1. A wind damage analysis method of a short building enclosure member considering typhoon holding time effect is characterized by comprising the following steps:

step 1: probability of impact damage to a flying object

In strong wind environment such as typhoon, airflow often wraps around stones, branches and other components and carcasses of houses, and the probability of single door and window being damaged by impact of thrown objects is

p_d＝1-exp(-a×N_a×b×c×d) (1)

In the formula, N_aRepresenting the amount of debris that the environment surrounding the premises is a potential flyaway; a ═ Φ [ (U-42.2)/4.69]Showing the proportion of debris in the incoming flow direction to the thrown objects; b is 0.4/62.52 (U-15.63) represents the number of house hits by the flier; c represents the ratio of the windward area of a single door window to the windward area of the wall surface; d ═ Φ [ (U-21.88)/3.13]Representing the probability that the momentum of the thrown object exceeds the impact resistance limit of the door and window; in the above relation, U represents the 10-minute average wind speed at a height of 10 meters, and Φ (-) represents the standard Gaussian cumulative distribution function;

step 2: determining wind pressure extremum distribution

Removing the influence of the throwing objects, wherein whether the component is damaged or not depends on the load and the bearing capacity; let x (t) represent the wind pressure time course of a certain component, its extreme variable is represented as W, its cumulative probability distribution generally obeys extreme I-type (Gumbel) distribution

F(W)＝exp{-exp[-(W-δ_w)/ψ_w]} (2)

In the formula, delta_wAnd psi_wRespectively are a position parameter and a scale parameter to be determined; according to a semi-empirical formula, the two parameters can be obtained by the following relationship

In the formula, mu and sigma are respectively the statistical mean value and standard deviation of the wind pressure time course x (t); c. C_l、d_lIs a constant coefficient, theta_l、Ω_lThe skewness α depending on the time interval x (t)₃Kurtosis alpha₄And the mean zero crossing number η₀The specific expression is detailed in table 1; directly calculating skewness alpha through member wind pressure time-course samples₃Kurtosis alpha₄And the average zero crossing number eta₀Obtaining an extreme value distribution function of the wind pressure through formulas (3) and (4);

TABLE 1 empirical formula parameters

Table 1 parameters of empirical formulae

And step 3: wind pressure multivariate extremum joint probability distribution

In a typhoon environment, the wind pressures of different members of a low building envelope are related in space, so that member damages are related to each other; in order to consider the influence of wind pressure correlation, a multivariate probability model based on Nataf transformation is adopted to establish multivariate extreme value joint distribution of the wind pressure of the enclosure member; let W be [ W ]₁,W₂,…W_i,…W_j,…W_N]Expressing the non-Gaussian wind pressure extreme value vector, N expressing the number of the enclosure components to be considered, W_iEdge probability distribution F of_Wi(W_i) I is 1,2, …, N is estimated by the method in step 2; defining a group of N-element standard Gaussian vectors Z ═ Z₁,Z₂,…Z_i,…Z_j,…Z_N]Of variable Z_iAnd variable W_iThere is an equiprobable relationship between:

by Nataf transformation^[18]Joint probability distribution F of non-Gaussian vectors W_W(w) is expressed as:

F_W(w)＝Φ_N(z,Σ_Z) (6)

in the formula phi_NRepresenting an N-ary standard Gaussian joint cumulative distribution function; sigma_ZMatrix of correlation coefficients representing a standard gaussian vector Z, the elements of whichShould be based on the correlation coefficient matrix sigma of the non-Gaussian vector W_WMiddle corresponding elementAnd determining the specific relation as follows:

in the formula, mu_i(μ_j) And σ_i(σ_j) Respectively represent W_i(W_j) Mean and standard deviation of;representing a binary standard Gaussian joint probability density function; due to extreme value of wind pressure W_iSubject to Gumbel distribution, the above relationship is approximated as:

and 4, step 4: calculating the internal pressure of the open-hole house

In the wind-induced damage process, the enclosure components are successively damaged by throwing or wind pressure, and the hole opening working condition and the internal pressure of the house are continuously changed; house internal pressure W⁽ⁱ⁾Approximately the weighted average of the external pressure and the area of each opening, specifically

In the formula, A_kRepresents the area of the kth opening;indicating the external pressure at the kth opening; n is a radical of_oIndicating the number of openings of the house;

and 5: wind damage loss gradual analysis process

Introducing the method in the steps 1-4 into a wind damage gradual analysis process, and evaluating the loss of the building enclosure structure in the typhoon holding process through the following 10 steps:

(1) the number of the building enclosure components needing to be considered is set to be N, wherein the number of the doors and windows is M, so that the impact damage influence of the throwing objects is analyzed; selecting the sample capacity in the simulation analysis as N, and defining N rows and N columns of destruction matrix D_(N×n)Defining an internal pressure vector with N sets of failure condition samples corresponding to N membersCorresponding to the house internal pressure in the n groups of simulations; the house is intact at the initial moment, and the internal pressure is considered to be 0;

(2) according to the probability distribution information of the resistance of the building members, randomly simulating the resistance of the N enclosure members to obtain N groups of samples, and generating a resistance matrix R_(N×n)；

(3) According to the records of typhoon wind speed and wind direction, dividing the typhoon wind speed into a plurality of time steps by taking 10 minutes as a unit, and gradually analyzing from the 1 st time step;

(4) when the kth time step is entered, selecting the average wind speed and the wind direction of typhoon at the kth time step, calculating the impact damage probability vectors of the thrown objects of M doors and windows by the method in the step 1, and obtaining an impact damage probability matrix B of the thrown objects by n-dimensional expansion_(M×n)(ii) a At the same time, n groups of samples of M standard uniformly distributed variables are simulated to generate a random sampling matrix S_(M×n)(ii) a By comparing the destruction probability matrix B_(M×n)And a random sampling matrix S_(M×n)Determining the damage condition of the door and the window according to the size of the parity elements; wherein the impact damage probability value is greater than the sample value, indicating damage;

(5) according to the average wind speed and the wind direction of typhoon at the kth time step and the house wind pressure coefficient information, firstly, the external wind pressure of the enclosure member is calculated, and then the external wind pressure extreme value distribution of the N members is estimated by using the method in the step 2;

(6) carrying out N-element extreme value simulation work based on Nataf transformation, wherein the specific process comprises the following steps: acquiring a correlation coefficient matrix sigma based on wind pressure data of each member_WCalculating according to the formula (8) to obtain a matrix sigma_ZAnd performing Cholesky decomposition on the matrix to obtain a lower triangular matrixGenerating N groups of samples of N-element independent standard Gaussian vectors by Monte Carlo simulation, i.e. generating matrix U_(N×n)(ii) a (III) by the relationObtaining a sample matrix Z of N-element related standard Gaussian vectors_(N×n)(ii) a Matrix Z_(N×n)After equal probability transformation (formula (5)), further obtaining a sample matrix W_(N×n)N groups of samples of the external wind pressure extreme values of the N components;

(7) the net wind pressure extreme value of the enclosure member is the sum of the external wind pressure extreme value and the internal pressure of the house, namely, the external wind pressure extreme value matrix W is used_(N×n)Each line of (2) and the current state internal pressure vectorAdding to obtain a sample matrix Q of a net wind pressure extreme value_(N×n)(ii) a By comparing the net wind pressure extreme value matrix Q_(N×n)And resistance matrix R_(N×n)Determining the size of the parity elements to determine the damage condition of the component; wherein the net wind pressure value is greater than the resistance value, indicating a failure;

(8) determining the component damage condition according to the analysis results of the impact damage and the wind pressure damage of the flying objects in the steps (4) and (7), and updating the damage matrix D according to the component damage condition_(N×n)；

(9) If the destruction matrix changes, the internal pressure vector is updated by equation (9) according to the current destruction stateRepeating the steps (7) to (8) until the damage matrix is not changed any more;

(10) enabling k to be k +1, entering the next time step, and repeating the steps (4) to (9) until all time step analysis is finished;

then, estimating the damage probability of a certain member by analyzing a certain row in the damage matrix, namely a plurality of groups of simulation results corresponding to the damage condition of the certain member; the loss rate of each type of component in a group of simulation results is estimated by analyzing a certain column in the damage matrix, namely a group of simulation results corresponding to the damage conditions of all components.

Background

The damage of the buildings such as the residences and the factory buildings is extremely high in the loss caused by typhoons, and the orderly development of local production and life is seriously influenced. The damage of a short house under typhoon attack usually starts from an enclosure structure, wherein the enclosure components such as doors, windows, roof cladding, roof panels and the like firstly rush to the enclosure structure^[2]The pressure in the house is changed sharply, rainwater is poured in, and subsequent disasters such as house collapse are caused. Effectively analyzing and predicting typhoon disaster loss of the enclosure structure of the short building has positive effects on local disaster prevention and reduction and risk control.

Worldwide, typhoon (hurricane in north america) disasters have severe impact on north american countries such as the united states, with total us economic losses due to hurricanes during the 1996 to 2012 years only, as high as $ 2500 billion, with about 4000 deaths (e.g., NIST (national Institute of Standards and technology), measurements science R & D roadmap for windstorm and total accumulation activity reduction, NIST GCR 14-973-13, Gaithersburg, MD: NIST, 2014). At present, two representative typhoon disaster prediction models exist: the document of valid of a probabilistic model for hurricane interior emission projects in Florida (Pinelli J, Gurley K, Subramaniana C, Hamid S, Pita G., Reliability Engineering and System Safety,93: 1896-. Document HAZUS-MH Hurricane Model method.II: Damage and Loss Estimation (Vickery P J, Skerlj P F, Lin J, Twindale Jr L A, Young Michael A, Lavelle F M. Natural Hazards Review,7(2):94-103,2006b) is a HAZUS-MH Model issued by the Federal Emergency administration (FEMA), which considers the hold-time effect of hurricanes, analyzes the whole process of the house destruction in the Hurricane period, provides a stepwise analysis simulation method based on a wind field time sequence, and is used for the cumulative Loss of the enclosure in the Hurricane period.

Both employ load specifications for loss analysis caused by wind pressure, and both develop a missile disaster model to account for missile impact damage. Studies have shown that Wind pressure values based on load specifications may underestimate The extreme effects of Wind pressure (e.g., St. Pierre L S, Kopp G A, Surry D, Ho T C E., The UWO constraint to The NIST airborne data base for Wind weights on low weights: Part 2.Comparison of data with loads, Journal of Wind Engineering and Industrial Aerodynamics,93(1), 31-59,2005). The research on the wind-induced vulnerability of the enclosure structure of the short house based on wind tunnel test data is developed primarily. The document Vulnerability analysis of steel roof closing: infiluence of wind direction (Ji X, Huang G, Zhang X, Kopp G A., Engineering Structures,156: 587-. In the document of Wind-induced halyard assessment for low-rise building construction and construction site options (Journal of structural engineering,146, (4):04020039,2020), a house internal pressure dynamic analysis method considering the randomness of the opening of an enclosure structure is provided, and the vulnerability of vulnerable components such as doors, windows and roofs of common wooden houses is evaluated.

Disclosure of Invention

In view of the fact that wind tunnel test data are not combined in the conventional typhoon-duration-effect-considered wind damage assessment of the low-rise building envelope structure, and the low efficiency of the cyclic operation is involved, in order to more practically and efficiently assess the typhoon-duration wind damage of the low-rise building envelope structure, an improved wind damage gradual analysis method is developed. The method comprises the following contents: calculating the impact damage probability of the throwing object, determining the distribution of wind pressure extreme values, calculating the combined probability distribution of wind pressure multiple extreme values, calculating the change of internal pressure and establishing a wind damage gradual analysis process.

The method comprises the following specific steps:

step 1: probability of impact damage to a flying object

In strong wind environments such as typhoons and the like, airflow often wraps around stones, branches and other components and debris of houses, flying objects formed by the debris can impact fragile components such as doors and windows with certain probability to cause damage, and the probability that a single door and window is impacted and damaged by the flying objects is

p_d＝1-exp(-a×N_a×b×c×d) (1)

In the formula, N_aRepresenting the amount of debris that the environment surrounding the premises is a potential flyaway; a ═ Φ [ (U-42.2)/4.69]Showing the proportion of debris in the incoming flow direction to the thrown objects; b is 0.4/62.52 (U-15.63) represents the number of house hits by the flier; c represents the ratio of the windward area of a single door window to the windward area of the wall surface; d ═ Φ [ (U-21.88)/3.13]Indicating the probability that the momentum of the thrown object exceeds the impact resistance limit of the window or door. In the above relation, U represents the 10-minute average wind speed at a height of 10 meters, and Φ (·) represents a standard gaussian variable cumulative distribution function.

Step 2: determining wind pressure extremum distribution

The component failure, apart from the effects of the throws, is also dependent on the load and bearing capacity. Let x (t) represent the wind pressure time course of a certain component, its extreme variable is represented as W, its cumulative probability distribution generally obeys extreme I-type (Gumbel) distribution

F(W)＝exp{-exp[-(W-δ_w)/ψ_w]} (2)

In the formula, delta_wAnd psi_wRespectively, a position parameter and a scale parameter to be determined. According to a semi-empirical formula, the two parameters can be obtained by the following relationship

In the formula, mu and sigma are respectively the statistical mean value and standard deviation of the wind pressure time course x (t); c. C_l、d_lIs a constant coefficient, theta_l、Ω_lThe skewness α depending on the time interval x (t)₃Kurtosis alpha₄And the mean zero crossing number η₀The specific expression is shown in table 1. Directly calculating skewness alpha through member wind pressure time-course samples₃Kurtosis alpha₄And the average zero crossing number eta₀And the extreme value distribution function of the wind pressure can be quickly obtained through the formulas (3) and (4).

TABLE 1 empirical formula parameters

Table 1 parameters of empirical formulae

And step 3: wind pressure multivariate extremum joint probability distribution

In a typhoon environment, there is a spatial correlation of the wind pressure of different members of a low building envelope, such that member breakages are linked to each other. In order to consider the influence of wind pressure correlation, multivariate probability models based on Nataf transformation are adopted to establish multivariate extreme value joint distribution of the wind pressure of the enclosure member. Let W be [ W ]₁,W₂,…W_i,…W_j,…W_N]Expressing the non-Gaussian wind pressure extreme value vector, N expressing the number of the enclosure components to be considered, W_iEdge probability distribution ofEstimated by the method in step 2. Defining a group of N-element standard Gaussian vectors Z ═ Z₁,Z₂,…Z_i,…Z_j,…Z_N]Of variable Z_iAnd variable W_iThere is an equiprobable relationship between:

by Nataf transformation^[18]Joint probability distribution F of non-Gaussian vectors W_W(w) is expressed as:

F_W(w)＝Φ_N(z,Σ_Z) (6)

in the formula phi_NRepresenting an N-ary standard Gaussian joint cumulative distribution function; sigma_ZMatrix of correlation coefficients representing a standard gaussian vector Z, the elements of whichShould be based on the correlation coefficient matrix sigma of the non-Gaussian vector W_WMiddle corresponding elementAnd determining the specific relation as follows:

in the formula, mu_i(μ_j) And σ_i(σ_j) Respectively represent W_i(W_j) Mean and standard deviation of;representing a binary standard gaussian joint probability density function. Due to extreme value of wind pressure W_iSubject to Gumbel distribution, the above relationship is approximated as:

and 4, step 4: calculating the internal pressure of the open-hole house

During wind-induced damage, the enclosure members are damaged successively by the thrown objects or wind pressure, and the opening condition and the internal pressure of the house are changed continuously. House internal pressure W⁽ⁱ⁾Approximately the weighted average of the external pressure and the area of each opening, specifically

In the formula, A_kRepresents the area of the kth opening;indicating the external pressure at the kth opening; n is a radical of_oIndicating the number of openings in the house.

And 5: wind damage loss gradual analysis process

Introducing the method in the steps 1-4 into a wind damage gradual analysis process, and evaluating the loss of the building enclosure structure in the typhoon holding process through the following 10 steps:

(1) the number of the building enclosure components needing to be considered is set to be N, wherein the number of the doors and windows is M, so that the impact damage influence of the throwing objects is analyzed; selecting the sample capacity in the simulation analysis as N, and defining N rows and N columns of destruction matrix D_(N×n)Defining an internal pressure vector with N sets of failure condition samples corresponding to N membersCorresponding to the house internal pressure in the n groups of simulations; the house was intact at the initial time and the internal pressure was considered to be 0.

(2) According to the probability distribution information of the resistance of the building members, randomly simulating the resistance of the N enclosure members to obtain N groups of samples, and generating a resistance matrix R_(N×n)。

(3) According to the records of typhoon wind speed and wind direction, the typhoon wind speed and wind direction are divided into a plurality of time steps by taking 10 minutes as a unit, and the analysis is carried out step by step from the 1 st time step.

(4) When the kth time step is entered, selecting the average wind speed and the wind direction of typhoon at the kth time step, calculating the impact damage probability vectors of the thrown objects of M doors and windows by the method in the step 1, and obtaining an impact damage probability matrix B of the thrown objects by n-dimensional expansion_(M×n)(ii) a At the same time, n groups of samples of M standard uniformly distributed variables are simulated to generate a random sampling matrix S_(M×n)(ii) a By comparing the destruction probability matrix B_(M×n)And a random sampling matrix S_(M×n)Determining the damage condition of the door and the window according to the size of the parity elements; wherein the impact failure probability value is greater thanThe sample value, representing a corruption.

(5) According to the average wind speed and the wind direction of typhoon at the kth time step and the house wind pressure coefficient information, the external wind pressure of the enclosure member is firstly calculated, and then the external wind pressure extreme value distribution of the N members is estimated by using the method in the step 2.

(6) Carrying out N-element extreme value simulation work based on Nataf transformation, wherein the specific process comprises the following steps: acquiring a correlation coefficient matrix sigma based on wind pressure data of each member_WCalculating according to the formula (8) to obtain a matrix sigma_ZAnd performing Cholesky decomposition on the matrix to obtain a lower triangular matrixGenerating N groups of samples of N-element independent standard Gaussian vectors by Monte Carlo simulation, i.e. generating matrix U_(N×n)(ii) a (III) by the relationObtaining a sample matrix Z of N-element related standard Gaussian vectors_(N×n)(ii) a Matrix Z_(N×n)After equal probability transformation (formula (5)), further obtaining a sample matrix W_(N×n)N groups of samples of the external wind pressure extremum of the N components.

(7) The net wind pressure extreme value of the enclosure member is the sum of the external wind pressure extreme value and the internal pressure of the house, namely, the external wind pressure extreme value matrix W is used_(N×n)Each line of (2) and the current state internal pressure vectorAdding to obtain a sample matrix Q of a net wind pressure extreme value_(N×n)(ii) a By comparing the net wind pressure extreme value matrix Q_(N×n)And resistance matrix R_(N×n)Determining the size of the parity elements to determine the damage condition of the component; wherein the net wind pressure value is greater than the resistance value, indicating a failure.

(8) Determining the component damage condition according to the analysis results of the impact damage and the wind pressure damage of the flying objects in the steps (4) and (7), and updating the damage matrix D according to the component damage condition_(N×n)。

(9) If the damage matrix changes, then based on the current damage status,updating the internal pressure vector by equation (9)And repeating the steps (7) to (8) until the destruction matrix is not changed any more.

(10) And (5) enabling k to be k +1, entering the next time step, and repeating the steps (4) to (9) until the analysis of all the time steps is finished.

Estimating the damage probability of a component by analyzing a certain row in a damage matrix, namely a plurality of groups of simulation results corresponding to the damage condition of the component; the loss rate of each type of component in a group of simulation results is estimated by analyzing a certain column in the damage matrix, namely a group of simulation results corresponding to the damage conditions of all components. .

The invention has the beneficial effects that:

based on wind tunnel test data, the method considers the aspects of typhoon holding effect, impact damage of flying and throwing objects, wind pressure damage and internal pressure change, thereby more practically evaluating the wind-induced disasters of the enclosure structure of the short house.

Drawings

FIG. 1 is a modified wind damage analysis process of a low building enclosure structure considering typhoon holding time effect

FIG. 2 is a prototype house

FIG. 3 shows the average wind speed for 10 minutes

FIG. 4 shows the average wind direction for 10 minutes

FIG. 5 is a graph of cumulative window failure probability trend over time

FIG. 6 is a cumulative failure probability of roof tiles at a time

FIG. 7 is the cumulative failure probability of the roof panel at a certain time

FIG. 8 is a graph showing the average loss rate of the envelope member

FIG. 9 is a standard deviation of a loss rate of the envelope

Detailed Description

The house prototype is shown in fig. 2 and has dimensions of 19.1m × 12.2m × 3.7m (length × width × height) and a roof slope of 1: 12. The roof consists of 80 roof panels, each of which is 1.22m by 2.44m in size. The upper layer of roof board is 0.33m × 1m asphalt tile. The numbers are W1-24 windows with the size of 1.83m multiplied by 1.07m are arranged on the peripheral wall surface of the W4. The mean obedience value of the resistance of the roof panel is 4.2kN/m²Normal distribution with a coefficient of variation of 0.28; the mean obeying value of the resistance of the asphalt shingle is 3.5kN/m²Normal distribution with a coefficient of variation of 0.3; the window resistance follows a normal distribution with an average value of 5kN/m2 and a coefficient of variation of 0.2. The university of western-safety University (UWO) developed an atmospheric boundary layer wind tunnel test on a 1:100 scaled model of the house. 335 wind pressure measuring points are arranged on the surface of the roof, and 116 measuring points are arranged on the wall surface, as shown by red "·" in figure 2. The test measures the house wind pressure coefficient under a plurality of wind directions by taking 5 degrees as increment. The sampling time is 100s, and the sampling frequency is 500 Hz. In the present example, the test result corresponding to the open terrain is selected, and the conversion of the reference wind speed to the eave height is about 6.1 m/s.

Selecting and analyzing the blackroom typhoon data logged in Guangdong province in 24 days 9 and 2008, intercepting the current day 2: 00 to 10: the average wind speed and average wind direction data of the meteorological observation tower (10m height) of 00-side young island are recorded for 10 minutes, as shown in fig. 3 and 4. As can be seen from the figure, the maximum wind speeds before and after the observation of the vicinity of the tower by the eye path are 44.2m/s (4:30am) and 38.1m/s (7:00am), respectively, during which the wind direction changes by nearly 180. The house is assumed to be located near the island meteorological observation tower, and the ridge line is parallel to 90 degrees (270 degrees) of the observed wind direction. Meanwhile, the types of the fields around the house are all regarded as open terrains. The boundary layer at the position is assumed to be an exponential wind cutting line, the index is 0.2, and the ratio of the wind speed at the eave height to the wind speed at the 10m height is about 0.82.

In the simulation flow shown in fig. 1, the extreme values of the external wind pressure of the window and the roof tile are simulated, and the extreme values of the external wind pressure of the roof panel should be weighted and averaged by the contribution area of the extreme values of the external wind pressure of the attached roof tile. The window and roof panels are affected by internal pressure, and the net wind pressure extreme value of the window and roof panels comprises the internal pressure; the roof tiles are only acted by external wind pressure, and the net wind pressure extreme value of the roof tiles is equal to the external wind pressure extreme value. The number of the throws is set to be 100, and the sample volume of the simulation is selected to be 20 ten thousand to ensure the convergence of the result.

1) Probability of damage to the containment structure

According to the destruction matrix recorded at each moment, the cumulative destruction probability of each component at each moment can be obtained. Fig. 5 shows the cumulative damage probability of windows on the surrounding wall surface as a function of time, and the cumulative damage probabilities of windows on different wall surfaces all increase with time, wherein the damage probabilities of all windows show an increasing trend at two moments 4:30 and 7:00 when the wind speed reaches the peak value. Fig. 6(a) (b) and 7(a) (b) show the cumulative failure probability of roof tiles and panels at 4:30 and 7:00, where the failure at 4:30 occurs mainly on the windward side of the roof, and at 7:00 the wind direction turns through about 180 °, and the failure on the other side of the roof is found to be greater. The resulting probability of failure of the roof tiles and panels in this typhoon event is shown in fig. 6(c) and 7 (c).

2) Loss rate of building envelope

And acquiring multiple groups of samples of the loss rate of each component in the typhoon event according to the damage matrix recorded at each moment, and further calculating the change condition of the loss rate statistic value of each component along with time. Generally, one is mainly concerned with the average level of loss and the level of variation, i.e., loss rate mean and standard deviation. The changes of the loss rate mean value and the standard deviation of each component along with time are respectively shown in the figure 8 and the figure 9. It can be seen that the average loss rate increases over time and increases significantly as the wind speeds peak at 4:30 and 7: 00. The standard deviation of the loss rate generally tends to increase, with a slight decrease occurring over the individual time periods. The final loss rate mean values (standard deviation) of the three types of components of the window, the roof tile and the roof panel are 73.31% (8.76%), 33.15% (4.43%) and 10.43% (3.37%), respectively. When only the influence of the maximum wind speed in a typhoon period is considered, namely the damage of the typhoon wind speed and the wind direction at the moment of 4:30 to a house is analyzed independently, the average loss rates of the windows, the roof tiles and the roof plates are respectively 24.57%, 12.94% and 5.4%. Compared with disaster loss analysis considering typhoon duration effect, the method only considers the maximum wind speed to underestimate the loss degree of various enclosure components.