【導(dǎo)讀】安德烈,和克里斯多夫,構(gòu)類型被認(rèn)為包括鋸成的木梁、膠合梁及各種類型的木梁板。視為隨機變量，因而，結(jié)構(gòu)特性是根據(jù)可靠性指標(biāo)來測定的。橋的恒載和交通活。載，都是基于先前的研究結(jié)果。材料的阻值是取自可用的測試得來的數(shù)據(jù)，這些。數(shù)據(jù)中包含了考慮有彈性反應(yīng)作用的數(shù)據(jù)。阻力的組成和結(jié)構(gòu)系統(tǒng)是基于可利用。的實驗數(shù)據(jù)和有限元分析的結(jié)果。阻力的統(tǒng)計參數(shù)是由梁板、梁體及個別的組件。對木橋進行可靠性分析設(shè)計應(yīng)依照AASHTO標(biāo)準(zhǔn)設(shè)計規(guī)范并且要注。因此，對于木結(jié)構(gòu)的可靠性水平的一致性問題還。注意：討論時間截至到2021年4月1日。擴大一個月的截止日期，必須向美國土木工程師協(xié)會總編輯提出書面申請。補了這一空缺并且提出了一些建議，從而使木橋在長期的可靠性上達(dá)成一致。性，要考慮到這些卡車的重量作用在橋上會產(chǎn)生不同程度的相互作用。

　　

【正文】 be consistent with those used to calibrate the steel and concrete sections of the LRFD code (Nowak 1999, 1993). The considered statistical parameters include the ratio of mean to nominal (design) value, called the bias factor, λ , and coef?cient of variation, V, that is the ratio of standard deviation to the mean. For wood and concrete (deck) ponents, bias factor λ = and coef?cient of variation V=。 for steel (girders),λ = and V=。 and for asphalt, mean thickness is taken as 90 mm and V=. Dead load is taken as normally distributed. The live load model is based on the available truck survey data as used in the calibration of the AASHTO Code (Nowak 1999,1993). The analysis of live load involves the determination of the load in each lane and load distribution to ponents. The probabilities of a simultaneous occurrence of more than one truck in adjacent lanes and a multiple truck occurrence in the same lane,were considered with various degrees of correlation between truck weights. For most wood bridges, however, only a single truck per lane needs to be considered, as the typical short spans result in the probability of two trucks in the same lane unlikely or even impossible. The simulations indicated that for bridges with girder spacing of – m (4–8 ft), two fully correlated trucks sidebyside govern. For the maximum 75 year moment,the results of analysis indicated that each truck in this bination is equivalent to the maximum two month truck. That is,considering the various binations of single and two sidebyside truck weights and probabilities of occurrence for each bination, two sidebyside trucks of equal weight which are each the weight of the maximum single truck expected to pass in a twomonth period, govern the load model in the reliability analysis. Bias factors were calculated as the ratio of mean maximum moment and design moment (applied to the entire bridge) speci?ed in the Code, for various time periods. It was found that bias factor varies with span length. For spans up to 30 m (100 ft), the results are shown in Fig. 4 for 1 and 75 year periods. The coef?cient of variation is shown in Fig. 5. Live load is approximately lognormal. 第 19頁 Fig. 4. Bias factor for live load 第 20頁 Fig. 5. Coef?cient of variation for live load As wood strength is affected by load duration, the live load duration is calculated for various time periods. Three values of the average daily truck traf?c (ADTT) are considered: low with ADTT=500, medium with ADTT=1,000, and high with ADTT=3,000. It is assumed that the percentage of the actual heavy trucks (only very heavy vehicles need to be considered) is 20%,and this corresponds to 100, 200, and 600 trucks per day for the three considered traf?c volumes, respectively. Note that these are high ADTT values for typical wood bridges, which are usually located on lowvolume roads and may experience only a fraction of the traf?c volume that highway bridges do. However, as current design procedures stipulate no restriction as to the use of wood bridges with regard to traf?c volume, for code calibration purposes it would be unconservative to base load duration on low traf?c volume roads only. Considering various span lengths and posted speed limits, it is assumed that the average duration of truck passage is about 1 s. For a typical simple span wood bridge,the load effect (bending moment) gradually increases from zero to maximum at midspan, then reduces back to zero. The actual duration of maximum live load effect is lower than crossing time and, therefore, on average exposure to the maximum live load effect is assumed equal to s. In most cases, this is a conservative assumption as the effected portion of the in?uence line for many ponents of wood bridges is smaller than the 第 21頁 whole span length. Therefore, the live load duration (corresponding to very heavy trucks) for 75 year period and for the three considered traf?c volumes is 1. Low ADTT(100 trucks)( s)(365 days)(75 years)=15 days。 2. Medium ADTT(200 trucks)( s)(365 days)(75 years)=30 days。 and 3. High ADTT(600 trucks)( s)(365 days)(75 years)=90 days. Although wood bridges are typically located on low volume roads, in the reliability analysis it is conservatively assumed that the live load duration is 2 months (between medium and high traf?c volumes). For short spans, live load is caused by axle loads or even wheel loads. Therefore, the live load model is determined by variations in wheel load rather than the entire truck or parameters for wheel load are derived from existing survey data (Nowak et al. l994). Based on axle load taken from ?eld measurements on bridges located in Michigan, as well as state police citation ?les for overload vehicles, the maximum observed axle load for a 1 year interval is close to 200 kN (40 kips), which produces 50 kN (10 kips) per wheel (two tires per wheel). Therefore, in this calibration, the mean maximum one year value for a wheel load is taken as 50 kN (10 kips). The coef?cient of variation is taken as (Nowak et al. 1994). Tire contact area is an important consideration for live load distribution to short span ponents. Based on the measurements reported by Pezo et al. (1989) and Sebaaly (1992),the transverse dimension (width) of the contact area is 185 mm ( in.) for each tire, with a 125 mm (5 in.) gap between tires for a dual tire wheel. A nearly linear relationship exists between the wheel load and length of the contact area. For a 50 kN (10 kips) wheel load, tire length is approximately 250 mm (10 in.). Therefore, in this study, the contact area for a