On September 19, we organized a scientific seminar on “A System Identification Package for Python: Actual Features and Future Developments “. The scientific seminar was led by Riccardo Bacci Di Capaci, an assistant professor at the University of Pisa, Italy. The scientific seminar was organized in the framework of the FrontSeat project as part of the seminar series on “Research Seminar on Smart Cybernetics”.
Systems identification methods gained new importance, especially for large multivariable processes, due to the widespread adoption of model-based control systems and MPC technologies. This talk discusses an established open-source System Identification Package for PYthon (SIPPY), which implements different methods to identify linear discrete-time multi-input multi-output systems. For input-output models, identification is performed by using least-squares regression (FIR and ARX models), iterative or recursive least-squares (ARMAX models), and optimization-based methods (BJ or more general formulations). For state space models, various subspace identification algorithms are implemented according to traditional methods (N4SID, MOESP, and CVA) and parsimonious methods that enforce causal projections. When the model order is not known a priori, three different information criteria can help the user in the choice of the most appropriate order. Data can be collected both in open-loop and closed-loop mode and then used for identification and validation purposes. Next efforts will be devoted to including more complex structures and approaches, such as zero-gain linear systems, nonlinear ARX models, and Artificial Neural Networks. Applications to industrial data from different fields show the effectiveness and computational efficiency of SIPPY.
This project has received funding from the European Union’s Horizon under grant no. 101079342 (Fostering Opportunities Towards Slovak Excellence in Advanced Control for Smart Industries).