Multi-fidelity deep neural network surrogate model for aerodynamic shape optimization

Development and implementation of an advanced, design-sensitive method for wing weight estimation

Progress in Additive Manufacturing of Energetic Materials: Creating the Reactive Microstructures with High Potential of Applications

Parameterization and optimization of hypersonic-gliding vehicle configurations during conceptual design

Fabrication and investigation of 3D-printed gun propellants

A new parameterization method to enhance shape optimization for aircraft design was developed based on the class-shape-transformation method. This method was implemented in a tool that combined all aspects of the aerodynamic design process: parameterization, aerodynamic analysis, and optimization. The parameterization method used a combination of Bernstein polynomials and B-splines to allow for both local and global control of a shape. Additionally, the use of B-splines made it possible to efficiently handle volume constraints, which are very common in aircraft design. The parameterization method was coupled to two different aerodynamic analysis tools: the commercial panel method code VSAERO and an in-house Euler code. As a first test case, a wing was optimized for maximum lift-to-drag ratio in subsonic conditions using VSAERO and various optimization algorithms. It was shown that the optimizer was capable of arriving at wing shapes that demonstrated laminar flow up to 80%, without violating the implemented volume constraints. As a second test case, an airfoil was optimized for lift-to-drag ratio in transonic conditions using the in-house Euler solver and a gradient-based optimizer. Refining the shape using B-splines proved to be an efficient way of increasing the design freedom, resulting in additional improvements in the drag divergence Mach number in the order of 0.02.

Extension to the Class-Shape-Transformation Method Based on B-Splines

A photocurable energetic resin was developed for photopolymerization additive manufacturing. The composition contains 50 wt % RDX, 25 wt % acrylate binder, and 25 wt % energetic plasticizer. The material was characterized in terms of compatibility, printability, mechanical properties and (ballistic) performance. The possibility of printing energetic items of increasing complexity was demonstrated through various print trials. The culmination of the research effort was the firing of a 30 mm gun setup with 3D-printed propellant.

Development of Propellant Compositions for Vat Photopolymerization Additive Manufacturing

The Multi-Disciplinary Design Optimisation (MDO) process can be supported by partial automation of analysis and optimisation steps. Design and Engineering Engines (DEE) are useful concepts to structure this type of automation. Within the DEE, a product can be parameterically defined using Knowledge Based Engineering (KBE). This parameteric product model needs to be initiated before global multidisciplinary optimisation can be performed. This paper presents the first phase in the development of an aerodynamic initiator tool. This tool combines all aspects of the aerodynamic design process, thereby allowing the designer to efficiently determine a feasible aircraft shape that can be used as an initial state for the MDO. This research is performed as part of the CleanEra project at the Faculty of Aerospace Engineering at Delft University of Technology. The Class-Shape-Transformation (CST) method is used to create a parameteric description of a blended-wing-body (BWB) aircraft. The method is then expanded to allow for more local control of the aircraft shape. The mathematical description of the BWB is then translated to an input for the panel method program VSAERO, which outputs a number of relevant flow characteristics. These are then fed into an optimisation algorithm which generates a new aircraft shape and the process is repeated. The CST method has proven to be very useful as a parameterisation tool. VSAERO is considered sufficient for the flow analysis for now, but should later be replaced by a Euler or Navier-Stokes code. Work on the optimiser will start later this year.

https://repository.tudelft.nl/islandora/object/uuid%3A107c7106-4fc7-4427-8d6d-7e56a9ec3d79/datastream/OBJ/download

Aerodynamic shape parameterisation and optimisation of novel configurations

Since common energetic materials are not easily compatible with existing additive manufacturing techniques, the Energetic Materials department of TNO in the Netherlands worked closely together with the Additive Manufacturing and Material Solutions groups of TNO to develop new types of energetic material that can be processed in standard 3D printers, aiming at better performance and greater flexibility in the manufacturing of energetic products. This contribution gives an overview of potential applications for additive manufacturing of energetic materials, ranging from rocket propellants to explosives, gun propellants, and other applications. The paper then briefly describes TNO’s research efforts since 2013 into AM of energetic materials. Next, the work performed on additive manufacturing of gun propellants, using stereolithography, is described in more detail, including the results of the material development and characterization, and performance test results, including test firings with a 30 mm gun. The paper concludes with an overview of the current status of the research at TNO, as well as a definition of the next steps to be taken and a vision of the future of additive manufacturing of energetic materials.

Developments in additive manufacturing of energetic materials at TNO

This thesis introduces a new parameterization technique based on the Class-Shape-Transformation (CST) method. The new technique consists of an extension to the CST method in the form of a refinement function based on B-splines. This Class-Shape-Refinement-Transformation (CSRT) method has the same advantages as the original CST method, while also allowing for local deformations in a shape. A number of test cases were performed using two different design frameworks with low and high fidelity. The low fidelity framework was based on a commercial panel method code and coupled to various optimization algorithms. The high fidelity framework used an in-house Euler code and employed adjoint optimization.

M.H. Straathof

Papers

Extension to the Class-Shape-Transformation Method Based on B-Splines

Development of Propellant Compositions for Vat Photopolymerization Additive Manufacturing

Aerodynamic shape parameterisation and optimisation of novel configurations

Developments in additive manufacturing of energetic materials at TNO

Shape Parameterization in Aircraft Design: A Novel Method, Based on B-Splines