This article proposes a novel approach to support Knowledge Based Engineering (KBE) application development based on Model-Based Systems Engineering (MBSE). In this methodology, the related knowledge is captured in a well-structured Systems Modeling Language (SysML) model, instead of (static) documents. The knowledge model is then automatically translated to application (skeleton) code using a model-to-code tool developed in this research. The proposed methodology is applied to a use case at GKN Fokker Elmo for the development of a KBE application to design Electrical Wiring Interconnection Systems (EWIS) architectures for aircraft. The results show that the proposed MBSE approach improves the knowledge acquisition process, reduces the time needed for developing new KBE applications (initial knowledge model and code skeleton) by almost 50%, and enables traceability of requirements within the KBE application and knowledge model. These benefits allow effective project-to-project knowledge transfer while mitigating the black-box effect often experienced by KBE application users. In the next phase of this research, reverse engineering capabilities will also be incorporated to enable code-to-model translation, so as to guarantee the application code and knowledge model synchronization throughout the application's lifetime. @en

By 2050, the aviation industry is expected to grow by 250-300% of its current air traffic. If left unchecked, the corresponding increase in atmospheric and noise pollution will have a catastrophic impact on the environment. Thus, improvements in aircraft design are urgently needed for a sustainable growth of the aviation industry. The performance and behaviour of existing tube-and-wing aircraft have been refined and improved so much in the past decades that further improvements are barely possible. Consequently, by 2050, breakthrough solutions in the form of unconventional design configurations, novel propulsion systems and operation models are required to meet the ambitious goals stated in Europe’s vision for Aviation. However, the design of unconventional aircraft is particularly challenging as it often involves integrated multi-functional components for which legacy data is unavailable. Thus, appropriate means to assess their performance and behaviour are needed to lower their industrial development risk. Sub-scale Flight Testing (SFT), an experimental approach involving the free-flight testing of sub-scale models with an on-board powerplant, shows promise in the evaluation of the in-flightmotion of a given aircraft configuration and its response to control inputs. In the past, SFT has been used in a wide-range of flight tests to study the effect of novel technologies on the aircraft flight behavior, to assess systems integration feasibility and as a proof-of concept for unconventional designs. The actual benefit and validity of SFT mainly depends on the design of the SFT model used for the test. A well designed SFT model can show a similar behaviour to the fullscale aircraft such that any observation on the scaled device can be directly used to predict the full-scale performance. However, ensuring similarity between SFT model and full-scale aircraft is challenging due to the differences in their size and flight conditions. In addition to guaranteeing similitude, SFT model must comply with multidisciplinary design requirements such as safe completion of the mission, adhering to restrictions imposed by local authorities and selecting suitable flight-control and measurement equipment...@en

Growing interest in unconventional aircraft designs coupled with miniaturization of electronics and advancements in manufacturing techniques have revived the interest in the use of Sub-scale Flight Testing (SFT) to study the flight behaviour of full-scale aircraft in the early stages of design process by means of free-flying sub-scale models. SFT is particularly useful in the study of unconventional aircraft configurations as their behaviour cannot be reliably predicted based on legacy aircraft designs. In this paper, we survey the evolution of various design approaches (from 1848 to 2021) used to ensure similitude between a sub-scale model and its full-scale counterpart, which is an essential requirement to effectively perform SFT. Next, we present an exhaustive list of existing sub-scale models used in SFT and analyse the key trends in their design approaches, test-objectives, and applications. From this review, we conclude that the state-of-the-art sub-scale model design methods available in literature have not been used extensively in practice. Furthermore, we argue that one sub-scale model is not sufficient to predict the complete flight behaviour of a full-scale aircraft, but a catalog of tailored sub-scale models is needed to predict full-scale behaviour. An introduction to the development of such a catalog is presented in this paper, but the development of a formal methodology remains an open challenge. Establishing an approach to develop and use a SFT catalog of models to predict full-scale aircraft behaviour will help engineers enhance confidence on their designs and make SFT a viable and attractive testing method in the early stages of design.

@en

Sub-scaled Physical Models (SPM) are often employed in wind-tunnel tests or in free-flight tests (physical tests) to predict flight behavior of aircraft Full-scale Design (FD). However, a quality prediction of both the static and dynamic behavior is to date an open challenge. In this research, a methodology for designing SPMs is proposed for those cases where dynamic similarity between SPM and FD cannot be achieved and legacy information to compare subscale flight results to FD is unavailable. Instead of attempting to use just one SPM to achieve complete similarity with full scale design, this methodology enables the design and comparison of multiple SPMs to determine the Sub-scale Design (SD) best suited the estimation of specific aspect the FD flight behavior. To this purpose, a metric called Degree of Similitude (DoS) is defined, to quantify the similarity of FD and SPM based on the aerodynamic coefficients that are relevant for a given test. The DoS estimation first requires the evaluation of relevant aerodynamic coefficients, by means of Computational Fluid Dynamics (CFD). CFD analysis, requires complex geometry generation, adequate grid generation, expensive calculation and laborious post processing. To this purpose, a Knowledge Based Engineering (KBE) tool called Multi-model Generator (MMG) is developed, to automate all the labor intensive tasks in the evaluation of the DoS including the integration of CFD tool. Validation of results produced by MMG-VSAERO tool-chain is performed by means of a wind-tunnel test campaign using a 8.8% aerodynamically scaled SPM of the Cessna Citation II 550 (citation). The results of this test are compared with flight test data of full-scale aircraft (which is co-owned and operated by Delft University of Technology). Furthermore, this SPM was compared with three other Sub-scale Designs (SD) to estimate their DoS with the full-scale aircraft for two different eigenmodes, namely short period mode and phugoid mode. Of the four SDs compared, it was found that the geometrically scaled SD showed highest DoS for short period motion and one of the aerodynamically scaled SD had highest DoS for phugoid motion. From the cases studied, it can already be inferred that geometrically scaled SDs are not always preferred and in many cases, aerodynamically scaled SDs can be much more similar to FD. This case study proved the convenience of the proposed coefficient DoS which, in the next phase of the research, will be used as objective function to design optimum SPMs for a given test. @en

@en

Sub-scale Flight Testing (SFT) is potentially useful in predicting aircraft flight behaviour, especially in the case of unconventional designs for which legacy information is unavailable and wind tunnel tests are unable to predict aircraft dynamics. A necessary condition for SFT is the design of properly scaled models. However, even in case of perfect scaling, the sub-scale model needs adequate flight performance and handling qualities to enable the execution of flight tests. Thus, the (static and dynamic) stability and control (S&C) and handling qualities (HQ) of sub-scale designs should be evaluated accurately as well as quickly, to allow conceptual design iterations. To this purpose, we propose the use of a 3D panel method (3DPM) for the generation of the non-linear aerodynamic database, in combination with a non-linear flight dynamics analysis. Two main challenges affect the proposed approach. The first concerns the validity of the low-fidelity 3DPM data for the assessment of the sub-scale design S&C and HQ. The second is about the time consuming and error-prone pre/post-processing activity demanded by the hundreds of analysis cases for the aerodynamic database generation. The first issue is investigated by predicting the longitudinal S&C performance and HQ of a sub-scale design using 3DPM analysis and comparing them with the prediction from wind-tunnel test (static) data supplemented by (dynamic) data from 3DPM. Both models appear trimmable and stable and the difference in their HQ are quantified, thus verifying the suitability of 3DPM analysis for sub-scale design assessment. The pre/post-processing challenge is tackled by the development of a knowledge-based engineering application to automate the aerodynamics database generation, reducing the time needed for geometry modeling, discretization and postprocessing of hundreds of cases from weeks to hours. The proposed methodology and its flexibility are demonstrated in this paper, where a commercial 3DPM code and an in-house developed non-linear flight dynamics analysis tool have been used to assess two sub-scale designs, one conventional and one based on the box-wing configuration. @en

This paper addresses the aerodynamic performance and numerical modeling of over-the-wing propellers. Installing the propeller above a wing has the potential to increase wing lift-to-drag ratio, high-lift capabilities, and to reduce flyover noise. However, the prediction of its performance is difficult, since research on the aerodynamic interaction effects of over-the-wing propellers has been limited so far. For this reason, an exploratory wind tunnel campaign was performed with a wing featuring a fowler flap. A single propeller was installed above the wing at different chordwise locations and inclination angles. Wing surface-pressure and wake-pressure measurements showed strong, bilateral aerodynamic coupling between the propeller and wing. A configuration with the propeller attached to the flap showed wing lift increases of 8% and 3% in cruise and high-lift conditions, respectively. The key findings of the wind tunnel campaign were used to validate a low-fidelity numerical tool, which combines a non-uniform inflow blade-element model for the propeller, a panel method for the wing, and a vortex lattice model for the propeller slipstream. The numerical model was used to assess the effect of propeller axial location and diameter. Results indicated that the optimal axial propeller position is near the trailing edge of the wing, and that reducing the propeller diameter at constant thrust coefficient at this location is beneficial for distributed propulsion applications. The tool allows a rapid computation of over-the-wing propeller and wing performance in cruise conditions. This enables an efficient design space exploration during the conceptual design process of such configurations.@en