Noncontact estimation of the onset of ice accretion in turbofan stators

The problem of crystal ice formation inside aircraft turbine engines is well-documented, and poses a significant risk to safety. The problem is not only one of power loss in flight, but the very real possibility that a flame-out event could occur due to ice accretion on compressor stators, with potentially catastrophic outcomes. Although many instrumentation systems have been developed for wing ice detection, incipient formation of crystal ice is somewhat more difficult to detect. This is compounded by the need for a noncontact sensor which is robust to in-flight conditions. This paper proposes an approach to the detection of ice formation based on microwave transmission characteristics across the first and possibly the second stage of the compressor stator. It is shown that noncontact detection is feasible under realistic conditions. The contribution of this paper is twofold. First, the microwave transmission approach is motivated using wind tunnel measurements, and appropriate frequency bands are determined. Next, a signal processing approach involving higher-order analysis of time-frequency distribution characteristics is then put forward. Experimental results are presented to support the hypothesis that multiband detection offers a workable approach to the incipient crystal-ice detection problem.