基于组合赋权法和云理论的震后重建住宅工程质量评价研究

针对地震灾区重建住宅工程的质量等级评价具有模糊性和随机性的特点，提出了关于云模型的震后重建住宅工程质量评价方法。着重考虑住宅安全性的需求，建立了震后重建住宅工程质量评价指标体系，首先采用层次分析法(AHP）与信息熵法分别确定主、客观权重，然后运用博弈论思想将主、客观权重优化组合，得到综合权重；最后以汶川地震后芦山某重建住宅工程为例，综合利用正、逆云发生器及虚拟云理论得到各级指标的评价云滴图，与质量等级标准云作相似性比较。结果表明:该重建社区的住宅质量等级总体为“良”，一级指标中抗震安全等级为“优”，重建效率评价为“中”。算例的结果表明该方法能够直观、生动地反映各级指标间的相互关系，有利于评估者对各指标因素的全局把握。

Evaluation on engineering quality of post-earthquake reconstruction residential buildings based on combined weight method and cloud theory

Aiming at the characteristics of fuzziness and randomness in the quality grade evaluation of post-earthquake reconstruction residential buildings in the earthquake-stricken area, an evaluation method on engineering quality of post-earthquake reconstruction residential buildings with regard to cloud model was put forward. Focusing on the requirement of residential building safety, an evaluation index system on engineering quality of post-earthquake reconstruction residential buildings was established. Firstly, the AHP and information entropy method were used to determine the subjective weights and objective weights respectively. Secondly, the optimization combination of subjective weights and objective weights was conducted based on the idea of game theory, then the comprehensive weight was obtained. Finally, taking an engineering of reconstruction residential building in Lushan after the Wenchuan earthquake as example, the evaluation cloud droplet graphs of indexes in each level were obtained by comprehensive utilization of the forward and backward cloud generator and virtual cloud theory, and the similarity comparison was conducted with the standard cloud of quality grade. The results showed that the overall quality grade of residential buildings in the reconstructed community is "good", and in the first level indexes, the anti-seismic safety grade is "excellent", and the reconstruction efficiency is "medium". The method can intuitively and vividly reflect the relationship between the indexes in each level, which is beneficial for the evaluator to grasp the overall situation of each index factor.