Modeling on the drum loading performance of the helix shearer based on Partial Least Square Regression

In order to get the accurate and quantificational loading performance model of the Shearer drum, the relevant abbreviated experiment is designed to simulate the drum loading coal process in the different working conditions according to the analogy theory. Select drum´s rotated speed, haulage speed and the angle of drum´s hub to be variables of the regression model. Experiment data are gotten by the orthogonal method. The regression model is identified by PLSR (Partial Least Square Regression) method to get model´s coefficients. The results of the simulation and experiment prove that this kind of method not only can build the model of loading performance correctly and effectively but also have the advantages of fewer experiments, less amount of calculation, stronger ability of the model forecast. The coefficients of the regression model have more actual meaning.