Prediction of the friction torque in grease lubricated angular contact ball bearings using grey system theory

The development of special grease makes it possible for angular contact ball bearings to operate at high speed and temperature; however, as an important performance parameter, friction torque of bearings lubricated with grease is much greater than that of bearings lubricated with oil-air, and heat generation due to frictional loss is also greater, so it is necessary to predict the friction torque occurring in grease lubricated angular contact ball bearings. Based on grey system theory, a new prediction methodology for bearing friction torque is proposed which capitalizes on the notion that the information about friction torque of angular contact ball bearing is generally poor, incomplete and uncertain. A grey prediction model, GM (1, N) model, is presented to predict the friction torque in grease lubricated angular contact ball bearings. Several experiments on the friction torque of grease lubricated angular contact ball bearings were conducted to model and validate the effectiveness of the GM (1, N) model through on-line and off-line approaches. Experimental results show that about 90% of bearing friction torque under varying speed can be predicted in the on-line prediction; above 85% of bearing friction torque under varying speed and different loads can be predicted in the off-line prediction. Comprehensive analysis shows that, the GM (1, N) model performs very well for both modeling data and model validation data under different loads, varying bearing speed and work cycles, the proposed methodology can be used to predict bearing friction torque with good accuracy and robustness.