interpolation analysis
It is suitable, according to data point fitting and continuous adjusting parameter obtained in the case where comprehensively considering variation function property
The variation function of near-optimization.Variation function formula indicates are as follows:
γ (h)=537 × [1.5 (h/30) -0.5 (h/30)3]
Variation function model curve figure is as shown in Figure 3.
Golden variance calculates in 3 grams
In step 3, the predicted value of the PM2.5 concentration of each grid is calculated, and the Kriging estimation for calculating each grid misses
Poor variance.The Kriging estimation error variance of all alternative mesh stations in survey region is calculated simultaneously by the principle of error variance
Divided rank is compared.
Step 3 implementation is as follows,
Under the conditions of existing for the variation function, according to the relational expression of covariance function c (h) and variation function γ (h): c (h)
=c (0)-γ (h), c (0) indicates covariance when distance between two points tend to 0 in formula, is definite value.Indicate general with variation function
Tong Kelijin equation group and Kriging estimation variance, it may be assumed that
Golden equation group in common gram:
Kriging estimation error variance:
γ (x in formula (5)i,xj) it is that variation function between point i and j, γ (x in formula 6 are controlled in statej,x0) it is point j and station to be inserted
Variation function between point, μ is Lagrange's multiplier.Golden equation matrix in being obtained gram by golden equation group in transformation gram, matrix
Form is such as shown in (7).
K λ=D
In formula, γijIndicate existing variation function between monitoring station i and j, γ (xi, x) and indicate existing monitoring station i
With the variation function of any mesh stations x to be inserted, K is lower section γijThe matrix constituted, D are lower section γ (xi, x) constituted
Matrix, λiIt is the weight coefficient of each space sample point.
Solving equations (7), find out weight coefficient λiWith Lagrange's multiplier μ, substitute into formula (5) and (6), find out estimated value with
Estimate variance.
In embodiment, after obtaining variation function, monitoring station is controlled to each standby according to 9 states in existing Nanchang urban district
Selective calling clicks through row interpolation, and 9 states are calculated by matrix golden in common gram and control website for the weight coefficient of sample point, calculate
The predicted value of the PM2.5 concentration of each grid, and calculate the Kriging estimation error variance of each grid.Pass through error variance
Principle calculates the Kriging estimation error variance of all alternative mesh stations in survey region and divided rank is compared.
The implementation of the step 3 is as follows:
In the evaluation of monitoring station optimization, mainly monitoring station is divided according to the variance of Kriging estimation error
Cloth optimizes.
Due to providing the variance of Kriging estimation error in Kriging regression method, and this characteristic is exactly Kriging regression
Where the advantage of method, be mainly characterized by: in Interpolation Process, variance of estimaion error size only with the quantity of known sample point,
Position distribution is related with variant structure, unrelated with the data value of known sample point.Golden variance can indicate in gram are as follows:
In formula, λiFor the weight of each space sample point, γ (xi,x0) it is between existing monitoring station i and mesh point to be inserted
Variation function, μ are Lagrange's multiplier.
That is, the distribution of point is more reasonable when the variance of Kriging estimation error becomes smaller, it can be from research area
The information that monitoring network obtains is also more.
4 interpretations of result and verifying
In step 4, calculating increases the overall region variance after different grades of alternative mesh stations newly, more whole variance with
Its relationship for corresponding to newly-increased site level, reaches verifying effect.
In embodiment, step 4 includes the following steps,
The analysis of 4.1 Kriging estimation error variances
It is carried out by the PM2.5 concentration data on 01 23rd, 2015 11:00 of 9, Nanchang urban district atmospheric monitoring website general
Logical Kriging regression and after calculating estimation error variance, by calculate each alternative mesh stations gram in golden error variance import
ArcGIS and in generating gram golden error variance isogram, specific picture and text step always provides in substantive examination reference.
4.2 result verification
In order to verify the correctness of newly-increased mobile site, to the Kriging estimation error of the grid in the upper right corner in survey region
Variance is divided into 1,2,3,4 four grade according to absolute value from small to large, chooses website respectively in each grade and is increased newly, specifically
Different grades of regional distribution chart is increased newly always to provide in substantive examination reference.
When it is implemented, software technology, which can be used, in the above process realizes automatic running.
Specific embodiment described herein is only an example for the spirit of the invention.The neck of technology belonging to the present invention
The technical staff in domain can make various modifications or additions to the described embodiments or replace by a similar method
In generation, however, it does not deviate from the spirit of the invention or beyond the scope of the appended claims.
Claims (4)
1. a kind of PM2.5 based on Kriging regression increases mobile site site selecting method newly, which comprises the following steps:
Step a, to primary data carry out preliminary analysis, including to the PM2.5 concentration data that monitoring station has been arranged pretreatment,
The division of distance statistics and survey region grid between website, using each grid as an alternative website；
Step b, calculates golden variation function in common gram, and monitoring station has been arranged including choosing the several of survey region a certain moment
Data and choose best variation function model and calculated；
Step c, calculates the predicted value of the PM2.5 concentration of each grid, and calculates the Kriging estimation error variance of each grid,
Calculate the Kriging estimation error variance of all alternative mesh stations in survey region by the principle of error variance and divide etc.
Grade is compared；
Step d, calculating increase the overall region variance after different grades of alternative mesh stations newly, and more whole variance is corresponding
The relationship of newly-increased site level reaches verifying effect.
2. a kind of PM2.5 based on Kriging regression as described in claim 1 increases mobile site site selecting method newly, feature exists
In: the implementation of the step b is as follows,
Assuming that regionalized variable Z (x) meets intrinsic hypothesis and second-order stationary it is assumed that its mathematic expectaion is m, covariance function c (h)
And variation function γ (h) exists, and is expressed as follows,
E [Z (x)]=m
C (h)=E [Z (x) Z (x+h)]-m2
In formula, h indicates the space length between regionalized variable；
Kriging regression method is by estimation region variable z (x) in website x to be inserted0The value of position is next approximate close to true value, in advance
It is as follows to survey formula,
In formula, Z*(x0) it is estimated value after interpolation, Z (xi) it is sampled value, total sample number n, λiIt is the power of each space sample point
Weight coefficient, weight coefficient λiCalculating must satisfy following two condition,
①Z*(xi) it is Z (xi) unbiased estimator；
2. estimate variance is minimum,
Estimate variance is expressed as follows with covariance function,
In formula, c (x0,x0) indicate mesh stations x to be inserted0Covariance, c (xi,xj) indicate the covariance for having between monitoring station,
c(xi,x0) indicate existing monitoring station and mesh stations x to be inserted0Between covariance；
Keep estimate variance minimum, according to lagrange's method of multipliers, enables the Lagrangian of hypothesisIn formula, μ is Lagrange multiplier,For estimate variance, enable F to λiIt is equal to 0 with the partial derivative of μ,
Golden equation group is as follows in obtaining gram,
After arrangement formula (4) it is as follows,
3. a kind of PM2.5 based on Kriging regression as described in claim 1 increases mobile site site selecting method newly, feature exists
In: step c implementation is as follows,
Under the conditions of existing for the variation function, according to the relational expression of covariance function c (h) and variation function γ (h): c (h)=c
(0)-γ (h), golden equation group and Kriging estimation variance are as follows in common gram indicated with variation function:
Golden equation group in common gram:
Kriging estimation error variance:
In formula, γ (xi,xj) it is existing variation function between monitoring station i and j, γ (xj,x0) be existing monitoring station j and to
Insert mesh stations x0Between variation function, μ is Lagrange's multiplier；Jin Fangcheng in being obtained gram by golden equation group in transformation gram
Matrix, matrix form such as (7) is shown,
Solving equations (7), find out weight coefficient λiWith Lagrange's multiplier μ, substitutes into formula (5) and (6), find out estimated value and estimation
Variance.
4. a kind of PM2.5 based on Kriging regression as described in claims 1 or 2 or 3 increases mobile site site selecting method newly,
Be characterized in that: the implementation of the step d is as follows,
In order to verify the correctness of newly-increased mobile site, to the Kriging estimation error variance of the grid in the upper right corner in survey region
It is divided into 1,2,3,4 four grade from small to large according to absolute value, chooses website respectively in each grade and increased newly；
Grid never is chosen as entire survey region after newly-increased website and the newly-increased grid of calculating with the region of variance grade respectively
The average Kriging estimation error variance of grid increases the entire survey region after the grid of different variance grades newly by comparison
Common Kriging estimation error variance obtains, in the variance higher grade where newly-increased grid, in average gram for increasing rear region newly
With regard to smaller, the Kriging estimation by primary data by grid in ordinary kriging interpolation rear region misses golden variance of estimaion error
The size of the variance absolute value of difference is directly proportional to the priority of newly-increased mobile site in survey region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811270277.0A CN109523066B (en) | 2018-10-29 | 2018-10-29 | PM2.5 newly-added mobile station address selection method based on Kriging interpolation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811270277.0A CN109523066B (en) | 2018-10-29 | 2018-10-29 | PM2.5 newly-added mobile station address selection method based on Kriging interpolation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109523066A true CN109523066A (en) | 2019-03-26 |
| CN109523066B CN109523066B (en) | 2023-06-20 |
Family
ID=65773110
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811270277.0A Active CN109523066B (en) | 2018-10-29 | 2018-10-29 | PM2.5 newly-added mobile station address selection method based on Kriging interpolation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109523066B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110210681A (en) * | 2019-06-11 | 2019-09-06 | 西安电子科技大学 | A kind of prediction technique of the monitoring station PM2.5 value based on distance |
| CN110766191A (en) * | 2019-08-27 | 2020-02-07 | 东华理工大学 | Newly-added PM2.5 fixed monitoring station site selection method based on space-time kriging interpolation |
| CN110990993A (en) * | 2019-10-21 | 2020-04-10 | 天津大学 | Method for estimating and designing ore body reserves based on Kriging method |
| CN112035553A (en) * | 2020-09-02 | 2020-12-04 | 中国电建集团成都勘测设计研究院有限公司 | Hydrological design value calculation method based on space statistics theory |
| CN112305163A (en) * | 2020-10-21 | 2021-02-02 | 西北工业大学 | Atmospheric pollution monitoring system based on fixed-wing unmanned aerial vehicle and data processing method |
| CN112434935A (en) * | 2020-11-20 | 2021-03-02 | 中国电波传播研究所（中国电子科技集团公司第二十二研究所） | Selectable PM2.5 concentration estimation method |
| CN116048106A (en) * | 2023-02-08 | 2023-05-02 | 海南大学 | Ocean observation control method and device for self-adaptive sampling of multi-underwater robot |
| CN117129638A (en) * | 2023-10-26 | 2023-11-28 | 江西怡杉环保股份有限公司 | Regional air environment quality monitoring method and system |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080167822A1 (en) * | 2004-08-03 | 2008-07-10 | Abhl | Climatic Forecast System |
| CN102521403A (en) * | 2011-12-26 | 2012-06-27 | 南京成风大气信息技术有限公司 | Refined meteorological information service system |
| CN103353923A (en) * | 2013-06-26 | 2013-10-16 | 中山大学 | Self-adaption spatial interpolation method and system based on spatial feature analysis |
| CN104252549A (en) * | 2013-06-28 | 2014-12-31 | 中国石油天然气股份有限公司 | Well spacing analysis method based on Kriging interpolation |
| CN104714257A (en) * | 2015-01-29 | 2015-06-17 | 哈尔滨工程大学 | Local geomagnetic map construction method based on multifractal kriging interpolation of gradual interpolation correction |
| JP2015161591A (en) * | 2014-02-27 | 2015-09-07 | Kddi株式会社 | Observed value processing device |
| CN105550784A (en) * | 2016-01-20 | 2016-05-04 | 中科宇图科技股份有限公司 | Distribution point optimizing method of air quality monitoring station |
| CN105678085A (en) * | 2016-01-12 | 2016-06-15 | 环境保护部卫星环境应用中心 | PM2.5 concentration estimation method and system |
| JP2017010125A (en) * | 2015-06-17 | 2017-01-12 | 日本電信電話株式会社 | Device, method and program for estimating air pollutant aspiration amount |
| CN106372277A (en) * | 2016-05-13 | 2017-02-01 | 新疆农业大学 | Variation function model optimization method in forest site index spatial-temporal estimation |
| CN106407633A (en) * | 2015-07-30 | 2017-02-15 | 中国科学院遥感与数字地球研究所 | Method and system for estimating ground PM2.5 based on space-time regression Kriging model |
| CN107315722A (en) * | 2017-07-03 | 2017-11-03 | 南京大学 | Based on gram in gold and the hydrographic(al) network Optimized model that is combined of information entropy theory |
| CN107391951A (en) * | 2017-08-03 | 2017-11-24 | 江苏卓易信息科技股份有限公司 | A kind of air pollution source tracing method based on the sequence of annular neighborhood gradient |
| WO2017210273A1 (en) * | 2016-05-31 | 2017-12-07 | Accuweather, Inc. | Method and system for predicting the impact of forecasted weather, environmental and/or geologic conditions |
| CN108062454A (en) * | 2018-01-19 | 2018-05-22 | 宁波市镇海规划勘测设计研究院 | Pollutant spatial and temporal distributions uncertainty characteristic analysis method, system and storage medium |
-
2018
- 2018-10-29 CN CN201811270277.0A patent/CN109523066B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080167822A1 (en) * | 2004-08-03 | 2008-07-10 | Abhl | Climatic Forecast System |
| CN102521403A (en) * | 2011-12-26 | 2012-06-27 | 南京成风大气信息技术有限公司 | Refined meteorological information service system |
| CN103353923A (en) * | 2013-06-26 | 2013-10-16 | 中山大学 | Self-adaption spatial interpolation method and system based on spatial feature analysis |
| CN104252549A (en) * | 2013-06-28 | 2014-12-31 | 中国石油天然气股份有限公司 | Well spacing analysis method based on Kriging interpolation |
| JP2015161591A (en) * | 2014-02-27 | 2015-09-07 | Kddi株式会社 | Observed value processing device |
| CN104714257A (en) * | 2015-01-29 | 2015-06-17 | 哈尔滨工程大学 | Local geomagnetic map construction method based on multifractal kriging interpolation of gradual interpolation correction |
| JP2017010125A (en) * | 2015-06-17 | 2017-01-12 | 日本電信電話株式会社 | Device, method and program for estimating air pollutant aspiration amount |
| CN106407633A (en) * | 2015-07-30 | 2017-02-15 | 中国科学院遥感与数字地球研究所 | Method and system for estimating ground PM2.5 based on space-time regression Kriging model |
| CN105678085A (en) * | 2016-01-12 | 2016-06-15 | 环境保护部卫星环境应用中心 | PM2.5 concentration estimation method and system |
| CN105550784A (en) * | 2016-01-20 | 2016-05-04 | 中科宇图科技股份有限公司 | Distribution point optimizing method of air quality monitoring station |
| CN106372277A (en) * | 2016-05-13 | 2017-02-01 | 新疆农业大学 | Variation function model optimization method in forest site index spatial-temporal estimation |
| WO2017210273A1 (en) * | 2016-05-31 | 2017-12-07 | Accuweather, Inc. | Method and system for predicting the impact of forecasted weather, environmental and/or geologic conditions |
| CN107315722A (en) * | 2017-07-03 | 2017-11-03 | 南京大学 | Based on gram in gold and the hydrographic(al) network Optimized model that is combined of information entropy theory |
| CN107391951A (en) * | 2017-08-03 | 2017-11-24 | 江苏卓易信息科技股份有限公司 | A kind of air pollution source tracing method based on the sequence of annular neighborhood gradient |
| CN108062454A (en) * | 2018-01-19 | 2018-05-22 | 宁波市镇海规划勘测设计研究院 | Pollutant spatial and temporal distributions uncertainty characteristic analysis method, system and storage medium |
Non-Patent Citations (4)
| Title |
|---|
| 卢月明: "面向大气污染指数分析的时空协同克里金插值方法", 《中国优秀博硕士学位论文全文数据库（硕士） 工程科技I辑》 * |
| 卢月明: "面向大气污染指数分析的时空协同克里金插值方法", 《中国优秀博硕士学位论文全文数据库（硕士） 工程科技I辑》, 15 October 2018 (2018-10-15), pages 2 - 3 * |
| 邓岳川等: "克里金插值的多路径误差空间分布特征", 《测绘科学》 * |
| 邓岳川等: "克里金插值的多路径误差空间分布特征", 《测绘科学》, 30 April 2018 (2018-04-30), pages 17 - 22 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110210681B (en) * | 2019-06-11 | 2023-06-27 | 西安电子科技大学 | Prediction method of PM2.5 value of monitoring station based on distance |
| CN110210681A (en) * | 2019-06-11 | 2019-09-06 | 西安电子科技大学 | A kind of prediction technique of the monitoring station PM2.5 value based on distance |
| CN110766191A (en) * | 2019-08-27 | 2020-02-07 | 东华理工大学 | Newly-added PM2.5 fixed monitoring station site selection method based on space-time kriging interpolation |
| CN110990993A (en) * | 2019-10-21 | 2020-04-10 | 天津大学 | Method for estimating and designing ore body reserves based on Kriging method |
| CN112035553A (en) * | 2020-09-02 | 2020-12-04 | 中国电建集团成都勘测设计研究院有限公司 | Hydrological design value calculation method based on space statistics theory |
| CN112035553B (en) * | 2020-09-02 | 2022-11-29 | 中国电建集团成都勘测设计研究院有限公司 | Hydrological design value calculation method based on space statistics theory |
| CN112305163A (en) * | 2020-10-21 | 2021-02-02 | 西北工业大学 | Atmospheric pollution monitoring system based on fixed-wing unmanned aerial vehicle and data processing method |
| CN112434935A (en) * | 2020-11-20 | 2021-03-02 | 中国电波传播研究所（中国电子科技集团公司第二十二研究所） | Selectable PM2.5 concentration estimation method |
| CN112434935B (en) * | 2020-11-20 | 2022-04-22 | 中国电波传播研究所（中国电子科技集团公司第二十二研究所） | Selectable PM2.5 concentration estimation method |
| CN116048106A (en) * | 2023-02-08 | 2023-05-02 | 海南大学 | Ocean observation control method and device for self-adaptive sampling of multi-underwater robot |
| CN116048106B (en) * | 2023-02-08 | 2024-04-12 | 海南大学 | Ocean observation control method and device for self-adaptive sampling of multi-underwater robot |
| CN117129638A (en) * | 2023-10-26 | 2023-11-28 | 江西怡杉环保股份有限公司 | Regional air environment quality monitoring method and system |
| CN117129638B (en) * | 2023-10-26 | 2024-01-12 | 江西怡杉环保股份有限公司 | Regional air environment quality monitoring method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109523066B (en) | 2023-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109523066A (en) | A kind of newly-increased mobile site site selecting method of the PM2.5 based on Kriging regression | |
| Ali et al. | Spatial–temporal characterization of rainfall in Pakistan during the past half-century (1961–2020) | |
| Chen et al. | High‐resolution gridded population projections for China under the shared socioeconomic pathways | |
| Sigaroodi et al. | Long-term precipitation forecast for drought relief using atmospheric circulation factors: a study on the Maharloo Basin in Iran | |
| Du et al. | Different flooding behaviors due to varied urbanization levels within river Basin: A case study from the Xiang river Basin, China | |
| Wu et al. | Estimating the probability of compound floods in estuarine regions | |
| CN112990976A (en) | Commercial network site selection method, system, equipment and medium based on open source data mining | |
| Masseran et al. | On spatial analysis of wind energy potential in Malaysia | |
| CN104392113B (en) | A kind of evaluation method of COASTAL SURFACE cold reactive antibodies wind speed | |
| Akçay et al. | Wind speed forecasting by subspace and nuclear norm optimization based algorithms | |
| Gao et al. | Using storm transposition to investigate the relationships between hydrologic responses and spatial moments of catchment rainfall | |
| Zhu et al. | Ridership prediction of urban rail transit stations based on AFC and POI data | |
| Xing et al. | Sustainable development evaluation of urban traffic system | |
| CN103150616A (en) | Digging method for distribution characteristics of prehistoric settlement site | |
| CN115239027A (en) | Method and device for air quality lattice ensemble prediction | |
| Xiao et al. | Impact of digital elevation model resolution on stream network parameters | |
| Wu et al. | Comparison analysis of sampling methods to estimate regional precipitation based on the Kriging interpolation methods: A case of northwestern China | |
| Wei et al. | Spatial–temporal evolution pattern and prediction analysis of flood disasters in China in recent 500 years | |
| Liu et al. | Estimating Beijing's travel delays at intersections with floating car data | |
| Cheng et al. | Runoff simulation by SWAT model based on parameters transfer method in ungauged catchments of middle reaches of Jialing River | |
| JP5864753B2 (en) | Physical quantity estimation system and physical quantity estimation method | |
| Wang et al. | Harnessing big data for sustainable urban management: A novel approach to gridded urban GDP dataset development | |
| Niu et al. | Research on dynamic simulation and prediction of urban expansion based on SLEUTH model | |
| Zhang et al. | Development of a global land suitability dataset for cultivation based on physiogeographic factors | |
| CN109947877A (en) | A kind of method and system improving GIS mobile terminal Orientation on map precision |
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 |