| Wind disasters are one of the main disasters facting mankind,and typhoons(i.e.,hurricanes,or strong tropical cyclones)are the most disasters.In recent years,low-rise light steel structures in coastal cities of China(Zhejiang,Hong Kong,Fuzhou,Xiamen,etc.)are extremely vulnerable to severe damage under the action of typhoons.Moreover,the heavy rain,floodings,and storm surge resulted from typhoons are responsible for more extensive and serious secondary damage,resulting in huge economic losses.To improve the resilience of houses and provide scientific basis for improving disaster prevention capacity and risk control in government,industry and insurance industries,it is necessary to conduct the vulnerability analysis of light steel structures under typhoons.The thesis focuses on four aspects: the typhoon wind field simulation,typhoon track simulation,extreme value estimation for non-Gaussian wind pressures and loss estimation for low-rise buildings.The main work is as follows:Firstly,empirical typhoon wind field models widely used in the engineering field and insurance industry have been introduced briefly.According to the general idea of the empirical wind field model by researchers at home and abroad,an improved typhoon empirical wind field model was proposed.In the model,the gradient wind field is first obtained based on the typhoon gradient equilibrium equation;then,inspired by the successful application of the Arya boundary layer model in planetary boundary layer model(PBL models),the theoretical boundary layer model was introduced when simulating the near-surface wind field.The boundary layer model is solved by the numerical method,and the results are simply analyticized for rapid simulation of typhoon wind field.Based on the observation data of the historical typhoon landing on the coastline of China,the accuracy of the refined empirical model was verified by comparing its accuracy with those from the PBL model and previous empirical model.Secondly,the general idea of typhoon risk analysis based on the typhoon track model by the Circular Sub-region Method(CSM)is introduced,the problems existing in the traditional typhoon local track model and the corresponding improved model have been proposed.The improved model mainly involves the simulation of the central pressure difference,the determination of the simulated circle size,and the consideration of the correlation among the typhoon key parameters.For the typhoon risk analysis,the data released by the China Meteorological Administration and the United States Joint Typhoon Warning Center(JTWC)were used to determine the typhoon filling model and maximum wind speed radius model,respectively.Finally,based on the models including improved wind field,improved local track,typhoon filling and maximum wind speed radius,the typhoon hazard analysis was conducted for coastal cities in southeast China,and the simulated extreme wind speeds in different directions and different return periods for different cities were given.Based on the simulation results,the influence of the correlation among the typhoon key parameters on the simulated extreme wind speed was discussed.Also,the simulation results are compared with the results from the codes,the traditional local track model and the full track model to validate the improved local track model.Wind pressures exhibit non-Gaussianity.The translation functions by Hermite polynomial model(HPM)which is widely used for the simulation and peak value estimation of the non-Gaussian wind pressure are required to be monotonic.For non-Gaussian wind pressures whose skewness and kurtosis fall out of the effective region of HPM,HPM is no longer applicable.The Johnson conversion model(JTM)has a larger effective region than HPM.In this paper,the simulation and peak value estimation of non-Gaussian wind pressures based on the Johnson transformation model is systematically studied.Firstly,the Johnson transformation model and its parameter estimation method are introduced.A framework of JTM-based non-Gaussian wind pressure simulation was established.In this framework,a set of closed-form formulas to determine the relationship between correlation coefficients of the non-Gaussian process and those of the Gaussian process were derived.At the same time,expressions for calculating the non-Gaussian peak factor based on JTM were derived.Finally,based on theoretical analysis and wind tunnel test data,the differences between JTM and HPM in non-Gaussian simulation and extreme value estimation have been systematically compared to evaluate the accuracy of JTM in non-Gaussian wind pressure simulation and extreme estimation.Based on the results of typhoon risk analysis in the coastal cities of southeast China and JTM,an effective wind-induced vulnerability analysis method for the envelope structure of light steel houses was proposed to consider many factors such as the opening conditions,internal pressure changes and the wind speed and direction variability.First,the wind-induced damage process of light steel building has been analyzed.Secondly,the opening condition is determined due to window/door failures due to windborne debris impact and wind pressures.Bernoulli equations are adopted to obtain internal pressures.Then the Nataf transformation method is used to model the correlation among wind loads so that associated failures of roof panels are considered.Finally,based on stochastic simulation,the wind-induced vulnerability analysis process is proposed;and the directional effect of the typhoon is further considered.Combined with specific numerical examples,the proposed method is illustrated,and the loss ratios for different cities in different return periods are given. |
