基于联系云-证据理论的岩爆烈度预测模型

A Connection Cloud-Evidence Theory Coupling Model for Prediction of Rockburst Intensity

岩爆机理复杂影响因素众多且呈现多种不确定性,应用云模型预测岩爆问题虽能刻画指标的随机性和模糊性,但很难模拟非正态分布的评价指标及存在冲突数据融合失真问题.为克服这些缺陷,探讨了岩爆烈度的联系云-证据预测模型.该模型首先基于联系数定量表达评价指标,通过联系云构建评价矩阵,并应用D-S证据理论得到基本概率赋值,进而基于距离函数的组合权重与融合均值证据预测样本的岩爆等级.实例应用及与其他方法对比结果表明,该模型应用于岩爆预测是有效可行的,且克服了传统云模型和证据理论的不足,为岩爆烈度分级预测提供了一种新的途径.

Rockburst mechanism is a complex problem involving various uncertain factors. Although the cloud model can deal with the randomness and fuzziness of indicators for prediction of rockburst intensity,it cannot simulate the state of evaluation indicators in a finite distribution interval and address the distortion of data fusion. Herein,a connection cloud-evidence theory coupling model was built to remedy these defects. In this model,evaluation indicators were firstly expressed quantitatively by connection numbers. Then the evaluation matrix was constructed with the connection cloud model and the basic probability assignment based on the evidence theory was obtained. Finally,with the combination weight obtained from a distance function,classification of the rockburst intensity was determined according to the mean evidence value. The case study and comparison with other methods show that,the proposed model is effective and feasible for the prediction of rockburst intensity. It can overcome the shortcomings of the normal cloud model and the evidence theory,making a novel method for comprehensive prediction of rockburst intensity.