This paper addresses the optimization problem of the overall design stage of Hydrogen-powered Unmanned Aerial Vehicles (H-UAVs) in the context of heterogeneous multi-source domains. It explores how to effectively utilize the transfer learning technology to establish a surrogate model and optimize H-UAVs in the presence of heterogeneous samples. To solve the problem of high cost of building a surrogate model due to heterogeneous samples during the evolution of hydrogen-powered UAVs, a framework for establishing a Multi-Source domain Fusion (DG-MSF) surrogate model is proposed based on Data Generation. The geodesic flow kernel method is used to map the heterogeneous source and target domains to a high-dimensional space to determine the relationship between multi-source domains. The marginal distribution-based data generation method is used to effectively integrate source domain information. A multi-layer perceptron neural network is built as a surrogate model, and is trained and fine-tuned through pre-training and fine-tuning methods to achieve efficient prediction of performance of H-UAVs. Finally, the optimization design of the H-UAV is carried out. The analysis results show that the proposed method can effectively utilize multi-source domain data to improve the efficiency of model training and prediction accuracy and the overall performance of H-UAVs, providing powerful technical support for the development of H-UAVs.

For the conceptual design phase of Unmanned Aerial Vehicles (UAVs), a process for conceptual design and configuration selection of Solar/Hydrogen powered UAVs (S/H-UAVs) is proposed. The design requirements of S/H-UAVs were analyzed firstly. The proposed process used Fuzzy Quality Function Deployment (FQFD) to establish logical and quantitative standards. Moreover, in order to appropriately describe the hesitancy of experts while making decision, it used Q-Rung Dual Hesitant Fuzzy Sets (QRDHFS) to score the correlationships. In addition, a decision-making framework is proposed to perform a logical selection of typical layouts based on defuzzification method and Technique for Order Preference by Similarity to the Ideal Solution (TOPSIS). The present process has been applied for S/H-UAVs. The resulting set of design requirements consists of three categories: Mission Requirements (MRs), Engineering Characteristics (ECs) and Technical Indicators (TIs). Four typical layouts of S/H-UAVs were sorted and determined. The performance of four typical layouts were evaluated and the Strut-Braced Wing (SBW) with external hydrogen storage was selected as the best layout for S/H-UAVs.

The trajectory related and Direct Current (DC) Electromagnetic Interference (EMI) of lithium battery, fuel cell and photovoltaic modules has a great influence on the small-scale Unmanned Aerial Vehicle (UAV) airborne magnetometer and is hard to be shielded, calibrated or filtered. Besides, the mechanisms underlying the DC EMI have been rarely investigated yet. To cope with this problem, this paper systematically studies the EMI models, and proposes an online 3-layer EMI reduction scheme. First, EMI coupled with UAV motion model and hybrid power system is established. Second, the mechanism EMI models of hybrid power system are established and verified based on the proposed concept “equivalent current”. Third, an online 3-layer EMI reduction scheme is proposed, including battery layer, trajectory planning layer and energy management layer. In the first main layer, EMI self-cancellation is realized by rotating battery inclinations and symmetrical circuits. In response to errors, the trajectory planning layer reduces the EMI intensity by optimizing an optimal trajectory, while the energy management layer prioritizes power allocation to power sources that can produce small and stable EMI. Simulation results of climb, level flight and descent illustrate the efficaciousness and applicability of the proposed online 3-layer EMI reduction scheme.