Fault detection and isolation for electro-mechanical actuators with metaheuristic algorithms: sparrow search, honey badger and dandelion optimizer

In the aerospace community, the More Electric Aircraft (MEA) concept represents one of the most popular research hotspots since its first introduction in the 1990s. MEAs, as the name implies, show an increased electrification in many subsystems. They typically entail the elimination of hydraulic power generation from the engine gearbox, the reduction of bleed air off- takes and the growing utilization of power electronics within the main engine's starter/generation system. The potential benefits range from an overall lower subsystem weight and a higher power management integration on board to Life Cycle Costs (LCCs) contractions, just to name a few. The Flight Control System (FCS) is not the first subsystem one may think about as far as electrification is concerned, but its implications on the hydraulic system sizing and design are definitely remarkable. In order to design a sized down hydraulic system, engineers and researchers are looking at electromechanical actuators (EMAs), which can seamlessly convert electrical energy into mechanical energy, without the need of hydraulic power. EMAs could be then employed to control primary and secondary flight controls surfaces. The main issue which is still preventing EMAs from taking over the flight control field in commercial aviation is linked to safety and reliability, and hence to certification. In fact, since EMA reliability cannot meet the required standards, practical EMA applications are severely limited. A way to boost the acceptance of these systems for safety critical applications could involve providing the aircraft FCS with additional checks and Fault Detection and Isolation (FDI) capabilities, integrated in a so-called Prognostic and Health Management (PHM) framework. This paper continues the already popular thread of FDI solutions for EMAs employing Metaheuristic Search Algorithms (MSAs) in a model-based FDI strategy by approaching new MSAs: Sparrow Search Algorithm (SSA), Honey Badger Algorithm (HBA), Dandelion optimizer algorithm (DOA). After a short introduction on EMA and MSA, the methodology is explained. Finally, a brief comparison between the methods and a conclusion section follows