A rockburst grade evaluation method based on principal component analysis and the catastrophe progression method

Abstract

Read online

Rockburst disasters always have a great influence on engineering practice. In order to accurately predict the occurrence of rockburst hazards, this paper proposes a rockburst rating evaluation method based on principal component analysis (PCA) and the catastrophe progression method, taking into account several influencing factors. In this paper, 15 indicators, such as strength brittleness factor (R), stress factor (P), and rock quality index (RQD) (reflecting the strength and fragmentation degree of rock mass), were selected from seven samples and were analyzed and downscaled by principal component analysis. Combined with the catastrophe progression method, each layer index was dimensionless and normalized to determine the mutation level value of each layer. Based on the principle of complementarity or non-complementarity, to determine the total mutation level value, the layer index was used to divide the rockblast-level interval and predict the rockblast level. The results show that the method proposed in this paper can be used to not only distinguish the importance of each of the same level of indicators but also avoid the impact of superimposed factor correlations between the same level of indicators, making the results more objective. This paper presents accurate rock explosion assessment results and an actual engineering situation. The number of factors affecting the assessment of the rock explosion level provides new insights.

Keywords

• principal component analysis
• catastrophe progression method
• rockburst classification
• prediction
• multi-factor
• predictive models