Discrete-time modeling and identification of continuous-time systems: a general framework

摘要: A general framework for modeling of a time-varying continuous-time SISO system from its sampled input and output while retaining the parameters with their physical interpretation is presented. The theory can be specialized to Poisson moment functional approach, integrated sampling instantaneous approach or use state variable filters. In all methods, initial conditions removed by applying an appropriate discrete-time operator. Digital filtering used explicitly implicitly reconstruct time-derivatives signals involved. thorough study approximations resulting converting model version It shown how these errors controlled that parameter estimates obtained arbitrary accuracy. digital exhibit optimal properties among methods fit in framework. Moreover, using filter designs instead numerical integration allows slower sampling. maximum likelihood estimator derived time-invariant systems errors-in-variables stochastic Finally, verified through simulations.