Loading...

Quantification of Partial Similitude and Multi-Fidelity Design Optimization of Dynamic Similar Models

Hajipourzadeh, Pedram | 2022

107 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 55413 (45)
  4. University: Sharif University of Technology
  5. Department: Aerospace Engineering
  6. Advisor(s): Banazadeh, Afshin
  7. Abstract:
  8. In this thesis, a framework for measuring the relative dynamic similarity and multi-fidelity design of the dynamic similar model of aircraft is presented. The conditions for establishing dynamic similarity between the scaled model and the original aircraft have been investigated for different applications and the methods for establishing relative dynamic similarity have been studied. In related researches conducted in the field of designing similar dynamic models of flying devices, the degree of relative similarity has not been measured and usually one or more scaling laws have been omitted according to the type of application. In this thesis, it has been shown that partial dynamic similarity or ignoring one or more scaling laws causes a significant difference in the dynamics and performance of the scale model. For this reason, a quantitative and measurable criterion has been proposed to determine the degree of relative similarity and the degree of importance of each of the scaling laws. By using the provided criterion, the relative similarity between the model and the original aircraft can be measured and optimized. Practical solutions to increase the relative dynamic similarity, using the weight and balance optimization of the model, as well as changing the geometry of the wing section, have been presented. According to the goals of designing a dynamic similar model, a framework has been proposed to optimize the design of the scale model, so that there is the most dynamic similarity between the scaled model and the original model, and on the other hand, other goals of the model design, including desirable dynamic behavior, dimensions and the construction cost is also optimized. The characteristics of the scale model, including the geometric scale, flight speed, weight and balance, and the cross-sectional area of the wing, have been optimized in such a way that the greatest relative dynamic similarity is achieved for the scale model. Due to the complexity and multi-subject design of the scale model, multi-fidelity optimization has been used. In this method, first by evaluating the cost function at limited points, the surrogate model with low accuracy of the cost function is developed, and then optimization is done with a combination of the evaluation of the surrogate model and the original cost function. The design method presented in this thesis leads to the design of a scale model with more dynamic similarity to the main aircraft compared to scale models with partial dynamic similarity.
  9. Keywords:
  10. Design Technique ; Multi-Fidelity Optimization ; Approximate Dynamic Similarity ; Dynamic Similar Model

 Digital Object List

 Bookmark

No TOC