Natural gas combined cycle (NGCC) generates around 34% of the total electricity in the United States [1], while it also plays a key role in CO2 emission in the power market [2], which motivates its combination with carbon capture (CC) techniques. One of the challenges in such combinations with traditional CC techniques is the volatile electricity demands caused by the increasing penetration level of the renewable electricity into the electricity supply mix [3]. Although the NGCC system can respond to the demand changes easily, traditional CC techniques, such as amine scrubbing, are usually less flexible and may perform poorly in changing operation conditions [4]. Therefore, there is a critical need to incorporate the NGCC systems with novel carbon capture techniques to guarantee that the NGCC power is generated in a carbon-neutral (or even carbon-negative) manner even in the volatile energy market.

The GaTech FLECCS team has proposed a retrofit design to existing NGCCs to allow them to have an overall negative CO2 emission while maintaining positive power generation by utilizing both post-combustion carbon capture (PCC) and direct air capture (DAC) techniques. The PCC unit applies traditional amine scrubbing and is used to remove 97% CO2 contained in the flue gas. The DAC unit is supported by advanced separation techniques [5] and is used to directly capture CO2 from the atmosphere, enabling overall negative CO2 emission. The DAC unit consumes power generated from the NGCC for adsorption and low-pressure (LP) steam from the heat recovery steam generator (HRSG) for sorbent regeneration. In addition, a modular design is applied to the DAC unit that allows it to operate in a high turn-down ratio. This gives extra flexibility to the system: when the power demand is low, it can consume more energy and LP steam to capture more CO2; when the demand is high, it can capture less CO2 and send more power to the grid. However, the flexibility also introduces extra dynamics that make the overall modeling and optimization problems more challenging.

In this work, we model the NGCC-PCC-DAC retrofit systematically and co-optimize the design and the operations of the system under different electricity price signals. Specifically, the operation of the modular DAC unit is modeled with two decisions (sorbents allocated for adsorption/regeneration) bounded by sorbent conservation over finely discretized periods. This allows the system to decouple the DAC swing cycle and flexibly choose when to consume power and steam for DAC CO2 capture to maximize the total profit with the rapidly changing power demands. Moreover, a start-up procedure is also modeled to accommodate extreme scenarios when the electricity price is zero for long periods and it is necessary to shut the system down to reduce cost, which introduces binary variables. The design decision in the optimization problem is the size of the DAC system, which is composed of three parts (the sorbent amount, the size of the adsorption system, and the size of the desorption system). The operation decisions are the hourly behaviors of each unit, while the hourly electricity price is provided for a whole year.

The corresponding optimization problem is a large-scale MILP problem and can be directly solved within a reasonable length of time. It is solved multiple times with different energy scenarios and different CO2 credits to reflect the system behavior under various energy conditions in the future. The solutions show that a large DAC unit is preferred in most situations with a medium CO2 price. In all scenarios, the retrofitted plant has longer dispatch time than the original NGCC and is more profitable. The modular DAC allows the system to switch quickly among several operating modes to deal with changes in electricity price: (1) when the price is high, the system tends to run NGCC at 100% and DAC at its minimum capacity to earn more money from selling power; (2) when the price drops, the DAC will be run at its full capacity to help the system remain profitable by capturing more CO2; (3) at very low electricity prices, the NGCC load can be dropped to its minimum stable level with the DAC at its maximum capacity, which can still be profitable with a mild CO2 price. In conclusion, this work reveals the feasibility of combining DAC with existing NGCCs to enable carbon-negative power generation, the advantages of the modular DAC design in response to varying power demands, and optimal solutions for the design and operations with different electricity price signals.

References

[1] U.S. Energy Information Administration, "Electric Power Annual 2020," 29 10 2021. [Online]. Available: https://www.eia.gov/electricity/annual/.

[2] U.S. Energy Information Administration, "U.S. Energy-Related Carbon Dioxide Emissions, 2020," 2021.

[3] S. A.-A. R. L. Juan Pablo Ríos-Ocampo, "Renewable energy penetration and energy security in electricity markets," International Journal of Energy Research, vol. 45, no. 12, pp. 17767-17783, 2021.

[4] M. W. P. L. J. S. Y. L. Xiao Wu, "Solvent-based post-combustion CO2 capture for power plants: A critical review and perspective on dynamic modelling, system identification, process control and flexible operation," Applied Energy, p. 113941, 2020.

[5] S. H. P. G. Z. C. W. J. a. R. P. L. Achintya R. Sujan, "Direct CO2 Capture from Air using Poly(ethylenimine)-Loaded Polymer/Silica Fiber Sorbents," ACS Sustainable Chemistry & Engineering, vol. 7, no. 5, p. 5264–5273, 2019.