A Prediction Method of Failure Depth of Coal Seam Floor Based on FA-GWO-SVM Model

Zhongsi Dou, Ruili Han, Yimeng Wang

Research output: Contribution to journalArticlepeer-review

1 Citation (Scopus)
30 Downloads (Pure)

Abstract

Accurately predicting the failure depth of the coal seam floor is an important premise to prevent water inrush from the coal seam floor and ensure safe and efficient mining operations. Based on the coal seam floor damage degree data collected from various mining areas in China, this study selected six indexes (coal seam mining thickness, coal seam dip angle, mining depth, working face slope length, floor damage resistance and presence of a cutting fault) to predict the value of the coal seam floor damage depth. Based on Support Vector Machine(SVM) model,the factor analysis method was applied to reduce the dimension and extract the original data variables. The extracted variables were used as the input for the SVM model. Afterwards, the Grey Wolf Optimizer (GWO) algorithm was adopted to optimize the parameters C and g, and the Factor Analysis(FA)-GWO-SVM coal seam floor failure depth prediction model was established. The reliability of the model was verified before proceeding with the investigation. Results indicate that the prediction model of the coal seam floor failure depth based on the FA-GWO-SVM method has a good generalization ability and a strong prediction performance for the new sample data. Compared with the traditional SVM model and the GWO-SVM model, it has the minimum MAPE, RMSE and MAE values. Furthermore, the learning ability, stability and prediction accuracy of the model are significantly improved. The model does not only overcome the drawbacks of the traditional prediction methods that do not consider the interaction of various factors but also simplifies the input scale of the SVM model. The issue of affecting the prediction accuracy due to the difficulty of the parameter optimization in the SVM model is solved using the GWO optimization technique, and the model’s prediction accuracy and operational performance are enhanced. This study provides an effective method for accurately predicting the failure depth of the coal seam floor.

Original languageEnglish
Pages (from-to)161-169
Number of pages9
JournalJournal of Engineering Science and Technology Review
Volume16
Issue number1
DOIs
Publication statusPublished - 2023

Keywords

  • Bottom plate failure depth
  • Factor analysis
  • Floor water inrush
  • Gray wolf optimization algorithm
  • Support vector machine

ASJC Scopus subject areas

  • General Engineering

Fingerprint

Dive into the research topics of 'A Prediction Method of Failure Depth of Coal Seam Floor Based on FA-GWO-SVM Model'. Together they form a unique fingerprint.

Cite this