Optimization in Conceptual Design of a BWB UAV Using an Adaptive Kriging‐Based Surrogate Model

To improve optimization efficiency in the conceptual design of Blended Wing Body (BWB) Unmanned Aerial Vehicles (UAVs), an adaptive Kriging‐based surrogate modeling framework is established. This framework addresses the high computational cost inherent in repeated numerical simulations in multidisciplinary analysis (MDA). First, a streamlined MDA model integrating geometry, mass, aerodynamics, and flight performance was developed. Subsequently, three optimization strategies, comprising a baseline non‐adaptive Kriging model, an exploration‐oriented strategy, and an exploitation‐oriented strategy, were implemented and compared for the objective of maximizing flight range. To ensure robustness and mitigate stochastic bias, 15 independent randomized trials were conducted for each strategy. Statistical results demonstrate that the exploitation‐oriented strategy delivers superior performance, with the median optimization result achieving a 14.5% range enhancement over the baseline design. Notably, for this representative optimal configuration, the discrepancy between the Kriging prediction and the numerical simulation is a mere 0.48%. Finally, a dual‐method parameter sensitivity analysis identifies fuel mass as the most influential parameter, followed by the center wing span. The proposed framework provides an efficient and robust methodology for the global optimization of BWB UAVs.