Cubesat development : integrating the current state-of-the-art in manufacturing techniques and engineering analysis software

Abstract

With the advancements in technologies over the last decade, computational power, machine learning and artificial intelligence has revolutionized software, manufacturing capabilities and automation, amongst others. While the production industry has been experiencing the Fourth Industrial Revolution and Industry 4.0, engineering development can also experience a similar paradigm shift with the use of advanced software, to create digital twins for advanced simulations, and use advanced manufacturing methods, such as additive manufacturing, to optimise product design and assist in product development. The aim of this research was to implement the current state-of-the-art in manufacturing techniques and engineering analysis software for CubeSat development. With Operation Phakisa underway, the devel-opment of CubeSat constellations is in focus at the African Space Innovation Centre (ASIC) at Cape Peninsula University of Technology (CPUT). To develop constellations efficiently, following the mo-tivating force of Operation Phakisa, rapid prototyping and additive manufacturing was to be imple-mented. Methods and procedures were to be determined and implemented in using additive manufac-turing to improve general CubeSat development. A CubeSat structure was to be designed to be suitable for additive manufacturing. The effect these additive manufacturing methods had on cost saving was to be determined. Advanced simulation software was to be used by developing a thermal digital twin of ZACube-2 CubeSat in the form of a finite element model and simulate the thermal orbital conditions to replicate that of the ZACube-2 orbit. The results of the simulation were to then be compared against actual data from ZACube-2, to validate the accuracy of the methods and model used for the digital twin. With the constant development of additive manufacturing abilities and the advancements in CubeSat development, this research has shown how additive manufacturing can have a significant contribution towards CubeSat development by detecting design issues early on in the project, reducing costs, lead times and optimizing the design process. A 3D printed structure printed out of ULTEM 9085 was de-signed that met the structural and compatibility requirements. A finite element model of the structure was developed using orthotropic material properties to represent the 3D printed material characteristics. Finite element analysis on the structure, using both SOL101 and SOL103 Nastran solvers for static and harmonic analysis respectively, resulted in a maximum stress, 30% less than yield when subjected to the launch conditions loading and with a natural harmonic mode of 193 Hz, above the 100 Hz require-ment. The structure was also designed to be compatible with commercial-off-the-shelf (COTS) compo-nents and proved to be more efficient to assemble and integrate by reducing the number of parts. The structure is also 14-39% lighter than COTS equivalents. The structure costs 57% more than manufac-turing an aluminium structure when outsourced, however the lead time can be reduced from 3 weeks to 2 days. This cost would be significantly lower however, should the structure be 3D printed in-house. Furthermore, the structure is lighter, easier to assembly and integrate and can be easily modified to suit the bus or payload needs. Building a Concept Model, an accurate 3D printed replica with actual connectors and harnesses, as a new deliverable at the end of Phase B of a CubeSat mission, resulted in reducing the extra costs by 88% for the MDASat-1 CubeSat mission. This was achieved by identifying errors that were not picked up in CAD but discovered when conducting a trial assembly of the Concept Model. ASIC also saved signif-icant time by identifying these issues early into the project rather than when building the engineering model (EM). Making use of the Concept Model saved costs towards the CubeSat mission as well as prepared and trained engineers for building the engineering and flight models of the CubeSats. The current state-of-the-art in advanced engineering analysis software have made structural and thermal analysis more accessible and user friendly for design engineering. This allows for improved, optimised designs as well as the validation of the system and spacecraft design. The digital twin of ZACube-2 was developed by creating a simplified finite element model (FEM) of the CubeSat using 2D, and 1D ele-ments. A simplified list of materials was used with accurate thermal and thermo-optical properties when defining the mesh of the FEM. A transient, orbital thermal simulation was conducted, replicating 15 orbits of ZACube-2, calculating the thermal radiation experienced by the CubeSat every 3 minutes. The digital twin showed to be very accurate when comparing the simulated results with temperature data acquired from ZACube-2, with a maximum variation at maximum and minimum temperatures of 2° C. Of the literature that was reviewed, none of the other researchers showed any form of validation of their results. Running the hypothetical dusk-dawn orbit simulation for ZACube-2 using the developed FEM showed that the satellite would be 15° C hotter in dusk-dawn than that of its current orbit. This would result in the overheating of components which could lead to catastrophic failure and finally mission failure. As part of this work a digital twin was developed for ASIC allowing them to better predict the thermal conditions the satellite components would be subjected to for a particular orbit. The methodol-ogies developed in this research to create a thermal digital twin of a CubeSat could be applied to future missions or by other researchers to more optimally predict the thermal environment. Educated decisions can be made regarding orbit selections or contingency plans can be made to control the excess heat, such as passive heat control measures. While Industry 4.0 is a term used predominantly in production and automation, some aspects of the revolution can be clearly seen to be implemented in the research and development engineering industry, with the advancements in additive manufacturing and the making use of the digital twin concept for engineering analysis, as demonstrated in this research.