Condition monitoring of hydraulic direction control valve

Abstract

In hydraulic systems, directional valves are responsible for regulating the flow of the fluid. Solenoid operated valves do so by mechanical movement of spool inside valve by the virtue of electromagnetic force generated by the applied control current. The deterioration in control current leads to the degradation in electromagnetic force and thus the spool takes longer to initiate as well as terminate the switching position. This delay or lag potentially causes the pressure, flow and power fluctuation and unintended impacts on the system. This project presents a comparative analysis of detecting these anomalies by acquiring pressure signals across the valve using extreme gradient boost (XGBoost) and 1-Dimensional Convolution Neural Network (CNN). Four handcrafted statistical features and four fractal dimensions train XGBoost whereas 1-D CNN with six hidden layers utilizes the raw signal of net pressure change across the valve. XGBoost predicts the switching behaviour at an accuracy of 99.68% and 1-D CNN performs at its maximum possible accuracy (100%). The very narrow gap signifies the nearly equal significance of both of these different category classifiers. As XGBoost cannot handle the raw signals, the pre-processing increases the time-consumption while 1-D CNN using its deep architecture efficiently maps the complexity of the hydraulic system using pressure signals and is robust to noisy data.