深基坑支护方案选型的区间证据推理方法

An Interval Evidential Reasoning-based Approach for Selecting Supporting Scheme for Deep Foundation Pit

在城市建筑密集区域进行地铁车站深基坑施工会对周边土体产生影响,进而危及临近建(构)筑物的安全,因此,合理选择深基坑支护方案显得尤为重要。以武汉地铁16号线某车站深基坑工程为背景,对深基坑支护方案的选型方法展开了研究。为了克服既有选型方法存在主观性强以及对不确定性因素考虑不足等缺陷,提出了一种深基坑支护方案选型的区间证据推理方法。该方法将混合型指标评价值统一转化为区间数形式,运用熵值法确定指标权重区间,在此基础上构建了指标评价值-重要度-置信结构的信度结构转换途径。将转换后的指标评价值视为方案决策的证据,运用区间证据推理算法融合得到各备选方案的综合等级效用值,通过比较综合等级效用值的大小实现方案排序。实例计算及与其他决策方法对比分析表明:该方法能够处理指标评价值为混合型的多属性决策问题,且决策过程受主观因素影响小;该方法适用于评价信息为非精确数的评价问题,且能将评价信息所具有的不确定性程度在评价结果中予以更直观的体现,即评价信息的不确定性越大,将导致证据合成结果的不确定性亦越大,从而使得评价结果的可信度更高。

As the construction of deep foundation pit for metro station in urban densely building area has adverse influence on the surrounding soil,and then endangers the adjacent buildings or structures,the reasonable selection of supporting scheme for deep foundation pit is particularly important.On the background of the deep foundation pit project for a station in metro line 16 in Wuhan,the selection method of deep foundation pit support scheme is studied.An interval evidential reasoning method aiming at overcoming some defects of the existing approaches,namely too subjective and taking insufficient account of the uncertain factors,is proposed.By using this method,the mixed indicator evaluation value is uniformly converted into the form of interval number,and the indicator weight interval is determined by using the entropy value method.On this basis,the reliability structure conversion path of indicator evaluation value-importanceconfidence structure is established.Regarding the converted indicator evaluation value as the evidence for scheme decisions,the comprehensive grade utility value of each alternative is obtained by using the interval evidence reasoning algorithm.Through comparing the utility values of comprehensive grade,the ranking of alternatives could be arrived.Comparative analysis of the case calculation and other decision-making methods shows that(1)the proposed method is able to deal with the multi-attribute decision making problem with mixed indicator evaluation value,and the decision-making process of the proposed method is less influenced by subjective factors;(2)this method is suitable for evaluation problems where the evaluation information is a non-precise number,and the degree of uncertainty of the evaluation information can be more intuitively reflected in the evaluation result,that is,the greater the uncertainty of the information used for evaluation,the greater the uncertainty of the evidential synthesis result,which allows a higher credibility of the evaluation result.