UNcertainty quantification and modelling Bias Inhibition by means of an Agnostic Synergistic Exploitation of multi-fidelity Data

Informazioni relative al progetto

UN-BIASED

ID dell’accordo di sovvenzione: 101059320

DOI

10.3030/101059320

Progetto chiuso

Data della firma CE 13 Giugno 2022

Data di avvio 4 Ottobre 2022

Data di completamento 3 Ottobre 2024

Finanziato da

• Marie Skłodowska-Curie Actions (MSCA)

Costo totale

Nessun dato

Contributo UE

€ 172 750,08

Coordinato da

POLITECNICO DI MILANO
Italy

Periodic Reporting for period 1 - UN-BIASED (UNcertainty quantification and modelling Bias Inhibition by means of an Agnostic Synergistic Exploitation of multi-fidelity Data)

Periodo di rendicontazione: 2022-10-04 al 2024-10-03

Sintesi del contesto e degli obiettivi generali del progetto

These are the brightest times of the data-science era, continuously striking priceless technological advancements towards the betterment of Society. Jim Gray, winner of the prestigious Turing award, recognised data science as the fourth paradigm of Science, together with experiments, theory and computation.
Interestingly, Scientific Modelling is still a heavily hypotheses-driven process, strongly biased by the subjective thinking of the human mind. The recent outburst of Data Science techniques opens the path to innovative modelling paradigms. Despite the complexity of the phenomena under investigation, data-driven regression procedures seek an unbiased implicit approach to our learning experience, based on raw data from actual observations.
The proposed action aims at developing an innovative Scientific Modelling (SM) paradigm closely entwining data-driven (DD) and hypotheses-driven (HD) techniques to potentially reduce, if not correct, possible cognitive biases concerning the Modeller’s subjective understanding of reality.
The goal is to demonstrate the proposed methodology on a complex application of great interest to the aerospace industry, namely, tilt-rotors and multi-rotor machines.
These are at the cutting edge of the modern aeronautic industry. Their unique capability of combining vertical take-off and landing with a high cruise speed, comfort, and range, makes them very attractive to the short-haul regional market, with a particular reference to electric Urban Air Mobility (UAM), search and rescue, emergency medical services and service to isolated areas.
Several multi-rotor configurations are presently developed for air-taxi applications. Despite the large amount of resources pledged by the industry, many challenges remain unanswered. In particular, performance predictions are hampered by the complexity of the aerodynamics of diverse flight configurations, e.g. hover and vertical-to-horizontal flight transition.
In this context, the industry calls for revolutionary modelling and design paradigms to improve the performance of multi-rotor machines.
This action aims to accommodate this need by crafting an agnostic multi-fidelity modelling framework establishing a synergy between the theory-to-data and the data-to-theory perspectives to identify and possibly mitigate epistemic uncertainty in experimental and computational models for tilt-rotors aerodynamics.

Lavoro eseguito dall’inizio del progetto fino alla fine del periodo coperto dalla relazione e principali risultati finora ottenuti

The activities carried out within the project aim to establish a novel modeling framework for mitigating possible modeling biases.
Beginning with the modelization of UAM applications into computerized models, the project builds a multi-fidelity framework for simulating the aerodynamics of single and multi-propeller configurations.
The framework allows to obtain predictions based on a different level of approximation of the complex physics underlying the aerodynamic flow developing around the aircraft (of parts of it).
This data serves the calibration of Data-Driven models which can either consist of heuristic corrections to analytic models or surrogate multi-fidelity models.
After, a novel formulation for building co-kriging surrogates is proposed and assessed.
The new formulation is summarized in Figure 1, in which the standard recursive method is reported on the left, whereas the proposed one is on the right.
The new formulation allows the construction of debiased co-kriging models. Namely, it allows to mitigate the effect of possible modeling bias resulting in an incorrect fidelity hierarchy among the available models.
This new formulation is exploited to improve the optimization process of aerodynamic shapes.
Up to this stage, the methodology is applied to simple aerodynamic shapes or an isolated portion of the aircraft geometry e.g. UAM propellers.

Progressi oltre lo stato dell’arte e potenziale impatto previsto (incluso l’impatto socioeconomico e le implicazioni sociali più ampie del progetto fino ad ora)

The main outcome is the formulation of a novel paradigm to build co-kiring models for taking advantage of data sets of different fidelity and cost.
This new formulation shows, on average, superior performances with regards to the standard approach.
Results may be applicable to improve the design process of aircraft characterized by a complex configuration, such as eVTOLS for UAM which take advantage of multiple lifting surfaces and multiple propellers.
Nonetheless, the readiness level of the proposed methodology is still low and further research is required.

Image of a UAM propeller from a CFD simulation with isomach surfaces highlighted.

Diagram reporting a comparison between the standard formulation (left) and the proposed one (right)