Fit ratio is one of the most popular criteria that evaluates a result of system identification in the time domain. This criterion is constructed by Root Mean Squared Error (RMSE) divided by the standard deviation of the measured signal. However the validity of fit ratio have not been discussed actively in the field of system identification. In this paper, we introduce some discussion about criteria like fit ratio from the field of physical geography. Then, we evaluated these criteria through two case studies. As a result, we found that fit ratio was not adequate for evaluation of the result of system identification.
It is well-known that preprocessing of the measured input-output data is very important to obtain accurate parameter estimates by system identification methods. This paper treats LTI (linear time-invariant) systems with integrators that appear in various engineering fields, for example, a physical system with rigid body motion and DC servo motors. However, it is known that it is difficult to identify the systems with integrators because they are unstable and astatic system. The purpose of this paper is to propose an appropriate preprocessing method and an identification procedure for the systems with integrators. First, preprocessing of input-output data for such systems is proposed, which removes low-frequency disturbances. Second, a new identification procedure for such systems is proposed, which applies an integrator on input data. Finally, effectiveness of the proposed methods is examined through numerical examples.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.