Short-term rockburst risk prediction using ensemble learning methods

  • PDF / 2,254,449 Bytes
  • 24 Pages / 439.37 x 666.142 pts Page_size
  • 30 Downloads / 224 Views

DOWNLOAD

REPORT


Short‑term rockburst risk prediction using ensemble learning methods Weizhang Liang1,2 · Asli Sari2 · Guoyan Zhao1   · Stephen D. McKinnon2 · Hao Wu3 Received: 26 March 2020 / Accepted: 21 August 2020 © Springer Nature B.V. 2020

Abstract Short-term rockburst risk prediction plays a crucial role in ensuring the safety of workers. However, it is a challenging task in deep rock engineering as it depends on many factors. More recently, machine learning approaches have started to be used to predict rockbursts. In this paper, ensemble learning methods including random forest (RF), adaptive boosting, gradient boosted decision tree (GBDT), extreme gradient boosting and light gradient boosting machine were adopted to predict short-term rockburst risk using microseismic data from the tunnels of Jinping-II hydropower project in China. First, labeled rockburst data with six indicators based on microseismic monitoring were collected. Then, the original rockburst data were randomly divided into training and test sets with a 70/30 sampling strategy. The hyperparameters of the ensemble learning methods were tuned with fivefold cross-validation during training. Finally, the predictive performance of each model was evaluated using classification accuracy, Cohen’s Kappa, precision, recall and F-measure metrics on the test set. The results showed that RF and GBDT possessed better overall performance. RF obtained the highest average accuracy of 0.8000 for all cases, whereas GBDT achieved the highest value for high (moderate and intense) risk cases with an accuracy of 0.9167. The proposed methodology can provide effective guidance for short-term rockburst risk management in deep underground projects. Keywords  Rockburst · Short-term risk · Ensemble learning · Prediction · Microseismic monitoring

Electronic supplementary material  The online version of this article (https​://doi.org/10.1007/s1106​ 9-020-04255​-7) contains supplementary material, which is available to authorized users. * Guoyan Zhao [email protected] * Hao Wu [email protected] 1

School of Resources and Safety Engineering, Central South University, Changsha 410083, China

2

The Robert M. Buchan Department of Mining, Queen’s University, Kingston K7L 3N6, Canada

3

School of Mines, China University of Mining and Technology, Xuzhou 221116, China



13

Vol.:(0123456789)



Natural Hazards

1 Introduction Rockbursting is a rock mass instability phenomenon accompanied by spalling, cracking, splitting or even ejection due to the sudden release of strain energy (Gong et al. 2019). For deep mines and tunnels, rockbursting is becoming an increasingly prominent problem. It has been reported that many countries have encountered the rockburst hazard, making it a worldwide challenge (Keneti and Sainsbury 2018; Hu et  al. 2019). For example, seismicity and rockbursting were deemed as the biggest threat to workers’ safety in Ontario’s underground mines based on an investigation by the Ontario Ministry of Labour in Canada (Ontario Ministry of Labour 2015). An intense rockburst can