Online Estimation and Identifiability of Cable Parameters Using Passive Monitoring

One method for diagnostic and prognostic evaluation of cable health is online estimation and tracking of physics-based parameters. In certain applications it is beneficial to passively monitor voltages and currents and subsequently estimate model parameters. However, passive monitoring requires a means to evaluate if and when the uncontrolled inputs are sufficient to excite the modeled dynamics. If the system dynamics cannot be observed the corresponding parameters cannot be estimated. In this paper, identifiability requirements of the observed input-output signals for standard models are investigated. Equivalent line circuit models are evaluated in forms allowing for recursive parameter estimation and with a minimum number of independent parameters.