Design and Testing of an Indirect Ice Detection Methodology

Distinct atmospheric conditions containing supercooled large droplets (SLD) have been identified as cause of severe accidents over the last decades as existing countermeasures even on modern aircraft are not necessarily effective against SLD-ice. Therefore, the detection of such conditions is crucial and required for future transport aircraft certification. However, the reliable detection is a very challenging task. The EU funded Horizon 2020 project SENS4ICE targets this gap with new ice detection approaches and innovative sensor hybridization. The indirect ice detection methodology presented herein is key to this approach and based on the changes of airplane flight characteristics under icing influence. A performance-based approach is chosen detecting an abnormal flight performance throughout the normal operational flight. It is solely based on a priori knowledge about the aircraft characteristic and the current measurable flight state. This paper provides a proof of concept for the performance-based ice detection: starting with the evaluation of operational flight data for different example aircraft the expectable flight performance variation within a fleet of same type is shown which must be smaller than the expected icing influence for reliable detection. Next, the implementation of the indirect ice detection system (IIDS) algorithms in SENS4ICE is detailed with certain regard to the flight test implementation for final validation. Finally, the initial methodology verification and validation results are presented and discussed.