In-depth Feature Selection for PHM System's Feasibility Study for Helicopters' Main and Tail Rotor Actuators



Published Jul 19, 2020
Andrea De Martin Giovanni Jacazio Andrea Nesci Massimo Sorli


As deeply complex machines subjected to heavy vibratory environment, helicopters require relative low mean time between overhaul and suffer from high maintenance cost and availability issues. So far, PHM for helicopters has been aimed at detecting the presence of structural defects in the most critical parts of the mechanical transmission conveying power from the engine to the rotor blades, but very little has been presented on other flight-critical components, such as the main and tail rotor actuators. The proposed paper is focused on preliminary diagnostics and prognostics considerations for a traditional configuration of hydraulic solution, where a tandem actuator is aided by a Stability and Command Augmentation System (SCAS) during operations.

At first, the case-study is introduced and the simulation model employed for the analysis is described. Hence, two different failure modes affecting the SCAS are investigated and the physical models used to describe their progression are presented. In-depth data mining is then applied to achieve an accurate feature selection from raw data and an original way to visualize features' performances through an accuracy-sensitivity plane is proposed. Lastly, a particle filtering approach is adopted for failure prognosis and its output evaluated through traditional PHM metrics to assess the algorithm effectiveness. The present research provides encouraging results regarding the opportunity of realising a PHM system for helicopters' flight control actuators without the need of additional sensors, which could make solutions based upon the presented work feasible for both in-service and future platforms.

How to Cite

De Martin, A., Jacazio, G., Nesci, A., & Sorli, M. (2020). In-depth Feature Selection for PHM System’s Feasibility Study for Helicopters’ Main and Tail Rotor Actuators. PHM Society European Conference, 5(1), 9.
Abstract 132 | PDF Downloads 113



prognostics, hydraulic flight control, stability and command augmentation system, particle filtering, feature selection

Technical Papers