The performance prediction of roadheader using Support Vector Machine

Instantaneous cutting rates (ICRs) roadheaders' is considered one of the important indicators in advance operations in tunnels and underground mines. This parameter depends on the two important factors of device specifications and geomechanical properties of the rock mass, and based on them, various researchers have presented various models for determining instantaneous cutting rates (ICRs) theoretically and experimentally. Instantaneous cutting rates (ICRs) can have various consequences such as increasing the cost of equipment and servicing and prolonging the excavating life. Considering the importance of time and cost management in tunneling and mining projects, the parameters affecting this factor should be coordinated in such a way that the instantaneous cutting rates (ICRs) in roadheaders are close to their optimal value. Various parameters are effective on the instantaneous cutting rates (ICRs) of roadheaders' cars, and it is not correct to study the effect of each of them on this factor as a unit, but the effect of them together should be considered. The database was then analyzed through SVM to yield an optimum predictive model for ICR (m3/h). Results showed that there is a close relation between actual (measured) data and predicted data with coefficient of determination (R2) of 0.89 and root mean square error (RMSE) =0.21.

The performance prediction of roadheader using Support Vector Machine Keywords:

Instantaneous cutting rates (ICRs) , Roadheaders , excavating , coefficient of determination