基于联系云-证据理论的岩爆烈度预测模型
摘要
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.