New approach of friction identification for electro-hydraulic servo system based on evolutionary algorithm and statistical logics with experiments

High-precision control of an Electro-hydraulic servo system (EHSS) should fully consider practical non-linearities, such as dynamic friction, in controller design. The LuGre model is widely used to describe non-linear friction, but parameter identification for this model remains a challenging task. This study proposes a novel identification approach for the LuGre model based on the Evolutionary algorithm (EA) and statistical logics. Identification experiments are also conducted for a practical EHSS. Static parameters are identified with constant velocity experiments. Moreover, the non-symmetry of friction in positive and negative rotary directions is fully considered. Dynamic parameters are identified with the pre-sliding process in the breakaway experiment. The EA is utilized to enhance the optimal estimation of friction parameters. Statistical logics are used to predict the deviations in the estimated results with high confidence interval. Furthermore, extensive simulation and experiment results validate the effectiveness of the proposed approach.