An Observer-based Approach to Modeling Error Identification

Information about the location of modeling errors is crucial for the efficient improvement of an invalid model. This article discusses how to pinpoint modeling errors through the comparison of experimental data with data obtained through simulation of the invalid model. An observer-based approach is developed. By designing a dedicated observer for the system using the invalid model, a signal vector is generated, on which each modeling error imposes an easily identifiable feature. An algorithm to analyze the featured signal is then developed. With this algorithm, the features of each of the modeling errors are extracted. To illustrate and verify the developed approach, a simulation study of a biological model is performed and promising results are obtained.