围岩稳定性评价的投影寻踪权重-属性区间识别模型

围岩的稳定性评价是一个复杂的不确定系统问题。采用粒子群算法优化投影寻踪方法，并结合属性区间识别理论，建立了围岩稳定性评价的投影寻踪权重-属性区间识别模型。选取5个围岩稳定性影响因子作为评价指标，通过构造属性测度区间函数计算单指标属性测度区间，利用基于粒子群优化的投影寻踪确定各评价指标的权重以计算综合属性测度区间，应用置信度准则和评分准则对围岩的稳定性进行属性识别。实例研究表明，该模型能有效的解决围岩稳定性评价问题，且评价结果科学可靠；评价模型采用投影寻踪确定权重，避免了权重确定中的主观性和随意性，保证了评价工作的客观性和准确性。

Attribute interval recognition model based on projection pursuit weight for evaluation of stability of surrounding rock

Evaluation of the stability of surrounding rock is a complex system problem of uncertainty. A new model for evaluating stability of surrounding rock is established based on the projection pursuit(PP) optimized by the particle swarm optimization(PSO) and the attribute interval recognition model(AIRM), which is named as attribute interval recognition model based on projection pursuit weight(AIRM-PPW). Based on the analyzing of main influencing factors of surrounding rock stability, five indices are chosen as the evaluation indices. The attribute measurement interval of single index is computed through constructing the attribute measurement interval functions and the synthetic attribute measurement interval is calculated by the weight of PP. The confidence criterion and the score criterion are used to recognize the stability of surrounding rock. The result from case study indicates that the model is effective to evaluate the stability of surrounding rock with reliable conclusion. Moreover, using weight of PP, the model could avoid subjectivity and arbitrary in the process of determination of index weight, and ensure objectivity and accuracy of the evaluation.