Predicting the Optical Performance of the Space Interferometry Mission Using a Modeling, Testing, and Validation Methodology

Abstract

This paper presents the modeling, testing, and validation methodologies developed to predict the optical performance of the Space Interferometry Mission (SIM) at the Jet Propulsion Laboratory (JPL). The modeling methodology combines structural, optical, and control system design within a common state space framework and incorporates reaction wheel assembly (RWA) disturbances to evaluate the end-to-end performance of the system requirements. The validation methodology uses the Micro-Precision Interferometer (MPI) testbed, which is a ground-based, representative hardware model of SIM. In this study, the integrated model of the MPI testbed was used to calculate the transfer functions from RWA input to optical performance output. The model-predicted transfer functions were compared with the MPI testbed measurements, and the accuracy of the integrated model was quantified using a metric that was based on output power of the transfer functions. The RWA disturbances were then propagated through the modeled and measured transfer functions to predict the optical performance of the MPI testbed. This method is called the “decoupled disturbance analysis” and relies on the “blocked” RWA disturbances, measured with the RWA hardmounted to a rigid surface. These predictions were compared with the actual (measured) optical performance of MPI, measured with the RWA mounted to MPI, to evaluate the accuracy of the decoupled disturbance analysis method. The results show that this method is not an accurate representation of the coupled boundary conditions that occurs when the RWA is mounted to the flexible MPI structure. In order to correct for the blocked RWA disturbance boundary conditions, the “coupled disturbance analysis” method was developed. This method uses “force filters” that depend on estimates of the interface accelerances of the RWA and the MPI structure to effectively transform the blocked RWA disturbance measurements into their corresponding “coupled” disturbances (the disturbances that would occur at the coupled RWA-MPI interface). Compared to the decoupled method, the coupled method more accurately predicts the system’s performance. Additionally, the RWA cross-spectral density terms were found to be influential in matching the performance predictions to the measured optical performance of MPI.