Plantwide real time optimization, control and estimation of a solar thermal power plant
This work will deal with plantwide simulation studies for real time optimization, control and estimation using a dynamic model of a solar thermal power plant. The model was developed by an earlier PhD student and mimicks an actual 1 MW hybrid solar thermal power plant designed, commissioned and installed by a team from IIT Bombay.
The plant is challenging to operate optimally as it has several control loops, only a few measurements, is affected by large magnitude disturbances, and is inherently dynamic in nature. The current work will utilize the developed dynamic model to demonstrate how set points for various control loops could be optimally computed, and how the interaction between various loops could be managed in a multivariate control setting (model predictive control) to maximize economic performance. The plantwide supervisory control layer will also be coupled with plantwide estimator which will estimate key variables/parameters in the plant and also identify faults. This information would then be utilized by the supervisory control layer.
Scope of the work: Computational/theoretical.
Pre-requisites: A keen interest in systems and control theory, and ability to do heavy duty programming as well as get deep into optimization and estimation ideas.