摘要
地震影响系数是构建地震反应谱的最重要参数之一,也是抗震房屋设计和减灾的基础,唐山市地震影响系数的分布受到场地条件、基岩条件和特殊地质现象等诸多因素的共同影响。通过分析唐山市地震影响系数分布的主要影响因素,得到影响唐山市地震影响系数分布的主要指标。考虑各指标与地震影响系数之间的非线性关联关系,应用人工神经网络方法构建了唐山市地震影响系数分布预测模型。根据实际资料,确定了模型的参数和多个系数矩阵,计算和分析了综合考虑场地和基岩等诸多条件的唐山市地震影响系数分布,并对唐山市的抗震设防规划提出了几点建议。
Earthquake influence coefficient is one of the most important factors of earthquake micro-zoning, which is the basis of earthquake-proof building design and disaster reduction. Distribution of earthquake influence coefficient is controlled by site condition, basement rock condition, and special geological condition, etc.. More attentions have been paid on site condition, but less on basement rock condition or special geological condition. The big earthquake in Tangshan city resulted in 243,000 lives lost in 1976, therefore, high attentions are paid to earthquake disaster resistance, and accurately calculation of earthquake influence coefficient is necessary for city planning and earthquake-proof building design. According to the condition of Tangshan city, distribution of basement rock fracture is calculated based on Griffith criteria, and controlling factors of the distribution of earthquake influence coefficient are discussed. Relativities among earthquake influence coefficient, site condition, basement rock condition and special geological condition are analyzed. It was found from results that the distribution of earthquake influence coefficient in Tangshan city are mainly controlled by displacement, fracture probability, site category and physiognomy. Because the relativities are complicated, predictive model is constructed based on artificial neural network, and distribution of earthquake influence coefficient in Tangshan city is calculated precisely. After repeated calculating of 197,012 times, precise results are obtained. Earthquake influence coefficient is larger than 0.48 in southeast of Tangshan city and even larger than 0.64 in some local regions. Also, it was found that earthquake influence coefficient diminishes gradually from southeast to northwest. Finally, the influence of earthquake influence coefficient distribution on city planning and engineering construction are analyzed and some advice are proposed.
出处
《防灾减灾工程学报》
CSCD
2005年第2期199-204,共6页
Journal of Disaster Prevention and Mitigation Engineering
基金
国家计委攻关计划项目(98-01-02)
河北省科学技术研究资助项目(03276901)
关键词
地震影响系数
反应谱
神经网络
唐山
地震小区划
Earthquake influence coefficient
response spectra
neural network
Tangshan
earthquake micro-zoning
