Thermal morphing anisogrid smart space structures part 2: Ranking of geometric parameter importance, trust region optimization, and performance evaluation

As future space mission structures are required to achieve more with scarcer resources, new structural configurations and modeling capabilities will be needed to meet the next generation space structural challenges. A paradigm shift is required away from the current structures that are static, heavy, and stiff, to innovative lightweight structures that meet requirements by intelligently adapting to the environment. As the complexity of these intelligent structures increases, the computational cost of the modeling and optimization efforts become increasingly demanding. Novel methods that identify and reduce the number of parameters to only those most critical considerably reduce these complex problems, allowing highly iterative evaluations and in-depth optimization efforts to be computationally feasible. This parameter ranking methodology will be demonstrated on the optimization of the thermal morphing anisogrid boom. The proposed novel morphing structure provides high precision morphing through the use of thermal strain as the sole actuation mechanism. The morphing concept uses the helical members in the anisogrid structure to provide complex constrained actuations that can achieve the six degree of freedom morphing capability. This structure provides a unique potential to develop an integrated structural morphing system, where the adaptive morphing capability is integrated directly into the primary structure. To identify parameters of interest, the Q-DEIM model reduction algorithm is implemented to rank the model parameters based on their impact on the morphing performance. This parameter ranking method provides insight into the system and enables the optimal allocation of computational and engineering resources to the most critical areas of the system for optimization. The methodology, in conjunction with a singular value decomposition (SVD), provides a ranking and identifies parameters of relative importance. The SVD is used to truncate the nine parameters problem at two locations, generating a five parameter optimization problem and a three parameter optimization problem. To evaluate the ranking, a parameter sweep in conjunction with a simple minimum cost function search algorithm will compare all 120 five parameter ranking orders to the Q-DEIM ranking. This reduced parameter set significantly reduces the parameter complexity and the computational cost of the model optimization. This paper will present the methodology to define the resulting performance of the optimal thermal morphing anisogrid structure, minimum morphing control, and the systems frequency response capability as a function of available power.