CN106096847A - A kind of fuzzy change weighs Engineering-geological environmental quality method - Google Patents

Abstract

The present invention relates to a kind of fuzzy change and weigh Engineering-geological environmental quality method, the method comprises the steps: to set up Engineering-geological environmental quality hierarchy structure model；The engineering geology data that in obtaining region to be evaluated, all indexs that participate in evaluation and electing are corresponding；Region to be evaluated is divided into m basic evaluation unit；Set up the evaluation index matrix of each basic evaluation unit and become power weight matrix；The evaluation index matrix corresponding according to base unit and change power weight matrix ask for the variable synthesis evaluation of estimate of each basic evaluation unit respectively；Variable synthesis evaluation of estimate according to each basic evaluation unit obtains the Urban Engineering Geological Environmental Quality grade of each basic evaluation unit in region to be evaluated；Use clustering procedure to be clustered by the basic evaluation unit with identical Urban Engineering Geological Environmental Quality grade, obtain Urban Engineering Geological Environmental Quality overall merit block plan.Compared with prior art, the present invention have that evaluation precision is high, the objective advantage such as reliably of evaluation result.

Description

A kind of fuzzy change weighs Engineering-geological environmental quality method

Technical field

The present invention relates to a kind of Engineering-geological environmental quality method, especially relate to a kind of fuzzy change and weigh engineering geology Method of environmental quality evaluation.

Background technology

City is human living and the regional space of economic activity high concentration, is also the topmost field of engineering construction. Engineering-geological environmental quality is an important process in overall city planning, scientific and reasonable Urban Engineering Geological Environmental Quality Evaluate, be conducive to evading the unfavorable factor to engineering construction in geological environment, Appropriate application land resource, promote engineering construction with The coordinated development of geological environment.Urban Engineering Geological Environmental Quality influence factor is numerous and diverse, and each factor also exists certain association each other Property, the influence degree of each factor is different, and factor of evaluation has the feature such as achievement data ambiguity and evaluation criterion ambiguity, It is difficult to classical mathematics in addition unified metric.Therefore, the theory of fuzzy mathematics and method are in Engineering-geological environmental quality Obtain widest application.At present, engineering geological environment evaluation is many is that platform uses structure to comment with GIS-Geographic Information System (GIS) Urban Engineering Geological Environmental Quality is evaluated by valency index system and FUZZY WEIGHTED scoring model.

The weight assignment method having considerable influence to evaluation result has: objective-weight method and subjectivity weigh method surely.Objective-weight Method mainly has: Information Entropy, PCA, average variance method etc.；Subjectivity weighs method surely mainly to be had: directly scoring, expert estimation Method, analytic hierarchy process (AHP), chain rate point system, contrast ranking method etc..Owing to the index that participates in evaluation and electing of Geological Environment Engineering mostly is misty idea There is no an exact value, and strong according to lazyness to data of objective-weight method, to data to specific strength, conflicting and discreteness requirement Higher and computational methods are complicated, the property of may participate in and poor universality.Therefore, objective-weight method is used relatively in Geological Environment Evaluation Few.Analytic hierarchy process (AHP) in subjective fixed power has can be by challenge stratification, and the characteristic of qualitative question quantification, therefore on ground Matter environmental evaluation is used wide.But each evaluation index only one of which weight in analytic hierarchy process (AHP), weight determines rear nothing Discussing how the desired value that participates in evaluation and electing changes, index weights always immobilizes, the quality evaluation of the Geological Environment Engineering finally given Precision and quality are difficult to effectively serve engineering construction.Therefore, it is necessary to use new technology and method that it is carried out More accurate quantitative evaluation, could provide more accurate science to depend on for new period reallocation of land utilization and engineering construction According to.

At present, the research field of Geological Environment Evaluation, there is no relevant evaluation methodology and carry out evaluation engineering Geological-environmental Quality. Acquisition methods many employings expert graded of weight or analytic hierarchy process (AHP) (AHP) in existing evaluation methodology are as big in Xi'an science and technology The patent " a kind of coal field geological environment evaluation on bearing capacity method " (CN104881571A) of the Sun Xueyang learned et al., China's stone The patent " the geological environment bearing capacity computation method of a kind of littoral zone " of the Song Dongmei of oil university (East China) et al. (CN104361225A), University of Science & Technology, Beijing thank plan literary composition et al. patent " interpretation of a kind of reservoir basin engineering geological survey with Evaluation methodology " (CN102707332A), Architectural Design & Research Institute of Tongji University (Group) Co., Ltd. Zheng Tao et al. invention special Profit " evaluation methodology of a kind of sponge city based on gray relative analytic hierarchy process (AHP) potential index " (CN105005690A).This Outward, Geological Environment Evaluation method also has the methods such as Multi-objective Decision Analysis method, artificial neural network method, logistic regression.Above In method more to studying on selecting index and integrated evaluating method, and evaluation result is had the weight assignment method of considerable influence Study less.

Summary of the invention

Defect that the purpose of the present invention is contemplated to overcome above-mentioned prior art to exist and provide a kind of and fuzzy become power engineering Geological Environmental Quality Assessment method.

The purpose of the present invention can be achieved through the following technical solutions:

A kind of fuzzy change weighs Engineering-geological environmental quality method, and the method comprises the steps:

(1) set up Engineering-geological environmental quality hierarchy structure model, refer to including general objective layer, sub-goal layer and evaluation Mark layer, the described general objective that general objective layer is this Engineering-geological environmental quality, described sub-goal layer is and general objective Corresponding multiple sub-goals, each sub-goal includes multiple index that participates in evaluation and electing, all index composition evaluation index layers that participate in evaluation and electing；

(2) all indexs pair that participate in evaluation and electing in Engineering-geological environmental quality hierarchy structure model in region to be evaluated are obtained The engineering geology data answered；

(3) engineering geology data step (2) obtained import GIS software, utilize GIS software to set up region to be evaluated Spatial database, and region to be evaluated is divided into m basic evaluation unit；

(4) the evaluation index square of each basic evaluation unit is set up according to the engineering geology data of each basic evaluation unit Battle array；

(5) the change power weight matrix that each basic evaluation unit is corresponding is asked for respectively；

(6) according to evaluation index matrix corresponding to base unit and become power weight matrix and ask for each basic evaluation respectively The variable synthesis evaluation of estimate of unit；

(7) by the variable synthesis evaluation of estimate of each basic evaluation unit respectively with set Urban Engineering Geological Environmental Quality grade Divide value compares, and obtains the Urban Engineering Geological Environmental Quality grade of each basic evaluation unit in region to be evaluated；

(8) clustering procedure is used to be clustered by the basic evaluation unit with identical Urban Engineering Geological Environmental Quality grade, To Urban Engineering Geological Environmental Quality overall merit block plan.

Step (3) comprises the steps:

(301) the engineering geology data in region to be evaluated being imported GIS software, the index that respectively participates in evaluation and electing treating evaluation region is built Vertical single index data layer；

(302) it is standardized single index data layer processing；

(303) each single index data layer treating evaluation region carries out rasterizing process, is divided in region to be evaluated For m basic evaluation unit.

Step (4) particularly as follows:

According to the index divided rank standard that participates in evaluation and electing, ask for regional evaluation index matrix X ' to be evaluated:

$X^{\′}=\left[\begin{array}{c}{X}_1^{\′}\\ X_2^{\′}\\ \mathrm{...}\\ X_m^{\′}\end{array}\right],$

Element in matrix X ' is designated as X '_i, X '_iFor the evaluation index matrix of i-th basic evaluation unit, i=1,2 ... Total number of evaluation unit based on m, m；

Specifically, the evaluation index matrix X ' of i-th basic evaluation unit_iFor:

X′_i=[x '_i1 x′_i2 … x′_in],

Matrix X '_iIn element be denoted as x '_ij, x '_ijRepresent that jth in i-th basic evaluation unit participates in evaluation and electing the index of index Value, j=1,2 ..., n, n represent total number of the index that participates in evaluation and electing.

Step (5) particularly as follows:

(501) analytic hierarchy process (AHP) is used to ask in each basic evaluation unit evaluation index layer respectively for general objective layer Often weighing weight matrix, often the element in power weight matrix is the often weight of index of respectively participating in evaluation and electing；

(502) strong punishment-stimulable type State Variable Weight function is built, respectively the often power weight square to each basic evaluation unit Each normal power weight in Zhen is redistributed, and then obtains the change power weight matrix of each basic evaluation unit, becomes power weight Element in matrix be respectively participate in evaluation and electing index change power weight.

The often power weight matrix asking for each basic evaluation unit in described step (501) is asked especially by following method Take:

(501a) 1～9 scaling laws in analytic hierarchy process (AHP) are used to set up sub-goal layer to always i-th basic evaluation unit The judgment matrix B-A of destination layer and the judgment matrix C-B, wherein i=1,2 of evaluation index layer antithetical phrase destination layer ... m, m are Total number of basic evaluation unit；

(501b) ask for often weighing weight matrix according to judgment matrix B-A and C-BWherein Chang Quanquan Weight matrixIn element be denoted asRepresent that jth in i-th basic evaluation unit participates in evaluation and electing the often weight of index, j=1, 2 ... n, n represent total number of the index that participates in evaluation and electing.

Described step (502) particularly as follows:

(502a) according to engineering geological environment evaluation feature and variable-weight theory, strong punishment-stimulable type State Variable Weight letter is built Number S (X):

$S\left(X\right)=\left\{\begin{array}{cc}\frac{a-b}{\mathrm{\λ}-\mathrm{\μ}}\mathrm{\μ}\ln \frac{\mathrm{\μ}}{x_j}+a,& x_j\∈(0,\mathrm{\μ}\]\\ \frac{b-a}{\mathrm{\λ}-\mathrm{\μ}}{x}_j+\frac{a\mathrm{\λ}-b\mathrm{\μ}}{\mathrm{\λ}-\mathrm{\μ}},& x_j\∈(\mathrm{\μ},\mathrm{\λ}\]\\ \frac{a-b}{2(\mathrm{\λ}-\mathrm{\μ})(\mathrm{\α}-\mathrm{\λ})}{(\mathrm{\α}-x_j)}^2+c,& x_j\∈(\mathrm{\λ},\mathrm{\α})\\ c,& x_j\∈\[\mathrm{\α},\mathrm{\β}\]\\ K(1-\mathrm{\β})\ln \frac{1-\mathrm{\β}}{1-x_j}+c,& x_j\∈(\mathrm{\β},1)\end{array}\right.,$

Wherein, x_jFor state value, 0 ＜ μ ＜ λ ＜ α ＜ β ＜ 1,0 ＜ c ＜ b ＜ a ＜ 1, μ is extra-heavy punishment level, and λ is strong Punishment level, α is for just to punish that level, β are stimulation level, and a, b are the State Variable Weight that μ, λ are corresponding with during α with c respectively state value Functional value, k is regulation coefficient；

(502b) the evaluation index matrix X ' of i-th basic evaluation unit is obtained_i, by X '_iMiddle element is denoted as x '_ij, x '_ijTable Show that jth in i-th basic evaluation unit participates in evaluation and electing the desired value of index, by x '_ijShape respectively as State Variable Weight function S (X) State value is asked for jth in i-th basic evaluation unit and is participated in evaluation and electing the State Variable Weight functional value S of index_ij(X), i=1,2 ... m, m Based on total number of evaluation unit, j=1,2 ..., n, n represent total number of the index that participates in evaluation and electing；

(502c) the change power weight matrix W of i-th basic evaluation unit is obtained_i=[w_i1 w_i2 … w_in], wherein become power Element in weight matrix is denoted as w_ij, w_ijRepresent i-th basic evaluation unit jth participate in evaluation and electing index change power weight, w_ijPass through Following formula obtains:

$w_{ij}=\frac{w_{ij}^0\·S_{ij}\left(X\right)}{\underset{j=1}{\overset{n}{\Σ}}(w_{ij}^0\·S_{ij}(X\left)\right)}.$

Described step (6) particularly as follows:

(601) the evaluation index matrix X ' of i-th basic evaluation unit is obtained respectively_iAnd become power weight matrix W_i, wherein I=1,2 ... total number of evaluation unit, X ' based on m, m_iMatrix, matrix X ' is tieed up for 1*n_iIn element representation be x '_ij, x′_ijThe desired value of the index that participates in evaluation and electing for jth in i-th basic evaluation unit, becomes power weight matrix W_iMatrix, this square is tieed up for 1*n Jth element in Zhen is denoted as w_ij, w_ijRepresent that jth in i-th basic evaluation unit participates in evaluation and electing the change power weight of index, j=1, 2 ... n, n represent total number of the index that participates in evaluation and electing；

(602) ask for jth in i-th basic evaluation unit participate in evaluation and electing index change power evaluation of estimate M_ij:

M_ij=w_ij·x_ij；

(603) variable synthesis evaluation of estimate M of i-th basic evaluation unit is asked for_i:

$M_i=\underset{j=1}{\overset{n}{\Σ}}{M}_{ij}.$

Compared with prior art, present invention have the advantage that

(1) region to be evaluated is divided into several basic evaluation unit by the present invention, respectively to each basic evaluation unit It is evaluated so that evaluation precision is high；

(2) present invention becomes power evaluation methodology to make up each index that participates in evaluation and electing in Urban Engineering Geological Environmental Quality overall merit the most true With regard to the defect of only one of which weighted value after Ding, it is impossible to the reflection indices internal diversity shadow to Urban Engineering Geological Environmental Quality Ring, more cannot reflect the feature that weight should change with the change of each index internal diversity, and then have influence on last evaluation The objectivity of result；

(3) present invention asks for setting up strong punishment-stimulable type State Variable Weight function when becoming power weight to the index correspondence that respectively participates in evaluation and electing Often power weight redistribute, " outstanding " value of the index that participates in evaluation and electing is given encouragement, the index that participates in evaluation and electing " difference " is punished, " the typically value " of the index that participates in evaluation and electing neither is punished and does not encourages, fully demonstrated indices internal diversity to engineering geology ring The globality impact of border quality, greatly improves credibility and the reliability of Engineering-geological environmental quality grade.

Accompanying drawing explanation

Fig. 1 is the basic flow sheet of the fuzzy change power Engineering-geological environmental quality method of the present invention；

Fig. 2 is the structural representation of the present embodiment Engineering-geological environmental quality hierarchy structure model；

Fig. 3 is the schematic diagram of strong punishment-stimulable type State Variable Weight function.

Detailed description of the invention

The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.

Embodiment

It is illustrated in figure 1 a kind of fuzzy basic flow sheet becoming power Engineering-geological environmental quality method of the present invention, should Method comprises the steps:

Step 1: set up Engineering-geological environmental quality hierarchy structure model, including general objective layer, sub-goal layer with comment Valency indicator layer, the described general objective that general objective layer is this Engineering-geological environmental quality, described sub-goal layer is with always Multiple sub-goals that target is corresponding, each sub-goal includes multiple index that participates in evaluation and electing, all index composition evaluation index layers that participate in evaluation and electing；

Step 2: obtain all fingers that participate in evaluation and electing in Engineering-geological environmental quality hierarchy structure model in region to be evaluated The engineering geology data that mark is corresponding；

Step 3: engineering geology data step 2 obtained import GIS software, utilize GIS software to set up region to be evaluated Spatial database, and region to be evaluated is divided into m basic evaluation unit；

Step 4: the evaluation setting up each basic evaluation unit according to the engineering geology data of each basic evaluation unit refers to Mark matrix；

Step 5: ask for the change power weight matrix that each basic evaluation unit is corresponding respectively；

Step 6: the evaluation index matrix corresponding according to base unit and change power weight matrix ask for each basis respectively The variable synthesis evaluation of estimate of evaluation unit；

Step 7: by the variable synthesis evaluation of estimate of each basic evaluation unit respectively with set Urban Engineering Geological Environmental Quality Grade classification value compares, and obtains the Urban Engineering Geological Environmental Quality grade of each basic evaluation unit in region to be evaluated；

Step 8: use clustering procedure to be clustered by the basic evaluation unit with identical Urban Engineering Geological Environmental Quality grade, Obtain Urban Engineering Geological Environmental Quality overall merit block plan.

In the present embodiment step 1, general objective layer is denoted as A, chooses topography and geomorphology (B₁), hydrogeologic condition (B₂), ground work Journey condition (B₃), Geological Environment Engineering problem (B₄), geological disaster (B₅) and anthropogenic influence (B₆) six affect Geological Environment Engineering The factor of quality is as the sub-goal layer (B of this appraisement system_i).And geomorphic unit (C₁), the gradient (C₂), ground elevation (C₃) make For topography and geomorphology (B₁) the index that participates in evaluation and electing；Groundwater level depth (C₄), distance (C away from surface water body₅) as hydrogeologic condition (B₂) the index that participates in evaluation and electing；Soil supporting layer thickness (C₆), ground Rock And Soil uniformity (C₇), foundation bearing capacity (C₈) as ground Engineering specifications (B₃) the index that participates in evaluation and electing；Collapsibility of loess (C₉), sand liquefaction (C₁₀), thickness of soft soil (C₁₁) as engineering geology Problem (B₄) the index that participates in evaluation and electing；Distance active fault distance (C₁₂), geological disaster point (C₁₃), cone of groundwater settling amount (C₁₄) As baleful geological processes (B₅) the index that participates in evaluation and electing；Human dried bloodstains intensity (C₁₅) as the effect of human activity (B₆) unique The index that participates in evaluation and electing, formed evaluation region multi-level evaluation index structural model, as shown in Figure 2.

Then step 2 obtains the engineering geology data that above-mentioned 15 indexs that participate in evaluation and electing are corresponding, continues executing with step 3.

Step 3 comprises the steps:

(301) the engineering geology data in region to be evaluated being imported GIS software, the index that respectively participates in evaluation and electing treating evaluation region is built Vertical single index data layer, i.e. utilizes GIS by the map data in initial data and to test data, through projection transform, information The processing procedures such as extraction, assignment, buffer zone analysis and DTM analysis, obtain evaluating required single index data layer；

(302) it is standardized single index data layer processing, so that each single index data layer has system The evaluation region border of one, unified coordinate system and unified projective parameter；

(303) each single index data layer treating evaluation region carries out rasterizing process, is divided in region to be evaluated For m basic evaluation unit.

The diversity affected Urban Engineering Geological Environmental Quality due to the index that respectively participates in evaluation and electing and complexity, cause respectively participating in evaluation and electing index There is obvious spatial diversity in assemblage characteristic, in therefore Engineering-geological environmental quality should implement to each discreet region, because of This needs survey region is divided into several little pels, i.e. basic evaluation unit.According to evaluation region Urban Engineering Geological Environmental Quality The spatial distribution characteristic of index, evaluation objective required precision and the computer system disposal ability to data, selection standard grid Study area is carried out dividing elements, the data of the index that respectively participates in evaluation and electing are converted into raster data.In the embodiment of the present invention with 400m × 400m grid grid is data storage and the basic evaluation unit of evaluation analysis, and this evaluation region is divided into 6048 basic evaluation lists Unit, the most above-mentioned m=6048.

Step 4 particularly as follows:

According to the index divided rank standard that participates in evaluation and electing, ask for regional evaluation index matrix X ' to be evaluated:

$X^{\′}=\left[\begin{array}{c}{X}_1^{\′}\\ X_2^{\′}\\ ...\\ X_m^{\′}\end{array}\right],$

Element in matrix X ' is designated as X '_i, X '_iFor the evaluation index matrix of i-th basic evaluation unit, i=1,2 ... Total number of evaluation unit based on m, m；

Specifically, the evaluation index matrix X ' of i-th basic evaluation unit_iFor:

X′_i=[x '_i1 x′_i2 … x′_in],

Matrix X '_iIn element be denoted as x '_ij, x '_ijRepresent that jth in i-th basic evaluation unit participates in evaluation and electing the index of index Value, j=1,2 ..., n, n represent total number of the index that participates in evaluation and electing.

Because the data mode of the index that participates in evaluation and electing is different, each evaluation index dimension is different, different dimension data need to be carried out standard Change processes.Before being standardized data processing, it is necessary to first determine the index grading standard that participates in evaluation and electing.According to " urban and rural planning Engineering geological investigation specification " (CJJ57-2012), with reference to domestic and international project Geological Environmental Quality Assessment index grading standard It is divided into several grade in conjunction with the expert opinion index that will participate in evaluation and electing.The index that will participate in evaluation and electing in the present embodiment is divided into from getting well to differing from: excellent, Good, in, general, differ from 5 ranks.The evaluation region of embodiment closes on great river, and the alluvial flat in river levee is flood flowing district, when evaluating Flood flowing district should be processed as a sensitizing range, each qualitative index is every relate to this region all as opinion rating In poor level process.As shown in table 1 according to the above index divided rank standard that participates in evaluation and electing formulated.

Table 1 Engineering-geological environmental quality participates in evaluation and electing index divided rank standard

In Engineering-geological environmental quality, data are divided into discrete type and the big class of continuous two.Index number for discrete type According to, 5 ranks in the present embodiment, the index that participates in evaluation and electing divided give value between 0～1 by conversion level；Finger for continuous Mark data, use extreme value standard technique to process, it is thus achieved that the value between 0～1.Assignment is the highest, represents that the indicator conditions that participates in evaluation and electing is the best；Assignment The lowest, represent that the indicator conditions that participates in evaluation and electing is the poorest.Being positive data separate (1) formula to polarity, data separate (2) formula that polarity is negative is entered After column criterionization processes, obtain evaluation index matrix X '_i:

X′_i=[x '_i1 x′_i2 … x′_in],

$x^{\′}=\frac{x_i-x_{min}}{x_{\max }-x_{min}}---\left(1\right)$

$x^{\′}=\frac{x_{max}-x_i}{x_{max}-x_{min}}---\left(2\right)$

X ' is the desired value of a certain index that participates in evaluation and electing；x_iActual value for a certain index that participates in evaluation and electing；x_minFor a certain index that participates in evaluation and electing Minima；x_maxMaximum for a certain index that participates in evaluation and electing.

Step 5 particularly as follows:

(501) analytic hierarchy process (AHP) is used to ask in each basic evaluation unit evaluation index layer respectively for general objective layer Often weighing weight matrix, often the element in power weight matrix is the often weight of index of respectively participating in evaluation and electing；

(502) strong punishment-stimulable type State Variable Weight function is built, respectively the often power weight square to each basic evaluation unit Each normal power weight in Zhen is redistributed, and then obtains the change power weight matrix of each basic evaluation unit, becomes power weight Element in matrix be respectively participate in evaluation and electing index change power weight.

The often power weight matrix asking for each basic evaluation unit in described step (501) is asked especially by following method Take:

(501a) 1～9 scaling laws in analytic hierarchy process (AHP) are used to set up sub-goal layer to always i-th basic evaluation unit The judgment matrix B-A of destination layer and the judgment matrix C-B, wherein i=1,2 of evaluation index layer antithetical phrase destination layer ... m, m are Total number of basic evaluation unit；

(501b) ask for often weighing weight matrix according to judgment matrix B-A and C-BWherein Chang Quanquan Weight matrixIn element be denoted asRepresent that jth in i-th basic evaluation unit participates in evaluation and electing the often weight of index, j= 1,2 ... n, n represent total number of the index that participates in evaluation and electing.

Specifically, 1～9 scaling laws in analytic hierarchy process (AHP) are used to construct multilevel iudge matrix two-by-two, 1～9 scaling law marks Degree implication is as shown in table 2.According to hierarchy structure model, for B layer and C layer by above for the same straton index straton index The ratio using 1～9 for criterion compares two-by-two, and indices relative importance in each layer is given certain judgement, Obtain the B layer judgment matrix [B to A layer₁-A, B₂-A ..., B_q-A] and the C layer judgment matrix [C to B layer₁-B₁, C₂-B₂.., C_k-B_q]。

Table 2 1～9 scaling law scale implication

Relative weight according to B-A and the C-B judgment matrix Calculation Estimation factor.Now say as a example by the weight calculation of B-A layer Bright calculation procedure, it is assumed that F ∈ M_n×nFor B-A judgment matrix, f_ij∈ F, i=1,2 ..., n；J=1,2 ..., n.Characteristic vector is just It is B₁, B₂..., B_nWeight vector to AMeridional (3) are done normalized calculating and are tried to achieve: W=(w₁, w₂,…,w_n)^TFor the characteristic vector of matrix F, the weight of the index that respectively participates in evaluation and electing.

$w_i=\frac{{\left(\underset{j=1}{\overset{n}{\Π}}{f}_{ij}\right)}^{1/n}}{\underset{i=1}{\overset{n}{\Σ}}{\left(\underset{j=1}{\overset{n}{\Π}}{f}_{ij}\right)}^{1/n}}---\left(3\right)$

Wherein f_ijElement for the i-th row jth row of matrix F.And the eigenvalue λ of matrix_max:

${\mathrm{\λ}}_{\max }=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}\frac{{\[FW\]}_i}{w_i}---\left(4\right)$

Wherein, [FW]_iFor the i-th row element in matrix F and character pair vector W^TThe sum of products, w_iFor weight vector W i-th The weight of individual index.

In order to reduce the discordance deviation that critical thinking occurs, the discordance of judgment matrix and randomness be carried out Inspection.

$CI=\frac{({\mathrm{\λ}}_{max}-n)}{n-1}---\left(5\right)$

$CR=\frac{CI}{RI}---\left(6\right)$

Wherein: CI is coincident indicator；λ_maxFor Maximum characteristic root；N is matrix exponent number；RI is that mean random concordance refers to Mark；CR is random Consistency Ratio.If random Consistency Ratio CR ＜ 0.10, it is judged that matrix has satisfactory consistency, required Weighted value is relatively more reasonable, and the value otherwise readjusting judgment matrix carries out multilevel iudge two-by-two again.

According to the Engineering-geological environmental quality index hierarchy structure model built in the embodiment of the present invention, to i-th Individual basic evaluation unit uses analytic hierarchy process (AHP) to obtain B-A judgment matrix and C-B judgment matrix is as follows:

B-A judgment matrix:

$\begin{array}{c}\left[\begin{array}{cccccc}1& 1/2& 1/3& 1/4& 1/5& 5\\ 2& 1& 1/2& 1/3& 1/4& 6\\ 3& 2& 1& 1/2& 1/2& 7\\ 4& 3& 2& 1& 1/3& 8\\ 5& 4& 2& 3& 1& 9\\ 1/5& 1/6& 1/7& 1/8& 1/9& 1\end{array}\right]\\ {\mathrm{\&lambda;}}_{max}=3.009,CI=0.0542,CR=0.0437<0.1\end{array};$

C-B judgment matrix:

$\begin{array}{c}C-B_1:\left[\begin{array}{ccc}1& 1/4& 3\\ 4& 1& 5\\ 1/3& 1/5& 1\end{array}\right]\\ {\mathrm{\&lambda;}}_{\max }=3.009;CI=0.0018\\ CR=0.0032<0.1\end{array};$

$\begin{array}{c}C-B_2:\left[\begin{array}{cc}1& 2\\ 1/2& 1\end{array}\right]\\ CR=0.0<0.1\end{array};$

$\begin{array}{c}C-B_3:\left[\begin{array}{ccc}1& 2& 1/2\\ 1/2& 1& 1/3\\ 2& 3& 1\end{array}\right]\\ \mathrm{\&lambda;}\max =3.0092;CI=0.0046\\ CR=0.0079<0.1\end{array};$

$\begin{array}{c}C-B_4:\left[\begin{array}{ccc}1& 1& 3\\ 1& 1& 5\\ 1/3& 1/5& 1\end{array}\right]\\ \mathrm{\&lambda;}\max =3.0183;CI=0.0091\\ CR=0.0158<0.1\end{array};$

$\begin{array}{c}C-B_5:\left[\begin{array}{ccc}1& 5/4& 3/2\\ 4/5& 1& 5/4\\ 2/3& 4/5& 1\end{array}\right]\\ \mathrm{\&lambda;}\max =3.0002;CI=0.00001\\ CR=0.0002<0.1\end{array};$

Finally obtain that C-A respectively participates in evaluation and electing index often weighs weighted value:

$\begin{array}{c}{W}_i^0=(0.0138,0.0357,0.0077,0.0654,0.0327,0.0521,0.0286,0.0944,0.1084,\\ 0.0947,0.0414,0.1734,0.1377,0.1093,0.0047).\end{array}$

Described step (502) particularly as follows:

(502a) according to engineering geological environment evaluation feature and variable-weight theory, strong punishment-stimulable type State Variable Weight letter is built Number S (X):

$S\left(X\right)=\left\{\begin{array}{cc}\frac{a-b}{\mathrm{\&lambda;}-\mathrm{\&mu;}}\mathrm{\&mu;}\ln \frac{\mathrm{\&mu;}}{x_j}+a,& x_j\&Element;(0,\mathrm{\&mu;}\&rsqb;\\ \frac{b-a}{\mathrm{\&lambda;}-\mathrm{\&mu;}}{x}_j+\frac{a\mathrm{\&lambda;}-b\mathrm{\&mu;}}{\mathrm{\&lambda;}-\mathrm{\&mu;}},& x_j\&Element;(\mathrm{\&mu;},\mathrm{\&lambda;}\&rsqb;\\ \frac{a-b}{2(\mathrm{\&lambda;}-\mathrm{\&mu;})(\mathrm{\&alpha;}-\mathrm{\&lambda;})}{(\mathrm{\&alpha;}-x_j)}^2+c,& x_j\&Element;(\mathrm{\&lambda;},\mathrm{\&alpha;})\\ c,& x_j\&Element;\&lsqb;\mathrm{\&alpha;},\mathrm{\&beta;}\&rsqb;\\ K(1-\mathrm{\&beta;})\ln \frac{1-\mathrm{\&beta;}}{1-x_j}+c,& x_j\&Element;(\mathrm{\&beta;},1)\end{array}\right.,$

Wherein, x_jFor state value, 0 ＜ μ ＜ λ ＜ α ＜ β ＜ 1,0 ＜ c ＜ b ＜ a ＜ 1, μ is extra-heavy punishment level, and λ is strong Punishment level, α is for just to punish that level, β are stimulation level, and a, b are the State Variable Weight that μ, λ are corresponding with during α with c respectively state value Functional value, k is regulation coefficient；

(502b) the evaluation index matrix X ' of i-th basic evaluation unit is obtained_i, by X '_iMiddle element is denoted as x '_ij, x '_ijTable Show that jth in i-th basic evaluation unit participates in evaluation and electing the desired value of index, by x '_ijShape respectively as State Variable Weight function S (X) State value is asked for jth in i-th basic evaluation unit and is participated in evaluation and electing the State Variable Weight functional value S of index_ij(X), i=1,2 ... m, m Based on total number of evaluation unit, j=1,2 ..., n, n represent total number of the index that participates in evaluation and electing；

(502c) the change power weight matrix W of i-th basic evaluation unit is obtained_i=[w_i1 w_i2 … w_in], wherein become power Element in weight matrix is denoted as w_ij, w_ijRepresent i-th basic evaluation unit jth participate in evaluation and electing index change power weight, w_ijPass through Following formula obtains:

$w_{ij}=\frac{w_{ij}^0\&CenterDot;S_{ij}\left(X\right)}{\underset{j=1}{\overset{n}{\&Sigma;}}(w_{ij}^0\&CenterDot;S_{ij}(X\left)\right)}.$

Specifically, when building strong punishment-stimulable type State Variable Weight function, according to Engineering-geological environmental quality feature And variable-weight theory, the local variable weight building evaluation region is theoretical；

If it is sub-goal that Engineering-geological environmental quality catalogue is designated as A, B, A={B₁, B₂..., B_q, I_q=(1, 2 ..., q), any one B_iAlso there are multiple evaluation index, i.e. B_i={ C₁,C₂,…,C_k}.Now might as well set the state of general objective A as X ={ [x₁, x₂..., x_n]^T, 0 ＜ x_j＜ 1, I_n=1,2 ..., n}, x_jIt is B_iState value, according to State Variable Weight principle, make It is defined below:

Definition 1: claim vectorIt is that a n ties up normal weight vector, ifHave1), and Meet

Definition 2: the given W:[0 that maps, 1]ⁿ→ (0,1)ⁿ, claim vector W (X) ≡ (W₁(X), W₂(X) ..., W_n(X)) it is one N ties up local variable weight vector, if meeting condition:

①W_j∈ (0,1), and

The most rightβ ∈ (0,1), and α≤β so that W_j(X) about x_jIn [0, α] upper monotone decreasing, and [β, 1] upper monotonic increase；If each W_j(X) all continuous about all arguments, then W (X) is called a continuous local variable weight Vector.

Definition 3: the given S:[0 that maps, 1]ⁿ→ [0 ,+∞]ⁿ, X | → S (X)=(S₁(X), S₂(X) ..., S_n(X)), S is claimed (X) it is that a n ties up continuous local state variable weight vector, if rightβ ∈ (0,1), and α ＜ β, meet condition:

1. as 0 ＜ x_jDuring ＜ α, S_j(X) ＞ 0, S '_j(X) ＜ 0,

2. as α≤x_jDuring≤β, S_j(X) ＞ 0, S '_j(X)=0；

3. as β ＜ x_jDuring ＜ 1, S_j(X) ＞ 0, S '_j(X) ＞ 0,

4. as S '_j(X)(x_j-x_k)+S_j(X) ＞ 0, j, k ∈ I_n, x_j≤x_k；

Wherein become power W (X) and meet the condition of definition 2, as each S_j(X) about the equal consecutive hours of all arguments, S (X) is called Local punishment stimulable type state variable weight vector, α is punishment level, and β is stimulation level.

For Urban Engineering Geological Environmental Quality generally speaking, the lowest if any 1～2 Danger Indexes values, which kind of no matter uses normal Power integrated evaluating method, is likely to be neutralized by the high level of other indexs so that final appraisal results deviation Practical Project geology Ambient conditions.Therefore, Urban Engineering Geological Environmental Quality change power evaluation should have following principle: when evaluating, as long as there being an index Value is special low, even the normal weights of this index are the least, final comprehensive evaluation value will be substantially reduced；But single desired value is extra-high to differ Making final evaluation of estimate increase substantially surely, in variable synthesis evaluation, punishment dynamics should be greater than dynamical encourage.According to engineering geology ring In the variable synthesis evaluation of border quality, punishment dynamics should be greater than the objective of dynamical encourage, and variable synthesis evaluation should have following spy Point: 1. sub-goal evaluates B_iEvaluation of estimate x_j∈ (0,1), stimulation level β are near 1, i.e. excited target interval (β, 1) is narrower, this interval State value by strong excitation change rapidly；2. punishment level is close with stimulation level, and the most qualified district [α, β] is the narrowest, this district Between state value be neither activated not pay for；3. the interval (0, α) that catches hell is wider, can be divided into just punish the stage (λ, α), Punish the stage by force (λ, μ], the extra-heavy punishment stage (0, μ] 3 stages.Power evaluation feature is become according to Urban Engineering Geological Environmental Quality, can If just the punishment stage is parabola, the strong punishment stage is straight line, and extra-heavy punishment stage and excitation phase are loaarithmic curve, and punish by force Penalize the contact of a curve in the straight line in stage and extra-heavy punishment stage in point (μ, a) and (λ, b), the most just has a following theorem:

Theorem 1 (theorem is coordinated in punishment-excitation segmentation): set State Variable Weight function S (X) and just punishing the stage, punishing rank by force Section, extra-heavy punishment stage, the variable weight function of excitation phase are respectively S₁(X)、S₂(X)、S₃(X)、S₄(X), and point (α, c), (λ, b), (μ, a), (β, c) place's smooth and continuous can be led, and is illustrated in figure 3 the schematic diagram of strong punishment-stimulable type State Variable Weight function. Wherein 0 ＜ μ ＜ λ ＜ α ＜ β ＜ 1,0 ＜ c ＜ b ＜ a ＜ 1, a, b, c are Evaluation Strategy, and k is regulation coefficient, just has:

$\frac{b-c}{\mathrm{\&alpha;}-\mathrm{\&lambda;}}=\frac{a-b}{2(\mathrm{\&lambda;}-\mathrm{\&mu;})}.$

Owing to just punishment stage condition variable weight function is parabola, can make and just punish that stage condition variable weight function isDue to S_j(X) (α, c), (λ, b), (μ, a) place's smooth and continuous can be led, and then has:

$\left\{\begin{array}{c}S\left(\mathrm{\&lambda;}\right)=a_0+a_1\mathrm{\&lambda;}+a_2{\mathrm{\&lambda;}}^2=b\\ S^{\&prime;}\left(\mathrm{\&lambda;}\right)=a_1+2a_2\mathrm{\&lambda;}=-\frac{a-b}{\mathrm{\&mu;}-\mathrm{\&lambda;}}\\ S\left(\mathrm{\&alpha;}\right)=a_0+a_1\mathrm{\&alpha;}+a_2{\mathrm{\&alpha;}}^2=c\\ S^{\&prime;}\left(\mathrm{\&alpha;}\right)=a_1+2a_2\mathrm{\&alpha;}=0\end{array}\right.---\left(7\right)$

Can be obtained by formula (7):

If State Variable Weight function S (X) can be led in interval (0,1) smooth and continuous, then:

$S\left(X\right)=\left\{\begin{array}{cc}\frac{a-b}{\mathrm{\&lambda;}-\mathrm{\&mu;}}\mathrm{\&mu;}\ln \frac{\mathrm{\&mu;}}{x_j}+a,& x_j\&Element;(0,\mathrm{\&mu;}\&rsqb;\\ \frac{b-a}{\mathrm{\&lambda;}-\mathrm{\&mu;}}{x}_j+\frac{a\mathrm{\&lambda;}-b\mathrm{\&mu;}}{\mathrm{\&lambda;}-\mathrm{\&mu;}},& x_j\&Element;(\mathrm{\&mu;},\mathrm{\&lambda;}\&rsqb;\\ \frac{a-b}{2(\mathrm{\&lambda;}-\mathrm{\&mu;})(\mathrm{\&alpha;}-\mathrm{\&lambda;})}{(\mathrm{\&alpha;}-x_j)}^2+c,& x_j\&Element;(\mathrm{\&lambda;},\mathrm{\&alpha;})\\ c,& x_j\&Element;\&lsqb;\mathrm{\&alpha;},\mathrm{\&beta;}\&rsqb;\\ K(1-\mathrm{\&beta;})\ln \frac{1-\mathrm{\&beta;}}{1-x_j}+c,& x_j\&Element;(\mathrm{\&beta;},1)\end{array}\right.---\left(8\right)$

As 0 ＜ c ＜ 1,0 ＜ 1-β≤c,Time, S_j(X) it is strong local punishment-stimulable type State Variable Weight function.The State Variable Weight function more than built be proved meet definition 3 in axiom 1., 2., 3., 4..

Geological Environment Engineering feature according to embodiments of the present invention and the feature of Engineering-geological environmental quality, take μ =0.2, λ=0.4, α=0.6, β=0.8, a=0.5, b=0.3, c=0.2, k=1.5, (8) formula can obtain State Variable Weight letter Number:

$S\left(X\right)=\left\{\begin{array}{cc}0.2\ln (0.2/x_j)+0.5,& x_j\&Element;(0,0.2\&rsqb;\\ -x_j+0.7,& x_j\&Element;(0.2,0.4\&rsqb;\\ 2.5x_j^2-3x_j+1.1,& x_j\&Element;(0.4,0.6)\\ 0.2,& x_j\&Element;\&lsqb;0.6,0.8\&rsqb;\\ 0.3\ln (0.2/(1-x_j\left)\right)+0.2,& x_j\&Element;(0.8,1)\end{array}\right.$

Step 6 particularly as follows:

(601) the evaluation index matrix X ' of i-th basic evaluation unit is obtained respectively_iAnd become power weight matrix W_i, wherein I=1,2 ... total number of evaluation unit, X ' based on m, m_iMatrix, matrix X ' is tieed up for 1*n_iIn element representation be x '_ij, x′_ijThe desired value of the index that participates in evaluation and electing for jth in i-th basic evaluation unit, becomes power weight matrix W_iMatrix, this square is tieed up for 1*n Jth element in Zhen is denoted as w_ij, w_ijRepresent that jth in i-th basic evaluation unit participates in evaluation and electing the change power weight of index, j=1, 2 ... n, n represent total number of the index that participates in evaluation and electing；

(602) ask for jth in i-th basic evaluation unit participate in evaluation and electing index change power evaluation of estimate M_ij:

M_ij=w_ij·x_ij；

(603) variable synthesis evaluation of estimate M of i-th basic evaluation unit is asked for_i:

$M_i=\underset{j=1}{\overset{n}{\&Sigma;}}{M}_{ij}.$

In the embodiment of the present invention, GIS uses using grid point-like unit as achievement data carrier, vector planar unit conduct Basic evaluation analytic unit, then obtains variable synthesis evaluation of estimate M of each basis valency unit_i, i ∈ [1,6048], its value becomes Change scope is (0,1), tries to achieve variable synthesis evaluation codomain and is: 0.23846～0.84981.The embodiment of the present invention use non-etc. Distance method, chooses 0.05 and divides spacing for intermediate grade, the Urban Engineering Geological Environmental Quality of evaluation region is divided into 5 grades Subregion, is followed successively by from getting well to differing from: ＞ 0.80 (excellent), 0.75～0.80 (good), 0.70～0.75 (in), 0.65～0.70 (one As), ＜ 0.65 (poor).

In order to the reasonability that variable synthesis is evaluated is described, choose some basic evaluation unit, use the present invention to become Quan work Journey Geological Environmental Quality Assessment method and often two kinds of methods of power Engineering-geological environmental quality calculate their comprehensive evaluation value, Evaluation result contrast table is as shown in table 3.Metrics evaluation standard and divided rank according to formulating are it can be seen that this basic evaluation list 8 desired values of unit belong to top grade, and 3 desired values belong to good level, and 2 desired values belong to general level, and 2 desired values belong to poor level, The Practical Project geological environment of this basic evaluation unit is very poor.When often power is evaluated, the poor level value of 2 indexs and 2 indexs General level value is neutralized by the top grade value of 8 indexs, evaluates risk factor and reduces.When using variable weight model to evaluate, State Variable Weight function S (X) these index often weights that participate in evaluation and electing being re-started distribution, due to the magnanimous dynamical encourage of punishment dynamics, the final of this unit is combined Closing evaluation of estimate to reduce rapidly, opinion rating also reduces to difference level from general level.In like manner, for some basic evaluation unit, if Its 15 desired values broadly fall into top grade, and the top grade value of the index that participates in evaluation and electing can be encouraged, to them by State Variable Weight function S (X) Often power weight also can redistribute, the final comprehensive evaluation value of this unit can increase rapidly, and their opinion rating is also Can increase accordingly.Find through two kinds of method evaluation result relative analyses, variable synthesis evaluation methodology " outstanding " value to index Give encouragement, index " difference " is punished, " the typically value " of index is neither punished and does not encourages, fully demonstrated every The globality of Urban Engineering Geological Environmental Quality is affected by index internal diversity, greatly improves Engineering-geological environmental quality The credibility of grade and reliability.

Table 3 i-th basic evaluation unit evaluation Comparative result

Claims (7)

1. fuzzy a change weighs Engineering-geological environmental quality method, it is characterised in that the method comprises the steps:

(1) Engineering-geological environmental quality hierarchy structure model is set up, including general objective layer, sub-goal layer and evaluation index Layer, the described general objective that general objective layer is this Engineering-geological environmental quality, described sub-goal layer is and general objective pair The multiple sub-goals answered, each sub-goal includes multiple index that participates in evaluation and electing, all index composition evaluation index layers that participate in evaluation and electing；

(2) all indexs that participate in evaluation and electing obtained in region to be evaluated in Engineering-geological environmental quality hierarchy structure model are corresponding Engineering geology data；

(3) engineering geology data step (2) obtained import GIS software, utilize GIS software to set up the space in region to be evaluated Data base, and region to be evaluated is divided into m basic evaluation unit；

(4) the evaluation index matrix of each basic evaluation unit is set up according to the engineering geology data of each basic evaluation unit；

(5) the change power weight matrix that each basic evaluation unit is corresponding is asked for respectively；

(6) according to evaluation index matrix corresponding to base unit and become power weight matrix and ask for each basic evaluation unit respectively Variable synthesis evaluation of estimate；

(7) by the variable synthesis evaluation of estimate of each basic evaluation unit respectively with set Urban Engineering Geological Environmental Quality grade classification Value compares, and obtains the Urban Engineering Geological Environmental Quality grade of each basic evaluation unit in region to be evaluated；

(8) use clustering procedure to be clustered by the basic evaluation unit with identical Urban Engineering Geological Environmental Quality grade, obtain work Journey complex assessment of geological environment block plan.

One the most according to claim 1 is fuzzy becomes power Engineering-geological environmental quality method, it is characterised in that step (3) comprise the steps:

(301) the engineering geology data in region to be evaluated are imported GIS software, treat the Index Establishment list that respectively participates in evaluation and electing of evaluation region Item achievement data figure layer；

(302) it is standardized single index data layer processing；

(303) each single index data layer treating evaluation region carries out rasterizing process, and region to be evaluated is divided into m Basic evaluation unit.

One the most according to claim 1 is fuzzy becomes power Engineering-geological environmental quality method, it is characterised in that step (4) particularly as follows:

According to the index divided rank standard that participates in evaluation and electing, ask for regional evaluation index matrix X ' to be evaluated:

$X^{\&prime;}=\left[\begin{array}{c}{X}_1^{\&prime;}\\ X_2^{\&prime;}\\ ...\\ X_m^{\&prime;}\end{array}\right],$

Element in matrix X ' is designated as X '_i, X '_iFor the evaluation index matrix of i-th basic evaluation unit, i=1,2 ... m, m Based on total number of evaluation unit；

Specifically, the evaluation index matrix X of i-th basic evaluation unit_i' it is:

X′_i=[x '_i1 x′_i2 … x′_in],

Matrix X '_iIn element be denoted as x '_ij, x '_ijRepresent that jth in i-th basic evaluation unit participates in evaluation and electing the desired value of index, j =1,2 ..., n, n represent total number of the index that participates in evaluation and electing.

One the most according to claim 1 is fuzzy becomes power Engineering-geological environmental quality method, it is characterised in that step (5) particularly as follows:

(501) analytic hierarchy process (AHP) is used to ask in each basic evaluation unit evaluation index layer respectively for the Chang Quan of general objective layer Weight matrix, often the element in power weight matrix is the often weight of index of respectively participating in evaluation and electing；

(502) strong punishment-stimulable type State Variable Weight function is built, respectively in the often power weight matrix of each basic evaluation unit Each normal power weight redistribute, and then obtain each basic evaluation unit change power weight matrix, become power weight matrix In element be respectively participate in evaluation and electing index change power weight.

One the most according to claim 4 is fuzzy becomes power Engineering-geological environmental quality method, it is characterised in that described Step (501) in ask for the often power weight matrix of each basic evaluation unit and ask for especially by following method:

(501a) 1～9 scaling laws in analytic hierarchy process (AHP) are used to set up sub-goal layer to general objective i-th basic evaluation unit The judgment matrix B-A of layer and the judgment matrix C-B, wherein i=1,2 of evaluation index layer antithetical phrase destination layer ... based on m, m Total number of evaluation unit；

(501b) ask for often weighing weight matrix according to judgment matrix B-A and C-BThe most often power weight square Battle array W_i ⁰In element be denoted as Represent that jth in i-th basic evaluation unit participates in evaluation and electing the often weight of index, j=1, 2 ... n, n represent total number of the index that participates in evaluation and electing.

One the most according to claim 5 is fuzzy becomes power Engineering-geological environmental quality method, it is characterised in that described Step (502) particularly as follows:

(502a) according to engineering geological environment evaluation feature and variable-weight theory, strong punishment-stimulable type State Variable Weight function S is built (X):

$S\left(X\right)=\left\{\begin{array}{cc}\frac{a-b}{\mathrm{\&lambda;}-\mathrm{\&mu;}}\mathrm{\&mu;}\ln \frac{\mathrm{\&mu;}}{x_j}+a,& x_j\&Element;(0,\mathrm{\&mu;}\&rsqb;\\ \frac{b-a}{\mathrm{\&lambda;}-\mathrm{\&mu;}}{x}_j+\frac{a\mathrm{\&lambda;}-b\mathrm{\&mu;}}{\mathrm{\&lambda;}-\mathrm{\&mu;}},& x_j\&Element;(\mathrm{\&mu;},\mathrm{\&lambda;}\&rsqb;\\ \frac{a-b}{2\left(\mathrm{\&lambda;}-\mathrm{\&mu;}\right)\left(\mathrm{\&alpha;}-\mathrm{\&lambda;}\right)}{\left(\mathrm{\&alpha;}-x_j\right)}^2+c,& x_j\&Element;\left(\mathrm{\&lambda;},\mathrm{\&alpha;}\right)\\ c,& x_j\&Element;\&lsqb;\mathrm{\&alpha;},\mathrm{\&beta;}\&rsqb;\\ k\left(1-\mathrm{\&beta;}\right)\ln \frac{1-\mathrm{\&beta;}}{1-x_j}+c,& x_j\&Element;\left(\mathrm{\&beta;},1\right)\end{array}\right.,$

Wherein, x_jFor state value, 0 ＜ μ ＜ λ ＜ α ＜ β ＜ 1,0 ＜ c ＜ b ＜ a ＜ 1, μ is extra-heavy punishment level, and λ is for punishing by force water Flat, α for just to punish that level, β are stimulation level, a, b and the State Variable Weight function that c respectively state value is that μ, λ are corresponding with during α Value, k is regulation coefficient；

(502b) the evaluation index matrix X ' of i-th basic evaluation unit is obtained_i, by X '_iMiddle element is denoted as x '_ij, x '_ijRepresent the In i basic evaluation unit, jth participates in evaluation and electing the desired value of index, by x '_ijState value respectively as State Variable Weight function S (X) Ask for jth in i-th basic evaluation unit to participate in evaluation and electing the State Variable Weight functional value S of index_ij(X), i=1,2 ... m, m are base Total number of plinth evaluation unit, j=1,2 ..., n, n represent total number of the index that participates in evaluation and electing；

(502c) the change power weight matrix W of i-th basic evaluation unit is obtained_i=[w_i1 w_i2 … w_in], wherein become power weight square Element in Zhen is denoted as w_ij, w_ijRepresent i-th basic evaluation unit jth participate in evaluation and electing index change power weight, w_ijObtained by following formula Take:

$w_{ij}=\frac{w_{ij}^0\&CenterDot;S_{ij}\left(X\right)}{\underset{j=1}{\overset{n}{\&Sigma;}}(w_{ij}^0\&CenterDot;S_{ij}(X\left)\right)}.$

One the most according to claim 1 is fuzzy becomes power Engineering-geological environmental quality method, it is characterised in that described Step (6) particularly as follows:

(601) the evaluation index matrix X ' of i-th basic evaluation unit is obtained respectively_iAnd become power weight matrix W_i, wherein i= 1,2 ... total number of evaluation unit, X ' based on m, m_iMatrix, matrix X ' is tieed up for 1*n_iIn element representation be x '_ij, x '_ij The desired value of the index that participates in evaluation and electing for jth in i-th basic evaluation unit, becomes power weight matrix W_iMatrix is tieed up, in this matrix for 1*n Jth element be denoted as w_ij, w_ijRepresent that jth in i-th basic evaluation unit participates in evaluation and electing the change power weight of index, j=1, 2 ... n, n represent total number of the index that participates in evaluation and electing；

(602) ask for jth in i-th basic evaluation unit participate in evaluation and electing index change power evaluation of estimate M_ij: M_ij=w_ij·x_ij；

(603) variable synthesis evaluation of estimate M of i-th basic evaluation unit is asked for_i:

$M_i=\underset{j=1}{\overset{n}{\&Sigma;}}{M}_{ij}.$