Empirical Model Estimation and Model Comparison on a Temperature Pilot Plant

Model of a plant is used to get an insight of the physical system behavior and there always a scope exists to improve the model parameters by using proper estimation techniques. First principle model is more accurate if the physics of the plant is well known and less complex whereas empirical model estimation is used to estimate the model if the plant is complex and will give an accurate result over the operating region of estimation. In this work, first principle method is used to find the model of a temperature plant and a user interface is developed to provide flexibility to estimate the model for a range of physical parameter. The user interface provides an online view of the model and its response along with a future to simulate the model behavior. In the second half of the work, an empirical model is estimated using open loop experiment data. The models are compared and found to be closed matching. This work provides a simple user interface to estimate the model and to understand the model response with a wide variety of features like data storage, trend plot and with open loop and closed loop response analysis.