Estimation and Evaluation of Energy Optimization Potential with Machine Learning Method for Active Hydrostatic Lubrication Control at Cradle Bearing in Axial Piston Pump

The tribology behavior influences the pump's efficiency, stability, functionality, and life cycle. Electronification extends the limitation of mechanical design and provides more optimization possibilities. This paper focuses on the active hydrostatic lubrication control (AHLC) with PID Controller for tribological pair between swashplate and cradle bearing in axial piston pump to minimize the energy dispatch. An novel approach is proposed for modeling the hydrostatic lubrication system at cradle bearing, optimization of the global optimal recess pressure, and evaluation of energy optimization potential with AHLC systematically. The new developed machine learning-based method considers the expensive simulation time and the data querying cost. The results show that the energy dispatch at the cradle bearing can be reduced enormously using AHLC.