(287a) Dynamic Data Reconciliation, Parameter Estimation, and Validation for the Dynamic Model of a Supercritical Pulverized Coal Power Plant

The need for process systems engineering tools to improve the performance of existing fossil-fueled power plants in the presence of renewables in the energy market is well documented [1]. A powerful, low risk, and cost-effective tool for the improvement of these existing systems is the development of highly predictive process models. In the past, many such models for coal-fired power plants have been developed [2], [3]. However, many of these models focus on the steady-state design and operation of traditional subcritical pulverized coal systems, with less consideration given to dynamic models of supercritical pulverized coal (SCPC) power plant. A few papers are focused on the impact of carbon capture on the SCPC plant [4]–[6] where simple models of the SCPC plant were used. One fairly comprehensive study has been done using the commercial software Apros, but the equipment models used lack some of the fidelity needed to capture the dynamics of the system [7]. For a general SCPC plant, a past contribution established a plantwide model with higher fidelity for some systems in the plant [8]. A particular gap in the literature is the lack of industrial validation for the models that have been developed.

In this presentation, a first-principles comprehensive dynamic model of a SCPC plant is described comprising detailed models of the following sections - boiler, turbine, feed water heaters, and the balance of the plant. Distributed models for each component of the boiler including the economizer, superheaters, water wall, and platen superheater are discussed. Each section of the boiler model also contains a detailed heat transfer model with due consideration of conductive/convective/radiative heat transfer as appropriate. The boiler model takes into account the changes in the heat transfer coefficients due to the hydrodynamics of the system and thermal properties of the steam that can vary rapidly during load-following especially under sliding pressure operation. The model of the steam turbine can capture the change in the performance characteristics of the leading, non-condensing, and condensing stages. In addition, the steam turbine model can detect and account for moisture, if present, in the stages that may be non-condensing under nominal condition.

The SCPC dynamic model is validated with the full-load and transient data from an industrial partner plant (IPP). The data does not satisfy material and energy balances motivating the solution of a dynamic data reconciliation problem to ensure closure of the balances [9], [10]. One difficulty in the dynamic data reconciliation for this system is the poor estimate of the holdup - especially thermal holdup in the boiler components - and the poor initial guess of the initial state of the system. It is also impractical to mimic the complex feed-forward-augmented feedback controllers and logic-based approaches used in the coordinated control system in the IPP. Furthermore, there are large number of model parameters, especially equipment specific parameters that are not available and must be estimated. Due to the large number of partial differential algebraic equations in the plant-wide model, the optimization problem for dynamic data reconciliation and parameter estimation is large. A two-stage optimization problem is solved for reconciling the dynamic data in the first stage followed by optimal estimation of the parameters. Uncertainties in the initial estimates of the material and thermal holdup are incorporated in the form of sensitivity information in the proposed approach for reducing the computational expense. This work shows that steady-state and transient performance of the IPP can be very well predicted by the model. The validated SCPC dynamic model is used for developing advanced model-based controllers for load-following operation. The controllers maximize the plant efficiency during transient operation while satisfying temperature constraints at various locations. Results are also shown for important calculated performance indices, such as the plant heat rate.

[1] A. Subramanian, T. Gundersen, and T. A. Adams II, “Modeling and Simulation of Energy Systems: A Review,” Processes, vol. 6, p. 238, 2018.

[2] E. Oko and M. Wang, “Dynamic modelling, validation and analysis of coal-fired subcritical power plant,” Fuel, vol. 135, pp. 292–300, Nov. 2014, doi: 10.1016/j.fuel.2014.06.055.

[3] C. Chen, Z. Zhou, and G. M. Bollas, “Dynamic modeling, simulation and optimization of a subcritical steam power plant. Part I: Plant model and regulatory control,” Energy Convers. Manag., vol. 145, pp. 324–334, Aug. 2017, doi: 10.1016/j.enconman.2017.04.078.

[4] A. K. Olaleye, E. Oko, M. Wang, and G. Kelsall, “Dynamic Modelling and Analysis of Supercritical Coal-Fired Power Plant Integrated with Post-combustion CO2 Capture,” in Clean Coal Technology and Sustainable Development, Singapore, 2016, pp. 359–363.

[5] K. Wellner, T. Marx-Schubach, and G. Schmitz, “Dynamic Behavior of Coal-Fired Power Plants with Postcombustion CO2 Capture,” Ind. Eng. Chem. Res., vol. 55, no. 46, pp. 12038–12045, Nov. 2016, doi: 10.1021/acs.iecr.6b02752.

[6] S. Ó. Garđarsdóttir, R. M. Montañés, F. Normann, L. O. Nord, and F. Johnsson, “Effects of CO2-Absorption Control Strategies on the Dynamic Performance of a Supercritical Pulverized-Coal-Fired Power Plant,” Ind. Eng. Chem. Res., vol. 56, no. 15, pp. 4415–4430, Apr. 2017, doi: 10.1021/acs.iecr.6b04928.

[7] J. Hentschel, H. Zindler, and H. Spliethoff, “Modelling and transient simulation of a supercritical coal-fired power plant: Dynamic response to extended secondary control power output,” Energy, vol. 137, pp. 927–940, Oct. 2017, doi: 10.1016/j.energy.2017.02.165.

[8] P. Sarda, E. Hedrick, K. Reynolds, D. Bhattacharyya, E. S. Zitney, and B. Omell, “Development of a Dynamic Model and Control System for Load-Following Studies of Supercritical Pulverized Coal Power Plants,” Processes, vol. 6, no. 11, 2018, doi: 10.3390/pr6110226.

[9] A. S. Chinen, J. C. Morgan, B. Omell, D. Bhattacharyya, and D. C. Miller, “Dynamic Data Reconciliation and Validation of a Dynamic Model for Solvent-Based CO2 Capture Using Pilot-Plant Data,” Ind. Eng. Chem. Res., vol. 58, no. 5, pp. 1978–1993, Feb. 2019, doi: 10.1021/acs.iecr.8b04489.

[10] P. Mobed, J. Maddala, R. Rengaswamy, D. Bhattacharyya, and R. Turton, “Data Reconciliation and Dynamic Modeling of a Sour Water Gas Shift Reactor,” Ind. Eng. Chem. Res., vol. 53, no. 51, pp. 19855–19869, Dec. 2014, doi: 10.1021/ie500739h.