基于改进FMNN模型的既有建筑可靠性鉴定.pdfVIP

学兔兔 江苏建筑 2014年第5期(总第 164期) 45 基于改进FMNN模型的既有建筑可靠性鉴定 张燕．施悦 (江苏省建筑科学研究院有限公司，江苏南京210008) f摘 要1 现阶段，由于自然因素和人为因素的影响。我国大量既有建筑的可靠性存在较大程度地削弱。这类建筑通常都 位于城市的重要地段、建造时间相对较早。为了尽可能地降低建筑物拆除重建所造成的资源浪费和不利环境影响，有必要采 取科学的评估鉴定方法对这类建筑的可靠性进行鉴定，及时采取加固处理措施。基于改进模糊Modular神经网络理论。对某 既有建筑的可靠性进行鉴定．通过将采集到的建筑结构可靠性性能指标输入至通过专家经验训练后的评估网络．最终根据 输出参数值得出鉴定结论．为后续建筑结构加固处理提供依据。 『关键{司1 改进的模糊Modular神经网~(FMNN)；既有建筑；可靠性鉴定 f中图分类-~]TU712．3 『文献标Lq~IA 『文章编号]1005—6270(2014)05—0045—02 The Appraisal of Reliability of Existing Building Based on Model of FMNN ZHANG Yan SHI Yue 0iangsu Research Institute of Building Science Co．，Ltd，Nanjing Jiangsu 210008 China) Abstract：At present，according to the influence of natural factors and human factors，the reliability of large number of existing buildings have been greatly weakened．In general，this kind of buildings were built relatively early in the downtown area．In order to reduce the waste of resources and pollution of environment caused by the demolition and reconstruction as far as possible，it is necessary to take scientific methods to evaluate the reliability of this kind of building and take reinforcement measures．The paper appraises the reliability of an existing building based on the theory of the improved Fuzzy Modular Neural Network(FMNN)．This method inputs the reliable indicators of the building to the network evaluation which has been trained by the experts experience and the appraiser can evaluate the reliability of the building according to the final output parameter．This paper provides the basis for the subsequent buildings which need to be appraised． Key words：the improved FMNN；the existing bui