CN103593560A - Comprehensive assessment and detection method for safety of existing glass curtain wall - Google Patents

Abstract

The invention relates to a comprehensive assessment and detection method for safety of an existing glass curtain wall. The method includes the steps that according to the relations among all index factors affecting the safety of the glass curtain wall, a hierarchical model of a hierarchical structure is established; index states of the bottom layer in the hierarchical model of the hierarchical structure are determined, assessed values of all indexes of the bottom layer are judged, and weights of all indexes of all layers in the hierarchical model of the hierarchical structure are calculated; according to the hierarchical model of the hierarchical structure, the assessed values of the bottom layer indexes and the weights of all the indexes, comprehensive assessment is performed on the safety of the glass curtain wall on the basis of a fuzzy theory. An area method is introduced into the glass curtain wall assessment, the comprehensive assessment and detection method for the safety of the glass curtain wall is established, the area method and the uncertain hierarchical method are combined, the qualitative indexes and quantitative indexes are fully considered, the safety level of the structure is judged through a maximum membership degree neighboring judgment method, and an assessed result expresses the safety of the glass curtain wall in a quantization mode, so that the assessment conclusion is more objective and rigorous.

Description

A kind of existing glass curtain wall security comprehensive assessment detection method

Technical field

The invention belongs to technical field of buildings, be specifically related to a kind of existing glass curtain wall security comprehensive assessment detection method.

Background technology

Abroad, from the middle and later periods in 20th century, the development of improving to glass curtain wall industry of the raising of material property and operating technique has brought powerful power, and curtain wall industry has obtained fast development, and has also formed seriation in the research of glass curtain wall.In the glass curtain wall safety issue Shang， U.S., carry out the research aspect serviceable life to curtain wall engineering, and formulated the in-situ check and test method of having issued relevant structural silicone adhesive strength character, ageing properties.

Present stage, various countries paid much attention to, under the prerequisite of glass curtain wall security, the method for evaluating safety of various glass curtain walls also have been carried out to a large amount of research.The research method that is applied to curtain wall field mainly comprises:

(1) research method based on Reliability Theory: this research method is the ingredient curtain wall panel building curtain wall system, curtain wall anchoring securing member, tackifier, curtain wall support system (is analysed frame, steelframe, prestressed cable etc.), observe respectively their behaviour in service, the health status of each ingredient is made respectively to appropriate evaluation, then, according to Structural Reliability Theory, in conjunction with the evaluation result to each ingredient of Curtain Wall, draw contribution-flexible strategy and life-span reduction coefficient that the health status of each ingredient is made the health status of whole curtain wall construction system, thereby can obtain the nominal residual life of whole curtain wall construction system.It is basis that the method needs a large amount of project data, and system dependability is calculated difficulty.

(2) MEDIC appraisal procedure: be used for the permanance of the relevant unit of predict good after using certain time limit, this is a kind of comprehensive to a large amount of investigation of existing structure and experience in the past, the method for simultaneously using relevant prior probability knowledge to assess the unit of existing structure.Basic Shang， China curtain wall development that the method must be based upon a large amount of engineering practice data only comprises 20 years history, and time data is abundant not.

(3) based on Visual Inspection method: the method for carrying out man-rate according to Visual Inspection. mainly by experienced curtain wall technology personnel, curtain wall is respectively formed to member and carry out detailed inspection, and evaluate according to check result.This kind of method only can be carried out qualitative evaluation to structure, but degree of injury that can not quantitative, and artificial objective factor is more.

(4) expertise rating method: existing structure is assessed with thering is the computer system that is equivalent to expertise and experience level.Due in curtain wall field, no matter be at design, construction or management aspect, the factor that all exists some probabilistic factors or theoretical research to relate in addition, these factors are difficult to calculate with numerical model mostly, adopt expertise evaluation rule well to solve, but the method popularization is poor.

(5) analytical hierarchy process: adopt the method to obtain each member significance level of structure and interactional relation, by multi-layer Fuzzy Evaluation and scoring, can the large numerous and diverse evaluation work of simplification amount.The method adopts complete determinacy quantization method, cannot get rid of people's impact on result on the uncertainty existing in the diversity of the understanding of things and Appraisal process and ambiguity.

Summary of the invention

The object of the invention is to solve the problems of the technologies described above and a kind of existing glass curtain wall security comprehensive assessment detection method is provided, the method combines qualitative index with quantitative target, area-method is combined with uncertain analytical hierarchy process, and close on the residing safe class of diagnostic method judgement structure by maximum membership degree, evaluation result is expressed the safety grades of glass curtain wall structure with quantized versions.

For achieving the above object, the present invention is by the following technical solutions:

A glass curtain wall security comprehensive assessment detection method, comprises the following steps:

Step 1: according to the relation affecting between each index factor of glass curtain wall security, set up hierarchical structure hierarchical model;

Step 2: determine bottom index state in described hierarchical structure hierarchical model, the assessed value of judgement bottom indices, calculates the weight of each index in each level in described hierarchical structure hierarchical model;

Step 3: according to the weight of described hierarchical structure hierarchical model, bottom index evaluation value and each index, utilize fuzzy theory to carry out comprehensive assessment to the security of existing glass curtain wall.

In described definite described hierarchical structure hierarchical model, bottom index state comprises definite non-quantizating index state and quantizating index state:

Determining that non-quantizating index state refers to by multidigit expert checks the presentation quality scene inspection of various members, scene utilizes interval number scoring to various member status, adopt the method for collection Data-Statistics to calculate, try to achieve the bottom index score value of comprehensive each expert opinion;

Determining quantizating index state refers to, each index attribute value is carried out to nondimensionalization processing, be converted into the score value between 0-100;

It is to adopt following methods to calculate that the described weight to each index in each level in described hierarchical structure hierarchical model is calculated:

Adopt the area-method in building to calculate weight quantitative bearing capacity index: according to the transparency area of impact after component failure, to carry out the relative importance of comparison means, calculate the weighted value of member.

$A_i=\frac{S_i}{\mathrm{\Σ}{S}_i},i=\mathrm{1,2,3}...,n$

In formula: Ai is the influence area of member weighted value ，SiWei unit component failure;

To qualitative index, the bottom index that comprises physical index and presentation quality adopts uncertain analytical hierarchy process to calculate weight: by expert, the situation of glass curtain wall is carried out to scene inspection investigation, determine bottom index weights in hierarchical model, adopt 1-9 scaling law to set up Uncertain Judgement Matrix, utilize uncertain analytical hierarchy process computation rule to calculate the corresponding weight of the judgment matrix interval that every expert provides, finally utilize the theoretical meter of collection Data-Statistics to calculate all experts' weight interval, draw the final weight between index.

Described step 3 specifically comprises the following steps:

By present stage, the security performance evaluation of China's glass curtain wall is divided into A, B, C, tetra-grades of D, adopting centesimal system to carry out assignment is A level (100～95), B level (85～75), C level (65～55) D level (45～0), sets up the membership function function of glass curtain wall safety evaluatio; Utilize following formula to carry out vectorial Fuzzy processing to the assessed value of bottom indices, draw respective mode gelatinization vector, judgement structure safety grades of living in, draws glass curtain wall safety evaluatio conclusion,

$\begin{array}{c}V=W.R=(W_1,W_2,...,W_m).{\left({\mathrm{\μ}}_{\mathrm{ij}}\right)}_{m\×5}\\ =(W_1,W_2,...,W_m).\left[\begin{array}{cccc}{\mathrm{\μ}}_{11}& {\mathrm{\μ}}_{12}& ...& {\mathrm{\μ}}_{15}\\ {\mathrm{\μ}}_{21}& {\mathrm{\μ}}_{22}& ...& {\mathrm{\μ}}_{25}\\ ...& ...& ...& ...\\ {\mathrm{\μ}}_{m1}& {\mathrm{\μ}}_{m2}& ...& {\mathrm{\μ}}_{m5}\end{array}\right]\\ =(V_1,V_2,V_3,V_4,V_5)\end{array}$

In formula, R is obfuscation vector, and Vi represents the final degree of membership of each index, and Wi represents the weight coefficient of index, and μ ij represents index evaluation value.

The present invention by introducing area-method in glass curtain wall assessment, set up existing glass curtain wall security comprehensive assessment detection method, area-method and uncertain stratification are combined, take into full account qualitative index and quantitative target, by maximum membership degree, close on the residing safe class of diagnostic method judgement structure, assessment result is expressed the security of glass curtain wall structure with quantized versions, make assessment result more objective more rigorous.

Accompanying drawing explanation

Figure 1 shows that the schematic diagram of existing glass curtain wall security comprehensive assessment detection method provided by the invention;

Figure 2 shows that hierarchical structure hierarchical model provided by the invention;

Figure 3 shows that membership function function provided by the invention.

Embodiment

Below, in conjunction with example, substantive distinguishing features of the present invention and advantage are further described, but the present invention is not limited to listed embodiment.

The present invention be take the thought of analytical hierarchy process and is basis, in conjunction with a large amount of experimental studies and finite element analysis, analyzed the factor that affects existing glass curtain wall security, simultaneously can Information Monitoring according to scene, set up Safety Evaluation and the index system of existing glass curtain wall; Combined curtain wall architectural characteristic is introduced area-method in glass curtain wall safety evaluation; The methods such as area-method and uncertain analytical hierarchy process, collection Data-Statistics theory, fuzzy theory are combined and are applied in the safety evaluation of curtain wall construction, reflected ambiguity and the uncertainty of structure safety effects factor; Set up existing glass curtain wall security comprehensive assessment detection method, and theoretical according to Fuzzy Reliability, qualitative index is combined with quantitative target, to the computing that increases by degrees of indexs at different levels in Analytic Hierarchy Process Model, and close on the residing safe class of diagnostic method judgement structure by maximum membership degree, assessment result is expressed the safety grades of glass curtain wall structure with quantized versions.

Consult Fig. 1, existing glass curtain wall security comprehensive assessment detection method, comprising:

Step 1, according to the relation affecting between each index factor of glass curtain wall security, set up hierarchical structure hierarchical model;

Utilize the analytical hierarchy process in systems engineering, by existing glass curtain wall structure to certain state influence degree close or contact factor formation one deck of putting together more closely, set up the hierarchical structure model of multilayer.This model is shown in Figure 2, can determine the set of factors U=(U of impact evaluation ₁, U ₂, U ₃), for the set of factors U that affects glass curtain wall security, its assessment factor subset comprises mechanical performance index U ₁, physical performance index U ₂, quality index appearance U ₃, a plurality of evaluation indexes are pressed attribute different grouping, and every group as a level, by top (destination layer U), middle layer (one-level evaluation index U _i, i=1,2 ..., n), the bottom (secondary evaluation index U _ij, i=1,2 ..., n, j=1,2 ..., form n) is lined up, and forms described staircase structure hierarchical model.

Step 2, determines bottom index state in described hierarchical structure hierarchical model, and the assessed value of judgement bottom indices, calculates the weight of each index in each level in described hierarchical structure hierarchical model;

In described definite described hierarchical structure hierarchical model, bottom index state comprises definite non-quantizating index state and quantizating index state;

Determining that non-quantizating index state refers to by multidigit expert checks the presentation quality scene inspection of various members, scene utilizes interval number scoring to various member status, adopt the method for collection Data-Statistics to calculate, try to achieve the bottom index score value of comprehensive each expert opinion;

Determining quantizating index state refers to, each index attribute value is carried out to nondimensionalization processing, be converted into the score value between 0-100.

Because each index is different to the importance degree of last layer target index, so the weight of each index is also different.With w=(w ₁, w ₂..., w _n) expression, wherein w _irepresent lower one deck index factor U _icorresponding weight.For the feature of different indexs, quantitative bearing capacity index is adopted to the area-method in building, by calculating the floor area of building affecting after each component failure, determine this member weight.To qualitative index, as the bottom index of physical index and presentation quality adopts uncertain analytical hierarchy process [7], weight is calculated.

Area-method is calculated weight

Glass curtain wall is to be suspended on architectural outer enclosed structure, bears side direction wind load and earthquake load.By Force transmission partses such as glass, crossbeam, column (pull bar, drag-line), crab-bolts, power passes in agent structure the most at last.Therefore certain component failure in structure, may cause other component failure on floor area of building, affects the security of structure.According to the transparency area of impact after component failure, carry out the relative importance of comparison means, area-method is calculated weight.Area-method computing formula:

$A_i=\frac{S_i}{\mathrm{\Σ}{S}_i},i=\mathrm{1,2,3}...,n......\left(1\right)$

Wherein, Ai is the influence area of member weighted value ，SiWei unit component failure.

Combined curtain wall structure feature of the present invention, solves middle employing area-method in the weight of each index of load-bearing capacity and calculates.

(1) transparency area that the inefficacy area after glassbreak is actual destruction.

(2) in Curtain wall with open frame, glass is directly supported on horizontal frame, and inefficacy area when crossbeam lost efficacy is supported whole upper piece transparency area.

(3) what column was born is the interior respectively glass gravity load of half of both sides lattice, and therefore, when column lost efficacy, inefficacy area is the interior respectively transparency area summation of half of lattice.

(4) pull bar is just the same with the stress performance of column in full glass curtain wall, by the respectively transparency area summation of half calculating in lattice.

(5) crab-bolt is fixed to structure in agent structure, be connected with column, once inefficacy pillar construction by complete failure, the transparency area of bearing is each half transparency area summation in the lattice of both sides.

Uncertain analytical hierarchy process calculates weight

So-called uncertain analytical hierarchy process adopts interval number Judgement Matricies exactly, and interval number meets the requirement of definition 1.

Definition 1: if A=is (A _ij) n * n, wherein A _ij=[a _ij, b _ij], claim that A is an Interval Judgment Matrix, and

1) A _ij=[a _ij, b _ij], and 9>=b _ij>=a _ij>=1/9;

2）A _ii=[1，1]i=1,2,…,n-1,n;

3）A _ij=1/A _ji,A _ji=[b _ij ^-1,a _ij ^-1]

According to above uncertain step analysis ratio juris, two two indexes are relatively adopted to Interval scales, set up Interval Judgment Matrix.Present stage, the solution interval method of counting judgment matrix weight mainly comprised the method for using Gradient Features vector to solve weight, utilized the algorithm of interval number to calculate weight, error analysis practical calculation method etc.In view of analytical hierarchy process influence factor is numerous and complicacy, select most widely used general, the easiest error analysis practical calculation method to calculate weight.

Uncertain analytical hierarchy process weight calculation formula is as follows:

Uncertain judgment matrix A=(A _ij) _{n * n}, A _ij=[a _ij, b _ij]

$m_{\mathrm{ij}}=2n\sqrt{\underset{k=1}{\overset{n}{\mathrm{\Π}}}\frac{a_{\mathrm{ik}}{b}_{\mathrm{ik}}}{a_{\mathrm{jk}}{b}_{\mathrm{jk}}}}......\left(2\right)$

$g_i={\left(\mathrm{\Π}\underset{j=1}{\overset{n}{m_{\mathrm{ij}}}}\right)}^{1/n}i=\mathrm{1,2},...,n......\left(3\right)$

${\left({\mathrm{\Δkg}}_i\right)}^2=\frac{1}{{\left[\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{m}_{\mathrm{ij}}\right]}^4}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\Δk}}^2\left(m_{\mathrm{ij}}\right),k=\mathrm{1,2}......\left(4\right)$

Δ ₁m _ij=（m _ij-a _ij），Δ ₂m _ij=（b _ij-m _ij）……（5）

The weight of interval judgment matrix is: w=(w ₁, w ₂..., w _j)

Wherein

$w_j=(w_1^{\left[j\right]},w_2^{\left[j\right]})=(g_j-{\mathrm{\Δ}}_1{g}_j,g_j+{\mathrm{\Δ}}_2{g}_j)j=\mathrm{1,2},...,n......\left(6\right)$

The n drawing weight is interval $[w_1^{\left[1\right]},w_2^{\left[1\right]}],\left[w_1^{\left[2\right]}{w}_2^{\left[2\right]}\right],......,[w_1^{\left[n\right]},w_2^{\left[n\right]}].$ Adopt collection Data-Statistics theoretical, according to formula (7), interval number is converted into precise figure and obtains comprehensive evaluation value.

$G\left(w\right)=\frac{1}{2}\frac{\underset{k=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ω}}_k[{\left(w_2^{\left(k\right)}\right)}^2-{\left(w_1^{\left(k\right)}\right)}^2]}{\underset{k=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\ω}}_k[w_2^{\left(k\right)}-w_1^{\left(k\right)}]}......\left(7\right)$

ω _kfor expert opinion coefficient, owing to being a concrete and complicated problem to individual expert level judgement, the present invention only considers that each expert level is the same, i.e. ω _kall equate.

Step 3, according to the weight of described hierarchical structure hierarchical model, bottom index evaluation value and each index, utilize fuzzy theory to carry out comprehensive assessment to the security of existing glass curtain wall;

Each index is divided into quantifiable indicator and non-quantizating index, because the evaluation criteria of bottom index is different, needs to process through nondimensionalization.For mechanical performance index and load-carrying properties index quantifiable indicator, adopting centesimal system and linear dimension is that 1 change is processed, and draws unified assessed value.For outward appearance damage criterion, adopt uncertain analytical hierarchy process by the assessment at expert scene, calculate the assessed value of each index; Then carry out Fuzzy processing, draw corresponding obfuscation vector, the residing safety grades of judgement structure.Wherein, key is to establish suitable membership function function.

The representative function form proposing according to professor Wang Guangyuan, by present stage, the security performance evaluation of China's glass curtain wall is divided into A, B, C, tetra-grades of D, adopting centesimal system to carry out assignment is A level (100～95), B level (85～75), C level (65～55) D level (45～0) has been set up the membership function function of glass curtain wall safety evaluation, see Fig. 3, then utilize following formula (8)～(11) to carry out vectorial Fuzzy processing

${\mathrm{\&mu;}}_a=\left\{\begin{array}{cc}1& 95\&le;s<100\\ \frac{s-85}{10}& 85\&le;s<95\end{array}\right.......\left(8\right)$

${\mathrm{\&mu;}}_b=\left\{\begin{array}{cc}\frac{95-s}{10}& 85\&le;s<95\\ 1& 75\&le;s<85\\ \frac{s-65}{10}& 65\&le;s<75\end{array}\right.......\left(9\right)$

${\mathrm{\&mu;}}_c=\left\{\begin{array}{cc}\frac{75-s}{10}& 65\&le;s<75\\ 1& 55\&le;s<65\\ \frac{s-45}{10}& 45\&le;s<55\end{array}\right.......\left(10\right)$

${\mathrm{\&mu;}}_d=\left\{\begin{array}{cc}\frac{55-s}{10}& 45\&le;s<55\\ 1& 0\&le;s<45\end{array}\right.......\left(11\right)$

Standard score is carried out to fuzzy synthesis processing, and then draw fuzzy synthesis computing formula, as formula (12), according to this formula, carry out fuzzy composition computing, obtain third level fuzzy evaluation result V, in like manner can obtain firsts and seconds fuzzy evaluation result;

$\begin{array}{c}V=W.R=(W_1,W_2,...,W_m).{\left(r_{\mathrm{ij}}\right)}_{m\&times;5}\\ =(W_1,W_2,...,W_m).\left[\begin{array}{cccc}{r}_{11}& r_{12}& ...& r_{15}\\ r_{21}& r_{22}& ...& r_{25}\\ ...& ...& ...& ...\\ r_{m1}& r_{m2}& ...& r_{m5}\end{array}\right]\end{array}=(V_1,V_2{,V}_3,V_4,V_5)......\left(12\right)$

In formula, R is obfuscation vector, and Vi represents the final degree of membership of each index, and Wi represents the weight coefficient of index, and μ ij represents index evaluation value.

Bonded area method of the present invention and uncertain analytical hierarchy process, carried out comprehensive assessment to the security of existing glass curtain wall, on the one hand, to load-bearing capacity index, adopts area-method computation structure weight, avoided the unworthiness of fuzzy evaluation method to force structure; On the other hand, for other qualitative index, adopt interval number to replace perfect number Judgement Matricies, calculate each index weights, improve science, objectivity and the accuracy of expert judgments.Actual assessment checking

The existing Glass Curtain Wall Projects of Beijing hospital, in August, 1996, be completed and use, two layers～tetra-layers is typical concealed frame glass curtain wall, the outer exterior material of curtain wall is all simple glass, curtain wall lattice (B * H) 1.0m * 1.5m, inner structure crossbeam and column adopt the aluminium alloy of 6063T5, floor height 3.6m.Scene is carried out sampling Detection in 5% ratio to the outward appearance of curtain wall and quality, and to part test specimen performance research experiment, outcome record is as follows:

The tensile strength test result of column is 246MPa, aluminum alloy frame hardness number is 12, crab-bolt pulling capacity is 19.7N, structural silicone adhesive tested for hardness is 36, and the mid-span deflection of glass under 1000Pa Action of Wind pressure is 3.79mm, and the mid-span deflection of column is 1.18mm, glue thickness 11mm, thickness of glass 7.2mm, colloid bonding thickness 6mm, Aluminum alloy column wall thickness 3.86mm.Field condition is described below:

1, curtain wall opening leaf connecting portion becomes flexible obscission;

2, cladding glass major part is that simple glass does not have tempering not meet safety standard;

3, there is large-scale water leakage in outer facade fluid sealant serious aging;

4, partly there is the aging potential safety hazard that exists in glass and auxiliary frame syndeton glue.

(1) non-bearing power index weights calculates

The safety evaluation model of setting up according to Fig. 2, the set of factors U=(U to impact evaluation ₁, U ₂, U ₃), and middle layer and bottom index carry out expert's marking and comprehensive assessment, sets up following uncertain judgment matrix.

As space is limited, only enumerate the Interval Judgment Matrix of firsts and seconds evaluation index.As shown in table 1～3.

Table 1 glass curtain wall outward appearance is damaged judgment matrix and the calculating of each index weights

Judgment matrix and the weight calculation of each index of table 2 glass curtain wall the first level factor collection

Judgment matrix and the weight calculation of each index of table 3 glass curtain wall the second level factor collection

(2) bearing capacity index weight calculation

This curtain wall is typical frame concealed curtain wall, and curtain wall lattice is even, and a root post connects two crossbeams and bears construction weight, according to area-method, calculates Impersonal-weight Parameter Theory, and the influence area after unit component failure is respectively: glass 1.5m ², crossbeam 1.5m ², column 3.0m ², crab-bolt 3.0m ², so the weighted value of each member is respectively: glass 0.17, crossbeam 0.17, column 0.33, crab-bolt 0.33.

(3) non-quantification and quantizating index bottom evaluation of estimate are calculated

For non-quantizating index, by expert, according to field condition, assess, set up uncertain judgment matrix as shown in table 4, the final evaluation of estimate of calculating each index according to formula (2)～(10) is in Table (5).

Table 4 glass curtain wall outward appearance is damaged the judgment matrix of each index

For quantizating index, according to field measurement value and centesimal system equal interval quantizing formula, calculate the bottom evaluation of estimate of quantizating index.

Obtain after the actual evaluation value of each bottom index, then utilize Fig. 3 membership function function and formula (11)～(14) to carry out vectorial Fuzzy processing, form bottom comment fuzzy matrix R, and according to formula (15), carry out fuzzy composition computing by gained bottom comment fuzzy matrix R, draw third level fuzzy evaluation result V, computation process is in Table 5.In like manner can obtain firsts and seconds fuzzy evaluation and the results are shown in Table 6 and table 7.

Three grades of set of factors Fuzzy comprehensive evaluation results of table 5

Table 6 the second level factor collection Fuzzy comprehensive evaluation result

Table 7 Fuzzy comprehensive evaluation result

According to the contiguous diagnostic method of maximum membership degree, this glass curtain wall security performance is for the being subordinate to property the highest (0.5) of B level, therefore the final assessment grade of this structure is B level, curtain wall construction agent structure can meet security requirement, but exterior trim glass and structure glue, there is serious aging in fluid sealant, recommended replacement, to guarantee integrally-built security.

In conjunction with Practical Project, on the basis of test figure and on-the-spot expert judgments, the existing glass curtain wall of application the method Dui Mou hospital carries out safety evaluation, evaluation result is consistent with the assessment result that employing Shanghai City glass curtain wall security performance detects assessment technology rules (DGT J08-803-2005), illustrates that the method has certain science, feasibility and applicable.

Claims (3)

1. an existing glass curtain wall security comprehensive assessment detection method, is characterized in that, comprises the following steps:

Step 1: according to the relation affecting between each index factor of glass curtain wall security, set up hierarchical structure hierarchical model;

Step 2: determine bottom index state in described hierarchical structure hierarchical model, the assessed value of judgement bottom indices, calculates the weight of each index in each level in described hierarchical structure hierarchical model;

Step 3: according to the weight of described hierarchical structure hierarchical model, bottom index evaluation value and each index, utilize fuzzy theory to carry out comprehensive assessment to the security of existing glass curtain wall.

2. the security comprehensive assessment detection method of existing glass curtain wall according to claim 1, is characterized in that, describedly determines in described hierarchical structure hierarchical model that bottom index state comprises and determines non-quantizating index state and definite quantizating index state:

Determining that non-quantizating index state refers to by multidigit expert checks the presentation quality scene inspection of various members, scene utilizes interval number scoring to various member status, adopt the method for collection Data-Statistics to calculate, try to achieve the bottom index score value of comprehensive each expert opinion;

Determining quantizating index state refers to, each index attribute value is carried out to nondimensionalization processing, be converted into the score value between 0-100;

The described weight to each index in each level in described hierarchical structure hierarchical model is calculated and is adopted following methods to calculate:

Adopt the area-method in building to calculate weight quantitative bearing capacity index: according to the transparency area of impact after component failure, to carry out the relative importance of comparison means, calculate the weighted value of member:

$A_i=\frac{S_i}{\mathrm{\&Sigma;}{S}_i},i=\mathrm{1,2,3}...,n$

In formula: Ai is the influence area of member weighted value ，SiWei unit component failure;

To qualitative index, the bottom index that comprises physical index and presentation quality adopts uncertain analytical hierarchy process to calculate weight: by expert, the situation of glass curtain wall is carried out to scene inspection investigation, determine bottom index weights in hierarchical model, adopt 1-9 scaling law to set up Uncertain Judgement Matrix, utilize uncertain analytical hierarchy process computation rule to calculate the corresponding weight of the judgment matrix interval that every expert provides $[w_1^{\left[1\right]},w_2^{\left[2\right]}],[w_1^{\left[2\right]},w_2^{\left[2\right]}],......,[w_1^{\left[n\right]},w_2^{\left[n\right]}],$ Finally utilize the theoretical meter of collection Data-Statistics to calculate all experts' weight interval, draw the final weight between index.

3. the security comprehensive assessment detection method of existing glass curtain wall according to claim 1, is characterized in that, described step 3 specifically comprises the following steps:

By present stage, the security performance evaluation of China's glass curtain wall is divided into A, B, C, tetra-grades of D, adopting centesimal system to carry out assignment is A level (100～95), B level (85～75), C level (65～55) D level (45～0), sets up the membership function function of glass curtain wall safety evaluatio;

Utilize following formula to carry out vectorial Fuzzy processing to the assessed value of bottom indices, draw respective mode gelatinization vector, judgement structure safety grades of living in, draws glass curtain wall safety evaluatio conclusion;

$\begin{array}{c}V=W.R=(W_1,W_2,...,W_m).{\left({\mathrm{\&mu;}}_{\mathrm{ij}}\right)}_{m\&times;5}\\ =(W_1,W_2,...,W_m).\left[\begin{array}{cccc}{\mathrm{\&mu;}}_{11}& {\mathrm{\&mu;}}_{12}& ...& {\mathrm{\&mu;}}_{15}\\ {\mathrm{\&mu;}}_{21}& {\mathrm{\&mu;}}_{22}& ...& {\mathrm{\&mu;}}_{25}\\ ...& ...& ...& ...\\ {\mathrm{\&mu;}}_{m1}& {\mathrm{\&mu;}}_{m2}& ...& {\mathrm{\&mu;}}_{m5}\end{array}\right]\\ =(V_1,V_2,V_3,V_4,V_5)\end{array}$

In formula, R is obfuscation vector, and Vi represents the final degree of membership of each index, and Wi represents the weight coefficient of index, and μ ij represents index evaluation value.

Images