Study On Vertical Earthquake Action Stipulated In Seismic Design Specification

In the seismic design calculations of engineering structure, complexstructures often need to take two or even three directions of combined effects ofearthquake action into consideration. Multi-directional inputs involves in how tofurther regulate the vertical seismic action on the basis of horizontal groundmotion’s provision. This paper points out that Guidelines for Seismic Design ofHighway Bridges stipulating the vertical earthquake action by using vertical tohorizontal spectra ratio function can be unreasonable and inconvenient flaws incertain conditions, then study the feasibility of stipulating the horizontal andvertical design response spectrum with two sets of parameters respectively.1513sets of NGA database’s three-component seismic records from64earthquakes released by PEER are selected in this paper to perform the statisticalanalysis of the response spectrum characteristics. To analyze t he responsespectrum characteristics of different basic accelerations, the records are groupedby magnitude, distance and site conditions, each group’s average responsespectrum is standardized to get the corresponding effective peak accelerationEPA. The EPA has6intervals of0.05,0.1,0.05,0.1,0.3and0.4g. According tothe results on the Class II site, considering the general relationship of EPA withmagnitude and epicentral distance, groups of the epicentral distance R∈(0,10], magnitude M∈[4.5,6.5] and R∈(10,30], M∈(6.5,7.5] are mergedinto interval0.2g, groups of R∈(10,30], M∈(5.5,6.5] and R∈(30,50],M∈(6.5,7.5] are into interval0.15g, groups of R∈(30,50], M∈(5.5,6.5]and R∈(50,80], M∈(6.5,7.5] are into interval0.1g; R∈(10,200], M∈[4.5,5.5], R∈(50,200], M∈(5.5,6.5], and R∈(80,200], M∈(6.5,7.5]are into interval0.05g. As for the horizontal ground motion data merged intoeach interval, they are divided into different site conditions to get the averageresponse spectrum, through unified standardization, the amplitudes andcharacteristic periods of the design response spectra are extracted. According toresponse spectrum amplitudes’ calculations, horizontal site coefficients of eachfield category are obtained.Grouping and merging are crucial in this study. After explaining the purposeof grouping, how to group and how to merge groups, the effects of merging scheme are verified with the aid of PGA C-B attenuation relationship publishedby NGA project. At the same time, the author figure out the Matlab codes whichcan perform the function of response spectra characteristic parameter extractionand fitting design response spectrum.Then, by analyzing the vertical ground motion records grouped by themagnitude and distance, based on the merging scheme, they are merged into eachinterval and divided into different site conditions to get the average responsespectrum, through unified standardization, the amplitudes and characteristicperiods of the design response spectra are extracted. According to responsespectrum amplitudes’ calculations, vertical site coefficients of each fieldcategory are obtained.In order to verify the reliability of the proposed site coefficient andcharacteristic period, the horizontal and vertical design standard response spectradetermined by design basic acceleration and site category are compared with theresponding average spectra. Results show that stipulating the parameters of thehorizontal and vertical response spectra, respectively, to determine the twodirection acceleration response spectrum, compared with stipulating thehorizontal acceleration response spectrum parameters, multiplying the horizontaldesign response spectrum by spectra ratio function to determine the verticalresponse spectrum, is more convenient, and more efficient to reflect thestatistical average characteristics of strong ground motion data.