Identifiability of Inertia with Partial

In dynamical systems models from biology to power systems, structural identifiability 4 is an essential step for understanding, probing, and even controlling the dynamical system. When the 5 system input is controllable, structural identifiability can be verified by computing well-established, 6 straightforward conditions. In many applications, however, the system input is stochastic, and these 7 previous checks for structural identifiability do not generalize to this case. In this work, we focus 8 on the identifiability of the inertia matrix and establish necessary and sufficient conditions for its 9 identifiability with partial measurements and stochastic input. We then demonstrate and corroborate 10 our results with numerical experiments on the Föppl-Jeffcott rotor dynamical systems model. 11