Diagnosability Optimal Design Of UAV Based On Minimal Structurally Overdetermined And Sine-Cosine Algorithm

Abstract To improve the diagnosability of UAVs, this research provides a diagnostic ability optimization design method that utilizes the minimal structurally overdetermined(MSO) and sine-cosine algorithm(SCA). Firstly, based on the structural analysis, a qualitative structural diagnosability assessment method based on MSO is suggested; secondly, a diagnosability quantitative assessment technique based on the Wasserstein distance is proposed; and lastly, to take into account the diagnosability requirements as well as diagnostic system design cost, a diagnostic optimization design strategy relying on SCA is proposed. Based on the fixed-wing UAV structural model, its diagnosability is optimized in the simulation stage. The simulation results show that adopting the qualitative-based optimization strategy makes the fault detection rate 100%, the fault isolation rate 85.71%, and the diagnosability cost shrinks by 70.59%. Adopting a quantitative-based optimization strategy results in a system with a 100% fault detection rate and 100% fault isolation rate.