CN109118004A - A kind of engineer construction addressing Suitable Area prediction technique - Google Patents
A kind of engineer construction addressing Suitable Area prediction technique Download PDFInfo
- Publication number
- CN109118004A CN109118004A CN201810936211.4A CN201810936211A CN109118004A CN 109118004 A CN109118004 A CN 109118004A CN 201810936211 A CN201810936211 A CN 201810936211A CN 109118004 A CN109118004 A CN 109118004A
- Authority
- CN
- China
- Prior art keywords
- factor
- addressing
- suitable area
- total number
- earth
- 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"
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Educational Administration (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 relates to the technical fields of engineer construction addressing Suitable Area prediction, more particularly to a kind of engineer construction addressing Suitable Area prediction technique, the step that the step of the step of including the steps that the step of influence index factor divides, each influence index factor divides correlation analysis and the selection of grid cell size, weighted information amount calculates and information content threshold value determine;The step of weighted information amount calculates, the information content including calculating each influence index factor;According to the weight of the information content and geologic elements reduction attribute, the weighted information amount of each influence index factor is calculated;All influence index factor Comprehensive weight information contents that each grid cell includes are calculated according to the weighted information amount.Present invention combination spatial weighting is theoretical, and the weight that Fuzzy and Rough set method is determined is added in conventional information amount model, and precision of prediction is greatly improved.
Description
Technical field
The present invention relates to the technical fields of engineer construction addressing Suitable Area prediction, and in particular to a kind of engineer construction addressing is suitable
Suitable area prediction technique.
Background technique
Forefathers are predicted it is proposed that excessively a variety of methods, such as Delphi method, analytic hierarchy process (AHP) in engineer construction addressing Suitable Area
(AHP), Fuzzy AHP (FAHP), entropy weight theory method, fuzzy mathematics and rough set theory method, evidence-right-weight " method, nerve
Network technique, particle swarm algorithm etc..There are many deficiencies in these methods, especially evaluation criterion weight distribution is asked in practical applications
Topic protrudes, and needs to reform improvement, this is also exactly the starting point of the invention.Forefathers' research method is combed, level point is broadly divided into
Analyse correlation technique, Probability & Statistics method, fuzzy mathematics and rough set theory method, machine learning method, space predicting method
Etc. classifications.
1) step analysis correlation technique and its deficiency
Delphi method is earliest Weight Determination, to the importance measures of decision influence index in management,
It is quoted afterwards by Other subjects, and is modified into chromatographic assays, the objectivity of Weight Determination is improved.
For Delphi method in index for selection or determining index weights, only with the knowledge and experience of brainstrust, it is objective fixed to lack
The measurement of amount, and often correlation is big, relevance is strong for the index chosen, and there are problems that assigning weight overlapping.
The prediction of engineer construction addressing Suitable Area, is related to many fuzzy targets, classical mathematics theory has been difficult to describe.With
It then will lead to weight sequencing result when AHP method operation and deviation occur, it is fuzzy to measure that fuzzy mathematics theory is introduced AHP method thus
Index, the result error of improved AHP method.
Step analysis correlation technique can not handle the index of incompleteness, and the inside that can not also refine between data is closed
System, so being difficult to be suitable for different Geological Environment Engineerings.
2) Probability & Statistics method and its deficiency
The prediction of engineer construction addressing Suitable Area can refer to and utilize built engineering when carrying out weight index selection
Evaluation index, to improve the reasonability and science that engineer construction addressing prediction evaluation index weight determines, for this purpose, scholars tie
It closes probability theory and statistical knowledge has invented entropy weight theory, evidence weight equal weight determines method.
Evidence weight, entropy weight theory are by data-driven, although preferably avoiding the subjectivity of weight setting, weight evidence
The considerations of dual mode of weight method lacks to known goal in research scale and differentiation, entropy weight theory are not suitable for evaluation index value then
Change very little or the case where the becoming smaller that become larger suddenly.
3) fuzzy mathematics and rough set theory method and its deficiency
Fuzzy mathematics and rough set theory are favored because of its advantage in terms of handling fuzzy and uncertain problem, but its
The information system of incompleteness can not be handled.Rough set theory then require processed data must be it is discrete, cannot handle
The data of continuous type.
4) machine learning method and its deficiency
The machine learning methods such as neural network, particle swarm algorithm are determined applied to engineer construction addressing evaluation criterion weight
Research, it is desirable that sample size is more, the quality of data is high, this is less in comparison and choice survey data, and there are when a large amount of qualitative datas,
The weight of each index can not be reasonably determined.
5) space predicting method and its deficiency
Since the last century 60's, American-European countries researchs and proposes space predicting method.But in space predicting method
Multivariate logistic regression method, artificial neural network method require that sample size is more, quality of data height, and be only applicable to specific sky
Between forecasting research field.
When the Factors Weighting addition method is predicted for engineer construction addressing Suitable Area, determined according to actual prospecting data
Threshold value, belongs to the qualitative scope for deriving and dividing, and subjective consciousness is stronger.
Conventional information amount method is equal weight superposition, realizes that there is no consider to sky in the prediction of engineer construction addressing Suitable Area
Between goal in research have controlling influence evaluation index contribution, be thus exaggerated space research target is occurred without or
The contribution for the evaluation index that rarer controlling influences.
2, fuzzy coarse central is theoretical
Fuzzy set and rough set support the mankind and indicate, handle, calculating incomplete and uncertain information theoretical frame, and two
The influence of person has apparent embodiment in human engineering application field.Although early stage is it is found that fuzzy set with rough set is mutually auxiliary phase
At relationship, rather than competitive relation;But the advantage of both the similitude of both early utilizations concept and combination goes to form one
A completely new application of mixture theory but encounters very big obstruction.Nevertheless, to the application of mixture theory of fuzzy set and rough set --- mould
The pionerring research for pasting rough set theory is risen in the 1990s and initial stage in 2000;Recently, interdisciplinary research is also from mould
Paste collection and the big important computations normal form of rough set two it is universal in the theoretical basis being benefited, and helped fuzzy coarse central perfect.Cause
This, more close to the fuzzy coarse central of the thoughtcast of the mankind in the letter for handling fuzzy, uncertainty, continuity, incompleteness
It is widely applied in breath system.
1) Fuzzy and Rough decision table
If by the object x of domain UiSee alternative address, conditional attribute collection A={ a1, a2, a3, in element afRegard as position,
The influence factors such as landform, traffic, decision attribute d regard suitable addressing degree as, then Fuzzy and Rough decision table such as table 1.
1 Fuzzy and Rough decision table of table
As can be seen from Table 1, addressing conclusion D is exactly the fuzzy set on domain U:
2) fuzzy equivalence relation class
It is concentrated in Fuzzy and Rough, by the lower and upper approximations concept for giving expression to Incomplete information and there is indistinguishability
Equivalence relation R expand to fuzzy equivalence relation class as follows, we can be with much between similitude, that is, object between evaluation object
Degree is similar, rather than the indistinguishability of object obscuring element.Thus it can define fuzzy equivalence relation class:
It defines 1: setting S as the fuzzy equivalence relation on U, [x]sFor fuzzy equivalence relation class, then have:
Enable fuzzy equivalence relation class [x]sFor F, then fuzzy equivalence relation class has following property:
(1)
(2)uF(x)∧us(x, y)≤uF(y);
(3)uF(x)∧uF(y)≤uS(x, y)
3) upper bound and lower bound is obscured
Define 2: setting (U, P) is Fuzzy Approximation Space, FiFor the fuzzy equivalence relation class for belonging to U/P,It is close under fuzzy P
Sihe obscures approximation on P and is defined as follows:
Wherein, X is about the positive domain of (U, P)PThe boundary of X, X areThis definition and clearly upper lower aprons
Collection is slightly different, because the approximation of each object is indefinite, it is possible to redefine are as follows:
In calculating process, not every y ∈ U requires to consider, only considers those uF(y) be non-zero y object, this
When object y be fuzzy equivalence relation class F an obscuring element.Referred to as fuzzy coarse central.By defining 2 aforementioned four public affairs
Formula can be seen that when whole equivalence classes is all clear, this definition becomes traditional rough set.At this point, considering clear lower close
As membership function:
This shows that it belongs to the P lower aprons of X if an object x belongs to the equivalence class of an X subset, obscures lower close
As characteristic it is identical with being clearly defined in clear situation.Fuzzy lower aprons can be rewritten are as follows:
Wherein, → it is known as fuzzy implication operator.Clearly, uF(x) and uX(x) value is 0 or 1, therefore, when
When at least one object fully belongs to F rather than X in its equivalence class F, u P XIt (x) is zero, this is complete with the definition of clear lower aprons
It is exactly the same.Equally, definition approximate on P can also be rewritten, makes it have the meaning actually calculated.
4) property of fuzzy coarse central
The negative domain of rough set, positive domain and Boundary Region can indicate with fuzzy membership functions, all elements of positive region
Degree of membership be 1, the degree of membership of borderline region element is 0.5, those degrees of membership for being included in negative region element are zero.Cause
This, it is necessary to allow the element in borderline region to have within the scope of 0-1 and be subordinate to angle value, rather than just 0.5.If rough set is
X, equivalence relation R, then fuzzy coarse central has following property:
(1)uY(R(X))=1;
(2)
(3)
5) obscures positive domain degree of membership and Fog property dependency degree
It defines 3: obscuring the degree of membership of positive field object x ∈ U are as follows:
When only the equivalence class belonging to the object x is the component part of positive region, it can just belong to positive region.
Definition 4: Fog property dependency degree function are as follows:
Fog property dependency degree, which corresponds to, to be determinedFuzzy cardinality divided by object in domain sum.It utilizes
Fog property dependency degree can establish the theoretical basis of geological property weight distribution.
3, information computation
The concept of information content is derived from the communications field, " the A Mathematical Theory created by Shannon in 1948
Of Communication " it is proposed in a text first.The main theory of information content is derived from Probability Statistics Theory and random process.Letter
Breath amount is introduced into disaster assessment of easy generation in the 1980's and risk spatial prediction is studied.With the hair of GIS and remote sensing technology
Exhibition, is widely used in the spatial prediction of other field.The core concept of information computation (IVM) is with space research target
According to influence index, the information content of each influence index factor segmentation is calculated by information content formula, uses the overlay analysis of GIS
A possibility that each influence index space overlapping gross information content of function calculating, evaluation space goal in research occurs.
Assuming that the impacted index x of space research target YiThe influence of (i=1,2,3 ..., n), xiTo space research target
Influence degree is different.Under specific geological conditions, always there is the combination of some influence indexs to will lead to space research mesh
Target occurs or occurs, and therefore, the target of information computation is by the general of the available space research target for having occurred or having occurred
Rate come determine combination in influence index information content.Information content is defined as:
Wherein, P (Y, x1, x2..., xn) it is known as x1, x2..., xnInfluence index combines the item that down space goal in research occurs
Part probability, the probability that P (Y) representation space goal in research occurs, I (Y, x1, x2..., xn) it is that influence index combines x1, x2..., xn
The information content that space research target is provided.According to condition probability formula, (formula 1) can be rewritten into (formula 2):
Wherein,Expression has an impact index x1In the presence of, influence index x2To extraterrestrial target provide information content,Expression has an impact indicator combination x1, x2..., xn-1In the presence of, influence index xnTo extraterrestrial target contribution
Information content.
Since the influence index for causing space research target to occur and the combination of corresponding influence index are very more, in order to true
Determine statistical sample and calculate information content, the establishment process of model is divided into three steps:
Firstly, calculating single influence index xiTo the information content of space research target contribution, formula is as follows:
Wherein, P (xi| Y) it is the index x that has an impact when space research target occursiProbability, P (xi) indicate in survey area
Middle influence index xiThe probability of appearance.However, spatial prediction is carried out on the basis of grid matrix, therefore, with sample frequency
Rate replaces many and diverse K-theoretic operation K in practical application, it may be assumed that
Wherein, S is survey area grid cell total number, and N is the grid list for occurring space research target in survey area
First total number, SiIt indicates to contain influence index x in survey areaiGrid cell total number, NiThere is space research target in expression
Contain influence index x in regioniGrid cell total number.
Second step is the gross information content I for calculating each influence indexi, formula is as follows:
Third step is to indicate that each grid cell may with by the calculated integrated traffic of GIS spatial overlay analysis
It will appear the degree of space research target.Ii< 0 indicates that the grid cell probability of space research target occurs lower than enumeration district
There is space research destination probability, I in being averaged for domaini=0 probability for indicating that space research target occurs in the grid cell is equal to tune
Look into the average probability in region, IiThe probability that > 0 indicates that space research target occurs in the grid cell is equal to being averaged for survey area
Probability.This shows that integrated traffic is bigger, and the probability for space research target occur is bigger.
Summary of the invention
The purpose of the present invention is to provide a kind of engineer construction addressing Suitable Area prediction techniques, theoretical in conjunction with spatial weighting,
The weight that Fuzzy and Rough set method is determined is added in conventional information amount model, and precision of prediction is greatly improved.
In order to reach above-mentioned technical purpose, the technical solution adopted in the present invention is as follows:
A kind of engineer construction addressing Suitable Area prediction technique, the prediction technique successively the following steps are included:
The step of the step of influence index factor divides, each influence index factor divide correlation analysis and grid cell are big
The step of small selection, it is characterised in that: further include the steps that the step that weighted information amount calculates and information content threshold value determines;
The step of weighted information amount calculates, comprising:
The earth's surface engineering delimited according to the total number of survey region cell, engineer construction addressing qualitative model or casual labour
Journey constructs addressing Suitable Area cell total number, the cell sum in survey region containing geologic elements reduction attribute factor
Mesh and earth's surface engineering or temporary project construct the cell total number that geologic elements reduction attribute factor is contained in addressing Suitable Area,
To calculate the information content of each influence index factor in geologic elements reduction attribute;
The earth's surface engineering or interim determined according to the information content of each influence index factor and by fuzzy coarse central method
The weight of engineering geology element reduction attribute, calculates the weighted information amount of each influence index factor;
Had an impact according to what the weighted information amount of each influence index factor calculated that each grid cell includes
Index factor Comprehensive weight information content;
The step that the information content threshold value determines, comprising:
Using some division points of jenks nature breakpoint method as engineer construction addressing Suitable Area information content threshold value.
Further, in the geologic elements reduction attribute information content of each influence index factor calculation formula are as follows:
Wherein, S indicates that the total number of survey region cell, N indicate the earth's surface that engineer construction addressing qualitative model delimited
Engineering or temporary project construct addressing Suitable Area cell total number, SiIt indicates to contain geologic elements reduction attribute in survey region
Factor xiCell total number, NiIndicate that earth's surface engineering or temporary project construct addressing Suitable Area and contain geologic elements reduction category
Sex factor xiCell total number.
Further, the calculation formula of the weighted information amount of each influence index factor are as follows:
I(xi, Y) and=wiln(Ni/N)/(Si/S)
Wherein, S indicates that the total number of survey region cell, N indicate the earth's surface that engineer construction addressing qualitative model delimited
Engineering or temporary project construct addressing Suitable Area cell total number, SiIt indicates to contain geologic elements reduction attribute in survey region
Factor xiCell total number, NiIndicate that earth's surface engineering or temporary project construct addressing Suitable Area and contain geologic elements reduction category
Sex factor xiCell total number, wiIndicate the earth's surface engineering determined by fuzzy coarse central method or temporary project geologic elements about
The weight of simple attribute.
Further, the calculating for all influence index factor Comprehensive weight information contents that each described grid cell includes is public
Formula are as follows:
Wherein, S indicates that the total number of survey region cell, N indicate the earth's surface that engineer construction addressing qualitative model delimited
Engineering or temporary project construct addressing Suitable Area cell total number, SiIt indicates to contain geologic elements reduction attribute in survey region
Factor xiCell total number, NiIndicate that earth's surface engineering or temporary project construct addressing Suitable Area and contain geologic elements reduction category
Sex factor xiCell total number, wiIndicate the earth's surface engineering determined by fuzzy coarse central method or temporary project geologic elements about
The weight of simple attribute;
Ii< 0 indicates that the grid cell probability of space research target occurs and occurs space lower than being averaged for survey area and grind
Study carefully destination probability, Ii=0 indicates that average probability of the probability equal to survey area of space research target, I occurs in the grid celli
> 0 indicates that average probability of the probability equal to survey area of space research target occurs in the grid cell.
The invention has the following beneficial effects:
Information computation is improved using fuzzy coarse central theory.The shortcomings that for the superposition of conventional information amount model equal weight,
It is theoretical in conjunction with spatial weighting for the contribution for embodying the index for thering is controlling to influence space research target, by Fuzzy and Rough
The weight that set method determines is added in conventional information amount model, and precision of prediction is greatly improved.
Based on rough set difference matrix geologic elements attribute selection and optimization method, based on the geology of fuzzy coarse central
On the basis of element reduction attribute weight determines method, propose that the engineer construction addressing of fuzzy coarse central improvement information computation is suitable for
Area's prediction technique solves the defect of information computation equal weight superposition, embodies between influence index to space research target
The otherness that influence degree occurs, realizes the high-precision forecast of engineer construction addressing Suitable Area.It preferably meets intricately
The demand of engineer construction addressing Suitable Area quick predict under matter environmental condition correctly implements engineer construction addressing for administrative staff and determines
Plan provides the foundation of science.
Detailed description of the invention
Fig. 1 is flow chart of the invention;
Fig. 2 is the wide pasture landforms schematic diagram of the embodiment of the present invention;
Fig. 3 is the Changjiang Village and advance village Suitable Area of the embodiment of the present invention;
Fig. 4 is the village Tan Dian, the town Yong Dian, kiln village, the village Liu Jiadian Suitable Area of the embodiment of the present invention;
Fig. 5 is the Weathering Degree of Rock Mass weighted information amount distribution map of the embodiment of the present invention;
Fig. 6 is the rock mass discontinuity Density Weighted information content distribution map of the embodiment of the present invention;
Fig. 7 is the rock mass discontinuity combination degree weighted information amount distribution map of the embodiment of the present invention;
Fig. 8 is the rock mass integrity index weighted information amount distribution map of the embodiment of the present invention;
Fig. 9 is the rock mass point load testing mean weighted information amount distribution map of the embodiment of the present invention;
Figure 10 is the rock mass saturated uniaxial compressive strength weighted information amount distribution map of the embodiment of the present invention;
Figure 11 is rock mass basic quality's index weighted information amount distribution map of the embodiment of the present invention;
Figure 12 is the rock mass hardness weighted information amount distribution map of the embodiment of the present invention;
Figure 13 is the land-based area water quality grade weighted information amount distribution map of the embodiment of the present invention;
Figure 14 is the land-based area water body water grade weighted information amount distribution map of the embodiment of the present invention;
Figure 15 is the groundwater level depth weighted information amount distribution map of the embodiment of the present invention;
Figure 16 is the geological disaster gradient weighted information amount distribution map of the embodiment of the present invention;
Figure 17 is the stock number weighted information amount distribution map of the embodiment of the present invention;
Figure 18 is the shield structure depth of stratum weighted information amount distribution map of the embodiment of the present invention;
Figure 19 is the underground engineering geologic elements reduction attribute synthesis weighted information amount distribution map of the embodiment of the present invention;
Figure 20 is that the underground engineering of the embodiment of the present invention constructs addressing Suitable Area;
Figure 21 is the gravelly soil weighted information amount distribution map of amount containing gravel of the embodiment of the present invention;
Figure 22 is the soil bearing capacity average value weighted information amount distribution map of the embodiment of the present invention;
Figure 23 is the soil body hardness weighted information amount distribution map of the embodiment of the present invention;
Figure 24 is the rock mass thickness of earth covering weighted information amount distribution map of the embodiment of the present invention;
Figure 25 is the Weathering Degree of Rock Mass weighted information amount distribution map of the embodiment of the present invention;
Figure 26 is the land-based area water quality grade weighted information amount distribution map of the embodiment of the present invention;
Figure 27 is the land-based area water body water grade weighted information amount distribution map of the embodiment of the present invention;
Figure 28 is the geological disaster gradient weighted information amount distribution map of the embodiment of the present invention;
Figure 29 is the stock number weighted information amount distribution map of the embodiment of the present invention;
Figure 30 is the earth's surface engineering or temporary project geologic elements reduction attribute synthesis weighted information of the embodiment of the present invention
Measure distribution map;
Figure 31 is that the earth's surface engineering of the embodiment of the present invention or temporary project construct addressing Suitable Area;
Figure 32 is that the underground engineering based on information computation prediction of the embodiment of the present invention constructs addressing Suitable Area;
Figure 33 constructs addressing for the underground engineering based on the prediction of Factors Weighting Additive Model of the embodiment of the present invention
Area;
Figure 34 a is the ROC curve of the IVM of the embodiment of the present invention;
Figure 34 b is the ROC curve of the FWAM of the embodiment of the present invention;
Figure 34 c is the ROC curve of the IVM-FRS of the embodiment of the present invention.
Specific embodiment
Below by specific embodiment combination attached drawing, the present invention will be described in detail, it should be noted that in the feelings not conflicted
Under condition, the feature in embodiment and embodiment in the present invention be can be combined with each other, and the scope of protection of the present invention is not limited thereto.
Limitation existing for the prediction technique of addressing Suitable Area is constructed for Traditional project, is proposed a kind of using fuzzy coarse central
Improve the engineer construction addressing Suitable Area prediction technique of information computation.It is folded that space overlapping in information computation belongs to equal weight
Add, the disadvantage is that easily reducing the contribution for the influence index for thering is controlling to influence space research target, and amplifies to sky
Between goal in research occur it is smaller or without controlling influence influence index contribution.In conjunction with spatial weighting theory, using fuzzy
Rough set improves information computation, can assign certain power when allowing to carry out GIS spatial overlay analysis to each influence index
Again it is spatial weighting, to embody the otherness for space research target occurring between influence index influence degree, makes engineer construction
The prediction of addressing Suitable Area is more scientific, reasonable, accurate.As shown in Figure 1, a kind of engineer construction addressing Suitable Area prediction technique, it should
Prediction technique successively the following steps are included:
Step S01: the influence index factor divides
Fuzzy coarse central improve information computation require to carry out the influence index of space research target state demarcation, that is, because
Son divides, so that influence index removes dimension factor, coequally participates in spatial overlay analysis.Influence index generally falls into qualitative category
Property two class of index and qualitative attribute index, these two types of indexs can be subdivided into three kinds of variable scales: continuous scale, ordinal scale and
Nominal mean power.For nominal mean power variable, the qualitative description according to nominal mean power variable can carry out factor division to it, and one
Kind feature or a kind of factor of status representative.For ordinal scale variable, the precedence categories according to ordinal scale variable can be right
It carries out factor division, and a kind of precedence categories represent a kind of factor.For continuous scale variable, data have dimension and numerical value is
Continuously, it is necessary to its continuum be divided to obtain the state interval factor.
The factor of influence index divides the result for affecting spatial weighting overlay analysis.With rock mass point load testing mean
As an example, sliding-model control is carried out to rock mass point load testing mean using equidistant discretization method, can be classified as
Tetra- sections 0-3.385,3.385-6.77,6.77-10.155,10.155-13.54 i.e. four factors, if this four factor pairs
Engineer construction addressing suitability influence difference degree is similar, then this factor division cannot correctly embody influence index to engineering structure
The influence of addressing is built, the factor divides unreasonable;If it is significant to influence difference degree, show that factor division has fully demonstrated influence and referred to
Target influences, and the factor divides reasonable.So fuzzy coarse central improve information computation to rock mass point load testing mean this
When the continuous scale variable of class carries out factor division, method of the reasonability of demarcation interval depending on its discretization.Continuous scale becomes
The factor of amount can be divided by the methods of equidistant discretization method, the heuristic discretization method of Nguyen, can also pass through experience
Method, statistic law divide.The factor division numbers of continuous scale variable are appropriate, and can sufficiently show different because subinterval is to engineering
Construct the influence difference of addressing suitability.
Since the practicability that fuzzy coarse central improves information computation prediction result depends on the reasonability of factor division, because
It is necessary to assess division methods for this.Factor information amount is its measure index to the contribution of engineer construction addressing suitability, difference
It is bigger, show that different factor pair engineer construction addressing suitability contribution differences are bigger.Here, Probability & Statistics theory can be borrowed
In standard deviation concept, using the standard deviation between factor information amount as the evaluation index for measuring division rule quality, good stroke
Divider then often makes the engineer construction addressing within the scope of influence index be suitable for that normal distribution is presented in points for investigation, utilizes the two sides
Method can assess the reasonability of division methods.
Step S02: each influence index factor divides correlation analysis
According to conditional probability, information computation requires non-correlation between each influence index after factor division.One good
Although factor division methods can embody the difference of different factor pair engineer construction addressing suitabilities contribution, not can be shown that
The mutual non-correlation of influence index after division.So to each influence index factor divide after, it is necessary to each influence index into
Row correlation analysis, with eligible probability to the correlation requirement of influence index.
IVM-FRS carries out correlation analysis to each influence index using GIS software, and factor involved in analysis refers to influence
Mark correlation has a significant impact, according to the requirement of information computation, it is necessary to be the influence index after the factor divides.For example,
Expert's experience have shown that the thickness of earth covering of rock mass and the point load testing mean of rock mass do not have any correlation, if carrying out shadow
It rings and all only divides two attributes into a factor when index factor divides, then the point load of the thickness of earth covering of rock mass and rock mass is surveyed
Trying average value division result will be just the same, this shows that the correlation of two attributes is consistent.Spatial Analyst in ArcGIS
The Band Collection Statistics that tool provides can analyze the correlation of each influence index, Band
Collection Statistics requires that each influence index distribution map is first switched to grid map by ArcToolbox crossover tool
Layer, the correlation of each influence index after then analysis factor divides, and generate correlation matrix.If the number in matrix is approximately equal to
1, show the two influence indexs correlation;If the number in matrix is approximately equal to 0, show that the two influence indexs are independent.
Step S03: grid cell size selection
The division of grid cell size is a step of IVM-FRS model key, is the basis of spatial overlay analysis, and
Carry out the basis of engineer construction addressing Suitable Area prediction, the reasonability divided directly affects the precision of influence index superposition, into
And influence the precision of engineer construction addressing Suitable Area.It is suitable for that the grid cell of size can fully demonstrate influence index factor pair engineering
The percentage contribution of addressing is constructed, the effect that will not ignore the influence index factor will not be both exaggerated.Grid cell divides too small, meeting
Increase the calculation amount of ArcGIS software, reduces computational efficiency, and may result in part geologic elements unit and be separated, nothing
Body of laws shows geologic elements unit to the mass action of engineer construction addressing.Grid cell division is excessive, may obscure each influence
The percentage contribution of index factor reduces the precision of engineer construction addressing Suitable Area prediction.
Basic statistics unit of the grid cell as spatial overlay analysis, forefathers have carried out systematic Study to it, and
It sums up and is suitable for the empirical equation that spatial prediction grid cell divides:
GS=7.49+0.0006S-2.0 × 10-9S2+2.9×10-15S3 (8)
In formula, GSIndicate to be suitable for grid cell size, S indicates the denominator of original contour line data precision.
Step S04: weighted information amount calculates
It includes three steps that weighted information amount, which calculates:
Step 1: the information content of each influence index factor in geologic elements reduction attribute is calculated according to the following formula:
Wherein, S indicates that the total number of survey region cell, N indicate the earth's surface that engineer construction addressing qualitative model delimited
Engineering or temporary project construct addressing Suitable Area cell total number, SiIt indicates to contain geologic elements reduction attribute in survey region
Factor xiCell total number, NiIndicate that earth's surface engineering or temporary project construct addressing Suitable Area and contain geologic elements reduction category
Sex factor xiCell total number.
Step 2: the weighted information amount of each influence index factor is calculated according to the following formula:
I(xi, Y) and=wiln(Ni/N)/(Si/S) (6)
Wherein, wiIndicate the power of the earth's surface engineering or temporary project geologic elements reduction attribute that are determined by fuzzy coarse central method
Weight.
Step 3: all influence index factor aggregative weighteds letter that each grid cell includes is calculated according to the following formula
Breath amount:
This step is realized by ArcGIS software raster overlay function, ultimately forms engineer construction addressing aggregative weighted information
Spirogram.Formula 6 and formula 7 are the core formula of the improvement information computation (IVM-FRS) based on fuzzy coarse central, according to this two
A formula can divide the grade of space research target generation.
In step 1 calculate the influence index factor information content when need to pay attention to a bit, if some factor of influence index or
Without engineer construction addressing Production Zones Suitable in person's combinations of factors, then it will appear the feelings that molecule is 0 when calculating information content with formula 4
Condition, the i.e. information content of the factor are negative infinite.It bears infinite information content and other influences index factor information content and carries out space overlapping
When can Comprehensive weight information content be always bear infinite situation, thus can offset other influences index to engineer construction addressing be suitable for
The contribution of property, and this has factor for bearing infinite information content itself less important.Infinite information is born therefore it is necessary to eliminate
Amount.In information computation, the minimum value of other factor information amounts of influence index is used to bear the infinite information content factor as this
Information magnitude.
Step S05: information content threshold value determines
Threshold definitions are the level thresholds that must be over when generating given effect or result.When a threshold is exceeded, in system
The state change of essence can occur for portion, and this variation would generally occur suddenly.A kind of certainty viewpoint is implied in this definition: being
The state of system can be predicted by comparing input value or one group of input value and threshold value, in addition be implied: when system future shape
When state development is not related to randomness, given input will have single possible output (being higher or lower than threshold value).Information content threshold
Comprehensive weight information content must be over when value is defined as generating engineer construction addressing Production Zones Suitable in Comprehensive weight information content figure
Value.Information content threshold value is determined using the method based on Probability & Statistics.When conventional method is there are when the shortcomings that high subjective,
Jenks nature breakpoint method provides a kind of objective method to determine the threshold value under complex situations.Jenks nature breakpoint optimizes
It is a kind of to be intended to determine that spatial data value to the data clustering method of different classes of optimal layout, owns by reducing every class to the greatest extent
The deviation of value and this classification average value, while the mean deviation of each class Yu other classes is improved to the maximum extent.In other words,
This method is intended to reduce the variance inside classification, and maximizes the variance between classification.It can be seen that from its core mathematics theory
Jenks nature breakpoint method divides Comprehensive weight information content classification based on probability distribution and statistical law and selects drawing between classification
Branch;Therefore, jenks nature breakpoint method overcomes the artificial subjectivity according to survey data selection information content threshold value, so that letter
The selection of breath amount threshold value is more objective, scientific, reasonable.
If Comprehensive weight information content can be divided into Ganlei with jenks nature breakpoint method by ArcGIS software.Root
According to jenks nature breakpoint method core concept, every a kind of information magnitude with adjacent category division points can make engineer construction addressing suitable
Great-jump-forward variation occurs for suitable property.So selecting some division points of jenks nature breakpoint method suitable as engineer construction addressing
Area's information content threshold value.
A specific embodiment is set forth below and further illustrates the present invention.
1, survey region geography geological environment
As shown in Figure 1, using Kuandian as engineer construction siting analysis area.The regional landforms geological form multiplicity, mainly
Landforms geological type has tectonic denudation low mountains and hills, Gao Qiu, mountain valley and basalt tableland:
(1) tectonic denudation low-relief terrain, be distributed in the county Kuan Dian south, east most area, be wide pasture major landform it
One.Height above sea level is in East and West direction spread between 500~1000m.300~500m of relative relief, the hillside gradient are mostly 20 °~30 °.
Vegetation is overall relatively to be developed, and 60% or more average coverage rate, tree species are mostly wildwood and artificial forest.
(2) crick area, be mainly distributed on the town Chang Dian and its on the south etc. ground.Height above sea level mostly between 100~500m, substantially along
Yalu River spread, rock type are mainly the migmatite and Metamorphic Rocks From Liaohe Group of the Proterozoic Era.Mountain shape is short, the wide slow, summit in brae
It is slightly round.Hills shape is perfectly round, undulation.
(3) mountain valley is distributed in Yalu River water system two sides along the river, and height above sea level is in 200m or less.Shape is in dendroid
Strid is mainly ploughed distributed area for wide pasture.
(4) basalt tableland, only minute quantity is distributed in the northwest corner Kuan Dian, is Quarternary Volcanoes basalt tableland development area.
Extinct volcano mouth is distributed in the county Kuan Dian more than 20, and volcanic cone, volcanic crater geomorphologic landscape are that eastern Liaoning is rare.
2, geologic elements reduction attribute factor divides
There is no nominal mean power attribute in geologic elements reduction attribute, therefore is drawn without the concern for the factor of nominal mean power attribute
Point.Specific geologic elements reduction attribute factor is divided, two kinds of situations can be divided into.It is divided according to the influence index factor former
Reason, for the ordinal scale variable in geologic elements reduction attribute, such as the soil body hardness of soil body element, the rock of rock mass element
Body rate of decay, rock mass discontinuity combination degree, rock mass hardness, the water quality grade of land-based area water body element, water body water
Grade is measured, the stock number of element of resource can divide the factor and is numbered according to its precedence categories.Geologic elements screening with it is excellent
In change, grade classification has been carried out to these ordinal scale attributes, this need apply this grade classification with regard to it is achievable this
The factor of a little ordinal scale attributes divides, as shown in table 2.
2 geologic elements reduction ordinal scale attribute factor of table divides
Scale variable continuous for geologic elements reduction attribute, as the gravelly soil amount containing gravel of soil body element, bearing capacity are average
Value, the thickness of earth covering of rock mass element, structure surface density, Perfection Index, point load testing mean, rock mass single shaft anti-saturation are strong
Degree, rock mass basic quality's index, the groundwater level depth of land-based area water body element, the gradient of geological disaster element, element of resource
Shield structure depth of stratum, can be using equidistant discretization method, the heuristic discretization method of Nguyen and empirical method, statistic law
Carry out factor division.Herein by taking rock mass basic quality's index as an example, standard deviation is maximum between selective factor B information content and geology is wanted
Engineer construction addressing in plain reduction range of attributes is suitable for the factor division methods that two standards of normal distribution are presented in points for investigation.
Area's survey data of analyzing and researching is it is found that the numberical range of rock mass basic quality's index is 255.38-547.76, it
It is divided into five factors.According to equidistant discretization theory can be divided into 255.38-313.856,313.856-372.332,
372.332-430.808, five factors of 430.808-489.284,489.284-547.76;According to the heuristic discretization of Nguyen
Theory can be divided into five factors of 255.38-326,326-391,391-413,413-455,455-547.76;According to expert
Experience can be divided into five factors of 255.38-321,321-386,386-425,425-463,463-547.76;According to system
Meter law theory can be divided into five factors of 255.38-316,316-384,384-448,448-489,489-547.76;Respectively
Numbering in sequence to the above-mentioned factor is 1,2,3,4,5.The each of above-mentioned four kinds of methods division is counted according to information computation theory
The data of factor engineer construction addressing Suitable Area, and the information content of each factor is calculated according to formula 4, calculated result is shown in Table 3 and table
4.Engineer construction addressing Suitable Area data are forefathers in the Changjiang Village that engineer construction addressing qualitative model delimited, advance village underground work
Journey constructs addressing Suitable Area (as shown in Figure 3) and the village Tan Dian, the town Yong Dian, kiln village, the village Liu Jiadian earth's surface engineering or temporary project
The engineer construction addressing Suitable Area data of addressing Suitable Area (as shown in Figure 4) are constructed, these data are also that fuzzy coarse central improves
The statistical sample of information computation.
The engineer construction addressing of 3 each factor of rock mass basic quality's index of table is suitable for points for investigation statistical magnitude
The factor | 1 | 2 | 3 | 4 | 5 |
Equidistant method | 1 | 36 | 37 | 33 | 3 |
Nguyen heuristics | 15 | 25 | 28 | 18 | 24 |
Empirical method | 11 | 15 | 23 | 26 | 35 |
Statistic law | 2 | 14 | 56 | 35 | 3 |
The information content of 4 each factor of rock mass basic quality's index of table
The factor | 1 | 2 | 3 | 4 | 5 | Standard deviation |
Equidistant method | -0.673 | 0.572 | 3.285 | 2.392 | -0.861 | 2.013 |
Nguyen heuristics | -2.836 | -0.489 | 1.382 | 1.871 | 2.165 | 1.897 |
Empirical method | -1.392 | -0.237 | 1.732 | 2.136 | 3.829 | 1.502 |
Statistic law | -3.526 | -1.139 | 2.725 | 3.481 | 2.749 | 2.185 |
It is learnt from table 3, the suitable points for investigation factor distributed number of statistic law most tends to normal distribution, learns statistics from table 3
The information content standard deviation of method is maximum;So the factor that the factor division result of statistic law is chosen as rock mass basic quality's index is drawn
Point.The continuous scale variable optimum factor division result of other geologic elements reduction attributes such as table 5, is wanted since 1 to each geology
The factor serialization row number of the plain continuous scale variable of reduction attribute.
The factor of the 5 continuous scale variable of geologic elements reduction attribute of table divides
3, geologic elements reduction attribute factor divides correlation analysis
According to IVM-FRS model needs, after geologic elements reduction attribute carries out factor division, it is necessary to carry out correlation to it
Analysis, to meet the requirement of formula 6 and 7 conditional probability of formula to geologic elements reduction attribute non-correlation.Correlation analysis
Influence index there are 7 factors to divide after geologic elements reduction attribute ordinal scale variable: soil body hardness, Weathering Zones of Igneous Rock
Degree, rock mass discontinuity combination degree, rock mass hardness, water quality grade, water body water grade, stock number, there is 11
The continuous scale variable of geologic elements reduction attribute after factor division: gravelly soil amount containing gravel, soil bearing capacity average value, rock mass cover
Soil thickness, rock mass structure surface density, rock mass integrity index, rock mass point load testing mean, rock mass single shaft anti-saturation intensity,
Rock mass basic quality's index, groundwater level depth, the geological disaster gradient, shield structure depth of stratum, above-mentioned 18 geologic elements
The factor division result of reduction attribute such as table 2 and table 5.
Each geologic elements reduction attribute factor division result distribution map is switched to grid map by ArcToolbox first, then
After statisticalling analyze 18 factors divisions using the Band Collection Statistics tool in Spatial Analyst
The correlation of geologic elements reduction attribute generates the correlation matrix between each geologic elements reduction attribute, as shown in table 6.
Geologic elements reduction attribute related coefficient after 18 factors that table 6 lists divide can be seen that correlation matrix pair
The element of linea angulata is 1, i.e. the related coefficient of 18 reduction attributes itself is 1, and other elements size levels off to 0, i.e. 18 reduction
Related coefficient approach 0 between attribute, this shows that 18 geologic elements reduction attributes after factor division are independent from each other.
4, cell divides
Non-correlation between geologic elements reduction attribute after factor division, so that it may which geologic elements reduction attribute is carried out
Spatial overlay analysis.The basis of spatial overlay analysis is to carry out cell division, unit to geologic elements reduction property distribution figure
The size of lattice wants the moderate information content size and each factor that each factor of geologic elements reduction attribute is fully demonstrated with guarantee
To the percentage contribution of engineer construction addressing suitability.Use space predicted grid dividing elements empirical equation 8 to geologic elements about
Simple property distribution figure carries out cell division, and research area's geologic elements reduction property distribution primitive beginning contour precision denominator is
80000, then: GS=7.49+0.0006 × 8 × 104-2.0×10-9×64×108+2.9×10-15×512×1012≈
69.7748, convenience of calculation when being spatial overlay analysis, cell size is divided into 70 meters.
5, geologic elements reduction attribute weight information computing
Before calculating geologic elements reduction attribute weight information content, it is suitable for what engineer construction addressing qualitative model was delimited
Geologic elements investigation point data in area is randomly divided into two groups, and the training group containing 70% data is used for IVM-FRS model
Training is to calculate weighted information amount, and the validation group containing remaining 30% data is for verifying IVM-FRS model validation.
Geologic elements reduction attribute correlation matrix after the division of 6 factor of table
1) underground engineering geologic elements reduction attribute factor weighted information amount calculates
The influence index that underground engineering constructs addressing has Weathering Degree of Rock Mass, rock mass structure surface density, rock mass discontinuity knot
Conjunction degree, rock mass integrity index, rock mass point load testing mean, rock mass saturated uniaxial compressive strength, rock mass basic quality
Index, rock mass hardness, land-based area water quality grade, land-based area water body water grade, groundwater level depth, geological disaster slope
14 degree, stock number, shield structure depth of stratum geologic elements reduction attributes.Underground engineering geologic elements reduction attribute factor adds
The calculating of power information content is divided into two steps:
First is the information content that 14 each factors of geologic elements reduction attribute are calculated according to formula 4;Research area is indicated with S
The total number of domain cell, the underground engineering that N indicates that engineer construction addressing qualitative model delimited construct addressing Suitable Area cell
Total number, SiIt indicates to contain geologic elements reduction attribute factor x in survey regioniCell total number, NiIndicate underground engineering
It constructs addressing Suitable Area and contains geologic elements reduction attribute factor xiCell total number.
Second is the weighted information amount that 14 each factors of geologic elements reduction attribute are calculated according to formula 6, uses wiIt indicates
By the weight for the underground engineering geologic elements reduction attribute that fuzzy coarse central method determines.Underground engineering geologic elements reduction attribute because
Sub- weighted information amount calculated result such as table 7.
7 underground engineering geologic elements reduction attribute factor weighted information amount of table
2) earth's surface engineering or temporary project geologic elements reduction attribute factor weighted information amount calculate
The influence index that earth's surface engineering or temporary project construct addressing have gravelly soil amount containing gravel, soil bearing capacity average value,
Soil body hardness, rock mass thickness of earth covering, Weathering Degree of Rock Mass, land-based area water quality grade, land-based area water body water grade, geology
9 the disaster gradient, stock number geologic elements reduction attributes.As underground engineering, earth's surface engineering or temporary project geologic elements
The calculating of reduction attribute factor weighted information amount is also classified into two steps:
First is the information content that 9 each factors of geologic elements reduction attribute are calculated according to formula 4;Research area is indicated with S
The total number of domain cell, it is suitable that N indicates that the earth's surface engineering that engineer construction addressing qualitative model delimited or temporary project construct addressing
Suitable area cell total number, SiIt indicates to contain geologic elements reduction attribute factor x in survey regioniCell total number, NiTable
Show that earth's surface engineering or temporary project construct addressing Suitable Area and contain geologic elements reduction attribute factor xiCell total number.
Second is the weighted information amount that 9 each factors of geologic elements reduction attribute are calculated according to formula 6, uses wiIndicate by
The weight of earth's surface engineering or temporary project geologic elements reduction attribute that fuzzy coarse central method determines.Earth's surface engineering or temporary project
The results are shown in Table 8 for the calculating of geologic elements reduction attribute factor weighted information amount.The meaning of weighted information magnitude is the same as ground in table
The meaning of lower engineering Factors Weighting information magnitude.
8 earth's surface engineering of table or temporary project geologic elements reduction attribute factor weighted information amount
6, engineer construction addressing Suitable Area is predicted
Realize that the basis of engineer construction addressing Suitable Area prediction is each by each geologic elements reduction attribute of space overlapping
Factors Weighting information spirogram obtains engineer construction addressing Comprehensive weight information content figure.This can pass through ArcGIS software raster overlay
Function is realized.It is crucial that using jenks nature breakpoint method by Comprehensive weight information content be divided into it is very big, big, medium,
Small, very small five grades, according to jenks nature breakpoint method core concept, the information of very big grade and big grade classification point
Magnitude can make engineer construction addressing suitability that great-jump-forward variation occur, so taking very big range suitable for engineer construction addressing
Area.
1) underground engineering constructs the prediction of addressing Suitable Area
It influences underground engineering and constructs Weathering Degree of Rock Mass weighted information amount point in 14 geologic elements reduction attributes of addressing
Butut as shown in figure 5, rock mass discontinuity Density Weighted information content distribution map as shown in fig. 6, rock mass discontinuity combination degree weight
Information content distribution map as shown in fig. 7, rock mass integrity index weighted information amount distribution map as shown in figure 8, rock mass point load test
Average value weighted information amount distribution map is as shown in figure 9, rock mass saturated uniaxial compressive strength weighted information amount distribution map such as Figure 10 institute
Show, rock mass basic quality's index weighted information amount distribution map is as shown in figure 11, and rock mass hardness weighted information amount distribution map is such as
Shown in Figure 12, land-based area water quality grade weighted information amount distribution map is as shown in figure 13, land-based area water body water grade weighted information
Amount distribution map is as shown in figure 14, and groundwater level depth weighted information amount distribution map is as shown in figure 15, geological disaster gradient weighting letter
Breath amount distribution map is as shown in figure 16, and stock number weighted information amount distribution map is as shown in figure 17, shield structure depth of stratum weighting letter
Breath amount distribution map is shown in Figure 18.By ArcGIS software to aforementioned 14 geologic elements reduction attribute weight information content distribution map grid
Superposition obtains underground engineering geologic elements reduction attribute synthesis weighted information amount distribution map, as shown in figure 19.
Underground engineering geologic elements reduction attribute synthesis weighted information amount is divided into very using jenks nature breakpoint method
Greatly, greatly, medium, small, very small five grades, be shown in Table 9;The division points information magnitude of negated often big grade and big grade is underground
Engineer construction addressing Suitable Area threshold value, value 0.607950;Negated often big rate range constructs addressing for underground engineering and is suitable for
Area, and be indicated on the remote sensing image by ortho-rectification with red, see Figure 20.
9 underground engineering geologic elements reduction attribute synthesis weighted information amount grade classification of table
2) earth's surface engineering or temporary project construct the prediction of addressing Suitable Area
It influences earth's surface engineering or temporary project constructs 9 geologic elements reduction attribute medium stone soil amounts containing gravel weighting of addressing
Information content distribution map is shown in that Figure 21, soil bearing capacity average value weighted information amount distribution map are shown in Figure 22, soil body hardness weighting letter
Breath amount distribution map is shown in that Figure 23, rock mass thickness of earth covering weighted information amount distribution map are shown in Figure 24, Weathering Degree of Rock Mass weighted information amount point
Butut is shown in that Figure 25, land-based area water quality grade weighted information amount distribution map are shown in Figure 26, land-based area water body water grade weighted information amount
Distribution map is shown in that Figure 27, geological disaster gradient weighted information amount distribution map are shown in that Figure 28, stock number weighted information amount distribution map are shown in Figure 29.
Raster overlay is carried out to aforementioned 9 geologic elements reduction attribute weight information content distribution map by ArcGIS software, obtains earth's surface
Engineering or temporary project geologic elements reduction attribute synthesis weighted information amount distribution map, as shown in figure 30.
Using jenks nature breakpoint method by earth's surface engineering or temporary project geologic elements reduction attribute synthesis weighted information amount
Very big, big, medium, small, very small five grades are divided into, are shown in Table 10;The division points of negated often big grade and big grade are believed
Breath magnitude is that earth's surface engineering or temporary project construct addressing Suitable Area threshold value, value 1.59172;Negated often big rate range is
Earth's surface engineering or temporary project construct addressing Suitable Area, and are indicated on the remote sensing image by ortho-rectification with green, see figure
31。
10 earth's surface engineering of table or temporary project geologic elements reduction attribute synthesis weighted information amount grade classification
7, engineer construction addressing Suitable Area prediction result is evaluated
There are two types of methods for the evaluation of engineer construction addressing Suitable Area prediction result, first is that passing through comparison using contrast verification method
The Suitable Area verifying that Suitable Area prediction result and qualitative model delimited improves information computation predictive engine based on fuzzy coarse central
Construct the validity of addressing Suitable Area;Second is that using ROC curve, that is, receiver operator characteristics' curve, by comparison IVM-FRS with
Area ratio under other model ROC curves evaluates the precision of IVM-FRS.The engineer construction addressing qualitative model that this section is mentioned
The underground engineering of delimitation constructs addressing Suitable Area and refers to Changjiang Village engineer construction addressing Suitable Area, and geographical coordinate is E124 °
53′ 27.81″ -124° 56′ 15.11″,N40° 36′ 01.29″ -40° 37′ 41.31″;Advance village engineer construction choosing
Location Suitable Area, geographical coordinate: E124 ° 53 ' 17.78 " -124 ° 56 ' 05.77 ", N40 ° 30 ' 37.28 " -40 °
31 ' 44.89 ", rectangular extent is as shown in Figure 3.The earth's surface engineering or temporary project of delimitation construct addressing Suitable Area and refer to smooth pasture
It is not stated since its range is in polygon in village, the town Yong Dian, kiln village and the village Liu Jiadian engineer construction addressing Suitable Area herein
Geographical coordinate, specific range are as shown in Figure 4.
The core concept of contrast verification method is to compare engineer construction addressing Suitable Area prediction result and geologic elements points for investigation
Data verification group shows that engineering structure can be effectively predicted in IVM-FRS model if validation group data is distributed in the Suitable Area of prediction
Build addressing Suitable Area.30% geologic elements points for investigation in the Suitable Area that we at random delimit engineer construction addressing qualitative model
Data are divided into validation group, in contrast verification method, use Ni(f=1,2 ..., n) indicates that n-th of investigation point data of validation group is
It is no to predict in Suitable Area, if Ni=1, then explanation is if Ni=0 shows not exist.Then count N numerical value and with validation group number
Data bulk compares, under the conditions of existing for the allowable error, if numerical value of N and validation group data quantitative proportion are higher than 96%, i.e.,
96% points for investigation is fallen in prediction Suitable Area in validation group, then shows IVM-FRS model prediction engineer construction addressing Suitable Area
It is effective.The numerical value of N for now counting underground engineering and earth's surface engineering or temporary project respectively, as a result such as table 11.
11 engineer construction addressing Suitable Area contrast verification method evaluation table of table
Works category | Numerical value of N | Validation group data quantity | Ratio |
Underground engineering | 86 | 86 | 100% |
Earth's surface engineering or temporary project | 52 | 53 | 98.1% |
The numerical value of N of underground engineering and validation group data quantitative proportion are 100% in table 11, earth's surface engineering or temporary project
Ratio is 98.1%, more than 98%, illustrates that IVM-FRS model prediction engineer construction addressing Suitable Area is effective.So figure
20 be that research area's underground engineering constructs addressing Suitable Area, and Figure 31 is that research area's earth's surface engineering or temporary project are constructed addressing and fitted
Yi Qu.The Suitable Area range of comparison diagram 20 and Fig. 3 and Figure 31 and Fig. 4 it can be found that the underground engineering of prediction to construct addressing suitable
Yi Qu is within the scope of Changjiang Village and two, advance village matrix Suitable Area, and is in irregular polygon shape;The earth's surface engineering of prediction
Or temporary project is constructed addressing Suitable Area and is within the scope of the village Tan Dian, the town Yong Dian, kiln village and the village Liu Jiadian Suitable Area, two are pre-
It is more accurate to survey Suitable Area range boundary.
ROC curve studies the useful tool of two-dimensional problem as one, extensive whether being suitable for such as engineer construction addressing
The performance for assessing spatial prediction model obtains true positive rate (TRR) and false positive rate (FRR) by the different threshold value of setting to draw
ROC curve processed.In our current research, the false positive values of X-axis are expressed as prediction Suitable Area geologic elements reduction attribute synthesis weighting letter
The ratio of breath amount and highest Comprehensive weight information content, it is accumulative that the true positives value of Y-axis is expressed as points for investigation quantity in prediction Suitable Area
Percentage.Spatial prediction model is assessed using the area ratio (AUC) under ROC curve, and AUC value range is 0.5-1, AUC
It is best for being worth maximum model.The result that AUC value is generated closer to 1 model is better;On the contrary, AUC value is closer to 0.5 meaning
Model generate result it is poorer;It is generally believed that the AUC value of model, which is greater than 0.7, shows that it has high-precision.
AUC is counted using ROC curve, assesses the Factors Weighting superposition mould of classical information computation (IVM), forefathers' exploitation
The precision and predictive ability for the improvement information computation based on fuzzy coarse central that type (FWAM) and the present invention study.Use information
The research area underground engineering of amount model prediction constructs that addressing Suitable Area is as shown in figure 32, based on the prediction of Factors Weighting Additive Model
It is as shown in figure 33 that research area's underground engineering constructs addressing Suitable Area.
The invention proposes fuzzy coarse centrals to improve information computation, to realize the high-precision of engineer construction addressing Suitable Area
Prediction.The core concept and calculation formula of conventional information amount model are described first, and are given in space lattice conditioned matrix
Under influence index information computing formula.It the shortcomings that for equal weight superposition, proposes fuzzy coarse central and improves information content mould
Type, and give the core formula 6 and formula 7 of IVM-FRS;IVM-FRS model is divided into the influence index factor and divides, respectively influences to refer to
Mark five steps such as the factor divides correlation analysis, the selection of grid cell size, weighted information amount calculates, information content threshold value determines
Suddenly;Give the standard deviation and engineer construction addressing between the influence index i.e. factor information amount for measuring factor division methods quality
Whether suitable points for investigation is presented normal distribution;The phase of each influence index is analyzed using Band Collection Statistics
Guan Xing, to realize non-correlation requirement of the IVM-FRS model to influence index;The experience divided using spatial prediction grid cell
Formula realizes the division to grid cell;Three steps of weighted information amount calculating are given, it is folded by ArcGIS software grid
Function is added to form engineer construction addressing Comprehensive weight information content figure;Using some division points conduct of jenks nature breakpoint method
Engineer construction addressing Suitable Area information content threshold value.
Using Kuandian as research area, the prediction of engineer construction addressing Suitable Area is realized.According to five, IVM-FRS model
Step carries out factor division to geologic elements reduction attribute, applies ordinal scale attribute ratings and divide to have obtained geologic elements about
Simple ordinal scale attribute factor division result has obtained the factor division result of the continuous scale variable of geologic elements reduction attribute;
The factor is carried out to geologic elements reduction attribute and divides correlation analysis, is united using Band Collection Statistics tool
Meter analysis has obtained the geologic elements reduction attribute correlation matrix after 18 factors divide, the results showed that 18 after factor division
Geologic elements reduction attribute is independent from each other;Cell division is carried out to geologic elements reduction property distribution figure, takes cell
Size is 70 meters;Underground engineering geologic elements reduction attribute factor weighted information amount and earth's surface engineering or temporary project is calculated
Geologic elements reduction attribute factor weighted information amount;It finally realizes underground engineering and constructs the prediction of addressing Suitable Area and earth's surface engineering
Or temporary project constructs the prediction of addressing Suitable Area.
Engineer construction addressing Suitable Area prediction result is evaluated, using the numerical value of contrast verification legally constituted authority meter underground engineering
N and validation group data quantitative proportion are 100%, and the ratio of earth's surface engineering or temporary project is 98.1%, demonstrate IVM-FRS mould
It is effective that type predictive engine, which constructs addressing Suitable Area,;Using the AUC value difference of ROC curve statistics IVM, FWAM and IVM-FRS
It is 0.834,0.792 and 0.878, demonstrating IVM-FRS has better performance than IVM, FWAM in terms of predictive ability, is three kinds
Predictive engine constructs the best model in addressing Suitable Area in model.
Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
Present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: it still may be used
To modify the technical solutions described in the foregoing embodiments or equivalent replacement of some of the technical features;
And these are modified or replaceed, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (4)
1. a kind of engineer construction addressing Suitable Area prediction technique, the prediction technique successively the following steps are included:
The step of the step of influence index factor divides, each influence index factor divide correlation analysis and the choosing of grid cell size
The step of selecting, it is characterised in that: further include the steps that the step that weighted information amount calculates and information content threshold value determines;
The step of weighted information amount calculates, comprising:
The earth's surface engineering or temporary project structure delimited according to the total number of survey region cell, engineer construction addressing qualitative model
Build addressing Suitable Area cell total number, in survey region the cell total number containing geologic elements reduction attribute factor and
Earth's surface engineering or temporary project construct the cell total number that geologic elements reduction attribute factor is contained in addressing Suitable Area, to calculate
The information content of each influence index factor in geologic elements reduction attribute;
According to the information content of each influence index factor and the earth's surface engineering or temporary project that are determined by fuzzy coarse central method
The weight of geologic elements reduction attribute calculates the weighted information amount of each influence index factor;
All influence indexs that each grid cell includes are calculated according to the weighted information amount of each influence index factor
Factor Comprehensive weight information content;
The step that the information content threshold value determines, comprising:
Using some division points of jenks nature breakpoint method as engineer construction addressing Suitable Area information content threshold value.
2. a kind of engineer construction addressing Suitable Area according to claim 1 prediction technique, which is characterized in that the geology is wanted
The calculation formula of the information content of each influence index factor in plain reduction attribute are as follows:
Wherein, S indicates that the total number of survey region cell, N indicate the earth's surface engineering that engineer construction addressing qualitative model delimited
Or temporary project constructs addressing Suitable Area cell total number, SiIt indicates to contain geologic elements reduction attribute factor in survey region
xiCell total number, NiIndicate earth's surface engineering or temporary project construct addressing Suitable Area contain geologic elements reduction attribute because
Sub- xiCell total number.
3. a kind of engineer construction addressing Suitable Area according to claim 1 prediction technique, which is characterized in that each shadow
Ring the calculation formula of the weighted information amount of index factor are as follows:
I(xi, Y) and=wiln(Ni/N)/(Si/S)
Wherein, S indicates that the total number of survey region cell, N indicate the earth's surface engineering that engineer construction addressing qualitative model delimited
Or temporary project constructs addressing Suitable Area cell total number, SiIt indicates to contain geologic elements reduction attribute factor in survey region
xiCell total number, NiIndicate earth's surface engineering or temporary project construct addressing Suitable Area contain geologic elements reduction attribute because
Sub- xiCell total number, wiIndicate the earth's surface engineering determined by fuzzy coarse central method or temporary project geologic elements reduction category
The weight of property.
4. a kind of engineer construction addressing Suitable Area according to claim 1 prediction technique, which is characterized in that it is described each
The calculation formula for all influence index factor Comprehensive weight information contents that grid cell includes are as follows:
Wherein, S indicates that the total number of survey region cell, N indicate the earth's surface engineering that engineer construction addressing qualitative model delimited
Or temporary project constructs addressing Suitable Area cell total number, SiIt indicates to contain geologic elements reduction attribute factor in survey region
xiCell total number, NiIndicate earth's surface engineering or temporary project construct addressing Suitable Area contain geologic elements reduction attribute because
Sub- xiCell total number, wiIndicate the earth's surface engineering determined by fuzzy coarse central method or temporary project geologic elements reduction category
The weight of property;
IiThere is space research mesh lower than being averaged for survey area in the probability that < 0 indicates that space research target occurs in the grid cell
Mark probability, Ii=0 indicates that average probability of the probability equal to survey area of space research target, I occurs in the grid celli0 table of >
Show that average probability of the probability equal to survey area of space research target occurs in the grid cell.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810936211.4A CN109118004B (en) | 2018-08-16 | 2018-08-16 | Prediction method for suitable area for engineering construction site selection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810936211.4A CN109118004B (en) | 2018-08-16 | 2018-08-16 | Prediction method for suitable area for engineering construction site selection |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109118004A true CN109118004A (en) | 2019-01-01 |
CN109118004B CN109118004B (en) | 2021-09-14 |
Family
ID=64852372
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810936211.4A Active CN109118004B (en) | 2018-08-16 | 2018-08-16 | Prediction method for suitable area for engineering construction site selection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109118004B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110675090A (en) * | 2019-10-14 | 2020-01-10 | 机械工业勘察设计研究院有限公司 | Ground source heat pump field applicability evaluation method |
CN111145452A (en) * | 2019-12-31 | 2020-05-12 | 中国银行股份有限公司 | Site selection method and device for self-service cash recycling machine capable of taking train tickets |
CN113505994A (en) * | 2021-07-12 | 2021-10-15 | 京创智慧科技有限责任公司 | Data processing method, device, equipment and storage medium |
CN116152009A (en) * | 2023-04-19 | 2023-05-23 | 铁正检测科技有限公司 | Tunnel geological monitoring management system based on big data |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030048260A1 (en) * | 2001-08-17 | 2003-03-13 | Alec Matusis | System and method for selecting actions based on the identification of user's fingers |
EP1448760A2 (en) * | 2001-11-14 | 2004-08-25 | Ciphergen Biosystems, Inc. | Methods for monitoring polypeptide production and purification using surface enhanced laser desorption/ionization mass spectrometry |
CN103324980A (en) * | 2013-04-25 | 2013-09-25 | 华北电力大学(保定) | Wind power station wind speed prediction method |
CN103426039A (en) * | 2013-09-08 | 2013-12-04 | 云南师范大学 | Method for establishing location selecting model of mountain photovoltaic power station |
CN103744850A (en) * | 2013-10-14 | 2014-04-23 | 国家电网公司 | Power grid disaster real-time regulating and control device and method based on intuition fuzzy rough set |
CN104239968A (en) * | 2014-09-02 | 2014-12-24 | 浙江大学 | Short-term load predicting method based on quick fuzzy rough set |
CN106228554A (en) * | 2016-07-20 | 2016-12-14 | 西安科技大学 | Fuzzy coarse central coal dust image partition methods based on many attribute reductions |
CN106600063A (en) * | 2016-12-16 | 2017-04-26 | 东北电力大学 | Space load prediction method based on fuzzy-rough set theory and spatio-temporal information |
CN107194420A (en) * | 2017-05-16 | 2017-09-22 | 浙江象立医疗科技有限公司 | A kind of Fuzzy and Rough concentrates the attribute selection method based on information gain-ratio |
US9945680B1 (en) * | 2010-11-12 | 2018-04-17 | Dp Technologies, Inc. | Location-based meeting system |
WO2018079940A1 (en) * | 2016-10-31 | 2018-05-03 | 에스케이플래닛 주식회사 | Bundle packaging apparatus based on characteristics of individual products, and method therefor |
-
2018
- 2018-08-16 CN CN201810936211.4A patent/CN109118004B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030048260A1 (en) * | 2001-08-17 | 2003-03-13 | Alec Matusis | System and method for selecting actions based on the identification of user's fingers |
EP1448760A2 (en) * | 2001-11-14 | 2004-08-25 | Ciphergen Biosystems, Inc. | Methods for monitoring polypeptide production and purification using surface enhanced laser desorption/ionization mass spectrometry |
US9945680B1 (en) * | 2010-11-12 | 2018-04-17 | Dp Technologies, Inc. | Location-based meeting system |
CN103324980A (en) * | 2013-04-25 | 2013-09-25 | 华北电力大学(保定) | Wind power station wind speed prediction method |
CN103426039A (en) * | 2013-09-08 | 2013-12-04 | 云南师范大学 | Method for establishing location selecting model of mountain photovoltaic power station |
CN103744850A (en) * | 2013-10-14 | 2014-04-23 | 国家电网公司 | Power grid disaster real-time regulating and control device and method based on intuition fuzzy rough set |
CN104239968A (en) * | 2014-09-02 | 2014-12-24 | 浙江大学 | Short-term load predicting method based on quick fuzzy rough set |
CN106228554A (en) * | 2016-07-20 | 2016-12-14 | 西安科技大学 | Fuzzy coarse central coal dust image partition methods based on many attribute reductions |
WO2018079940A1 (en) * | 2016-10-31 | 2018-05-03 | 에스케이플래닛 주식회사 | Bundle packaging apparatus based on characteristics of individual products, and method therefor |
CN106600063A (en) * | 2016-12-16 | 2017-04-26 | 东北电力大学 | Space load prediction method based on fuzzy-rough set theory and spatio-temporal information |
CN107194420A (en) * | 2017-05-16 | 2017-09-22 | 浙江象立医疗科技有限公司 | A kind of Fuzzy and Rough concentrates the attribute selection method based on information gain-ratio |
Non-Patent Citations (3)
Title |
---|
孙红政 等: "《地质要素在工程构筑选址中的影响特性分析》", 《测绘与空间地理信息》 * |
染文学: "《基于粗糙集和交互式多目标模糊决策的物流配送中心选址研究》", 《中国物流学术前沿报告》 * |
田智慧等: "《基于GIS的高校内ATM网点选址分析》", 《西北师范大学学报》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110675090A (en) * | 2019-10-14 | 2020-01-10 | 机械工业勘察设计研究院有限公司 | Ground source heat pump field applicability evaluation method |
CN111145452A (en) * | 2019-12-31 | 2020-05-12 | 中国银行股份有限公司 | Site selection method and device for self-service cash recycling machine capable of taking train tickets |
CN113505994A (en) * | 2021-07-12 | 2021-10-15 | 京创智慧科技有限责任公司 | Data processing method, device, equipment and storage medium |
CN116152009A (en) * | 2023-04-19 | 2023-05-23 | 铁正检测科技有限公司 | Tunnel geological monitoring management system based on big data |
Also Published As
Publication number | Publication date |
---|---|
CN109118004B (en) | 2021-09-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yao et al. | Simulating urban land-use changes at a large scale by integrating dynamic land parcel subdivision and vector-based cellular automata | |
Gigović et al. | Application of the GIS-DANP-MABAC multi-criteria model for selecting the location of wind farms: A case study of Vojvodina, Serbia | |
Chabuk et al. | GIS-based assessment of combined AHP and SAW methods for selecting suitable sites for landfill in Al-Musayiab Qadhaa, Babylon, Iraq | |
Torkashvand et al. | DRASTIC framework improvement using stepwise weight assessment ratio analysis (SWARA) and combination of genetic algorithm and entropy | |
Abudeif et al. | Multicriteria decision analysis based on analytic hierarchy process in GIS environment for siting nuclear power plant in Egypt | |
Donevska et al. | Regional non-hazardous landfill site selection by integrating fuzzy logic, AHP and geographic information systems | |
Moeinaddini et al. | Siting MSW landfill using weighted linear combination and analytical hierarchy process (AHP) methodology in GIS environment (case study: Karaj) | |
Zhang et al. | Simulation and analysis of urban growth scenarios for the Greater Shanghai Area, China | |
CN111898315B (en) | Landslide susceptibility assessment method based on fractal-machine learning hybrid model | |
Ghamgosar et al. | Multicriteria decision making based on analytical hierarchy process (AHP) in GIS for tourism | |
CN109146204A (en) | A kind of wind power plant booster stations automatic addressing method of comprehensiveestimation | |
CN109118004A (en) | A kind of engineer construction addressing Suitable Area prediction technique | |
Omar et al. | Markov-CA model using analytical hierarchy process and multiregression technique | |
Shi et al. | Study on evaluation of regional cultivated land quality based on resource-asset-capital attributes and its spatial mechanism | |
Demir | Landslide susceptibility mapping by using statistical analysis in the North Anatolian Fault Zone (NAFZ) on the northern part of Suşehri Town, Turkey | |
CN111882244A (en) | Construction method of multi-source homeland development risk assessment system based on hierarchical framework | |
Sikder et al. | Geospatial analysis of building structures in megacity Dhaka: the use of spatial statistics for promoting data-driven decision-making | |
CN113240257A (en) | Territorial space partitioning method and device based on minimum cumulative resistance model | |
Alavipoor et al. | A geographic information system for gas power plant location using analytical hierarchy process and fuzzy logic | |
Anagnostopoulos et al. | Land suitability analysis for natural wastewater treatment systems using a new GIS add-in for supporting criterion weight elicitation methods | |
CN112508399A (en) | Underground space development suitability evaluation method and system | |
Shenavr et al. | Comparison of Multi-criteria evaluation (AHP and WLC approaches) for land capability assessment of urban development in GIS | |
Asefi et al. | A multi-criteria decision support framework for municipal solid waste landfill siting: a case study of New South Wales (Australia) | |
Cheng et al. | Updating conventional soil maps by mining soil–environment relationships from individual soil polygons | |
Rostami et al. | Landslide susceptibility mapping by using fuzzy logic: a case study of Cham-gardalan catchment, Ilam, Iran |
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 |