The Prognostic Health Management (PHM) Lab is dedicated to the study of machine health monitoring of engineering systems. This research comprises of fault detection & isolation, state-of-health assessment, and remaining useful life prediction across several technologies. The primary test setup in this lab is a MOOG industrial MaxForce Electromechanical Actuator (EMA) that is used to represent an all-electric flight actuator presently used on secondary control surfaces on many aircraft. The system is instrumented with a variety of sensors including accelerometers, phase current, LVDTs, and rotary encoders that are connect to a dSPACE rapid prototype control & data acquisition system with Matlab/Simulink software interface. Present graduate-level research is being conducted into various mechanical faults of these units, such as bearing and ball screw, through data-driven and model-based Bayesian classification methodologies. More electric aircraft is the desire of many in the aerospace industry with EMAs being a primary component. For this technology to be commercially accepted improved integrated health monitoring is crucial. In addition to the graduate research, undergraduate independent studies, senior design teams, and summer coops have continued to advance the labs capability through advanced sensor design and installation, EMA test fixture upgrades, and benchtop EMA prototype design and fabrication for rapid turnaround of conceptual sensor and algorithm development of targeted failure modes.
In addition the PHM lab is the home to another industry focused research project on state-of-health prediction in the area of automotive fuel cell science. Through a project with General Motors - Fuel Cell Activities (GM-FCA) graduate research is continuing by applying health monitoring techniques to model-based simulations of an automotive fuel cell powertrain focused on degradation modes. Actual full-scale (100kW+) fuel cell module data that has been run in test stands at GM-FCA is used to validate the proposed predictive algorithms. In order for this transportation method to reach the marketplace health prediction will go a long way to solving many of the durability questions behind this transformative science. Dr. Jason Kolodziej manages this lab.
The following have provided support for this lab: