Modeling of hydraulic systems tailored to diagnostic fault detection systems

The consistent and reliable operation of hydraulic componentry is paramount for many systems. From spacecraft to the most basic automotive bottle jack an undetected failure can have significant consequences if not noticed in time. Many hydraulic systems have diagnostics capabilities but these are normally very limited in scope. They can detect events only related to specifically monitored components or general system failures. Typically these diagnostic systems are designed after the fact and tuned to meet goals. When hydraulic systems are designed, simulation models are frequently used to gain some idea of the finished systems performance. Rarely is the simulation model designed to accommodated and optimize a diagnostic capability. The number and placement of diagnostic sensors can have a significant effect on the ability of a diagnostic system to resolve faults early in their evolution cycle. This paper describes a technique developed at the Penn State Applied Research Laboratories Systems Operations and Automation Department for the design of hydraulic simulation models that pre-incorporate fault diagnostic advanced design features. This technique is being applied to an US Army M1120 Heavy Expanded Mobility Tactical Truck (HEMTT) Load Handling System (LHS) supply vehicle that utilizes a palletized hydraulic loading system. The test vehicle has a hydraulic system that was fully instrumented with sensors for this work. This paper addresses a piece in the diagnostic puzzle that has until now bee little looked at. Namely what additional features in a simulation model will allow for optimal placement and monitoring of the hydraulic system. The researchers have found that this typically leads the readdressing or removing certain engineering assumptions that are typically made when designing simulation models. When this is done the models are more flexible when it comes to diagnostic implementation.