(689d) Single and Multi-Objective Supervisory Model Predictive Control of Steam Temperature Cycles

In supervisory control applications, advanced controllers are generally used as the preceding layer for common controllers (i.e., in programmable logic controllers, distributed control systems, etc.). In a number of applications, supervisory controllers manipulate the behavior of underlaying feedback controllers by providing estimated setpoints based on pre-defined criteria. The type of setpoints and frequency of setpoint updates may vary depending on the process and desired performance.

In this work, a supervisory control system is designed for steam temperature control cycles. In particular, the supervisory layer consists of a model predictive control algorithm that includes an underlying model of the process. A parameter estimation technique is used to recursively estimate the open-loop ARX model parameters of the supervisory control loop as it operates at off-design conditions. The estimated ARX model is then provided to the model predictive control to update the controller design in order to deal with the changing states of the process. The supervisory control layer provides setpoints for proportional-integral-derivative controllers for the desuperheater outlet temperature and/or final steam temperature of a power plant steam generation process. This process consists of a drum where steam is evaporated, superheaters for increasing heat content of the generated steam, desuperheater for regulating the superheated steam temperature, and turbine-generator system for estimating the generated power¹.

The performance of the developed control scheme is evaluated considering two cases: (i) single objective control of final steam temperature; and (ii) multi-objective control of final steam temperature and desuperheater outlet steam temperature. Results for these cases, including validation of the proposed approaches with industrial and existing literature data as available, will be discussed.

Reference

¹Agbleze, S., Lima, F. V., Indrawan, N., Panday, R., Pezzini, P., Bonilla-Alvarado, H., Bryden, K. M., Tucker, D., & Shadle, L. J. (2020). Modeling and control of subcritical coal-fired power plant components for fault detection. ASME 2020 Power Conference. https://doi.org/10.1115/power2020-16571