Machine Learning-Based Approach to Determine the Deployment Stability of Bistable Composite Tape Springs in Coiling State

Composite tape springs are capable of self-deployment and can be fixed in place due to their bistability feature, allowing for stable storage and deployment. Thanks to these advantages, extensive research has been conducted on applying composite tape springs to space structures where storage efficiency is crucial. These applications include deployable antennas, solar arrays, and other space structures that require compact stowage during launch and reliable deployment in orbit. However, in certain cases, despite possessing bistability, some composite tape springs exhibit unstable deployment behavior known as the blossoming phenomenon. This phenomenon can lead to unstable deployment, potentially resulting in mission failure in space, making analysis and prevention essential to ensure mission success. In this paper, we investigate the blossoming phenomenon by analyzing the deployment behavior and strain energy history of composite tape springs with various geometric parameters through detailed deployment simulations. We also propose a method to predict deployment behavior and assess stability using machine learning techniques, which have the potential to significantly improve design efficiency. First, finite element modeling of composite tape springs with various geometric parameters, such as radius of curvature and subtended angle (see Figure 1, 2), was conducted at specific intervals to observe their deployment behavior. This approach allows for a comprehensive understanding of how geometric parameters can affect the deployment performance of composite tape spring. Subsequently, the bistability of each model was evaluated using a discriminant function, which helps in determining the likelihood of stable deployment. By performing coiling, holding, and deployment simulations, we compared and analyzed the deployment behavior and strain energy history based on the geometric parameters to establish criteria for stable and unstable behavior. The analysis revealed that the reduction trend of strain energy during the deployment process varies depending on the geometric parameters, and this variation critically influences the stability of the deployment behavior. (Figure presented). Additionally, we proposed a stability assessment method using advanced machine learning techniques, where the geometric parameters of the composite tape spring are input to predict the strain energy history and evaluate deployment stability with greater accuracy. The results of this study provide important insights for ensuring deployment stability in the design of composite tape springs and can be used as guidelines to achieve deployment stability according to specific design objectives. Furthermore, these findings have broader implications for the development of advanced space structures, potentially informing the design of next-generation deployable systems, and could pave the way for future research into more sophisticated deployment mechanisms in space exploration.