Controller tuning plays an important role when designing or adjusting a control system. Using a model for that purpose is more convenient than tuning the controller on a real system, especially for the case of remote assistance. It can be performed on model using various methods, therefore, it is necessary to evaluate the performance of the model using available methods for the case study. Flow control is one of the most common processes on board a ship, therefore it is worth evaluating the performances of the model for the ship digital twin applications. The aim of this paper is to evaluate the previously developed transfer function model using common well-known tuning methods, in order to get a clear insight in the model applicability for the purpose of controller tuning and to determine the optimal method for that purpose. By analyzing the response of the system with different controller parameters, it is possible to make a conclusion about the performance of the model, as well as the conclusion about the tuning methods. The results of the evaluation are compared both in Matlab and in a real flow control process on a GUNT 674 flow control unit.

System identification plays an important role in system control and controller tuning. Data-driven identification provides a behavioristic model of an identified system or process that can be useful for remote controller tuning by simulating the response of an identified system or process. The identification can be performed using various methods, therefore, it is necessary to examine available methods for the case study. The aim of this paper is to describe and perform the identification procedure for the case study of flow control process using the Matlab's System Identification Toolbox. As a result, a continuous-time transfer function model is formed, which is to be used for PID controller tuning and a digital twin development purposes in further work.