GRNN-based Hydraulic Hose Bulk Modulus Measurement and Prediction Method.

Small-scale hydraulic actuation systems are especially suitable for narrow space and harsh environment arrangement. Due to its flexibility, thin and long hose is used to connect components, uncertainty of the system model resulted from the hose expansion has severe impacts on control properties of the nonlinear system. In order to solve the uncertainty problem, this paper proposes a measurement method of hose bulk elastic modulus and a hose modeling method based on Generalized Regression Neural Network (GRNN). By constructing a dedicate test device, the pressure and volume changes of the hose are measured, and the raw data needed for the hose modeling is obtained. The simulations show that the hose model based on GRNN has good effects on predicting the volume dynamic change process, and could be applied to the modeling and simulation of hydraulic system with flexible hose satisfactorily.