Carbon–Electricity–Heat Coupling Process for Full Unit Carbon Capture: A 1000 MW Case in China

Abstract

Carbon capture is pivotal for achieving carbon neutrality; however, its high energy consumption severely limits the operational flexibility of power plants and remains a key challenge. This study, targeting a full flue gas carbon capture scenario for a 1000 MW coal-fired power plant, identified the dual-element (“steam” and “power generation”) coupling convergence mechanism. Based on this mechanism, a comprehensive set of mathematical model equations for the “carbon–electricity–heat” coupling process is established. This model quantifies the dynamic relationship between key operational parameters (such as unit load, capture rate, and thermal consumption level) and system performance metrics (such as power output and specific power penalty). To address the challenge of flexible operation, this paper further proposes two innovative coupled modes: steam thermal storage and chemical solvent storage. Model-based quantitative analysis indicated the following: (1) The power generation impact rate under full THA conditions (25.7%) is lower than that under 30% THA conditions (27.7%), with the specific power penalty for carbon capture decreasing from 420.7 kW·h/tCO₂ to 366.7 kW·h/tCO₂. (2) Thermal consumption levels of the capture system are a critical influencing factor; each 0.1 GJ/tCO₂ increase in thermal consumption leads to an approximate 2.83% rise in unit electricity consumption. (3) Steam thermal storage mode effectively reduces peak-period capture energy consumption, while the chemical solvent storage mode almost fully eliminates the impact on peak power generation and provides optimal deep peak-shaving capability and operational safety. Furthermore, these modeling results provide a basis for decision-making in plant operations.

1. Introduction

The global energy system is under profound pressure to achieve a low-carbon transition, with carbon emissions from coal-fired power plants representing a particularly critical challenge. Although improving energy efficiency and developing renewable energy are considered fundamental pathways for climate change mitigation, the full flue gas deployment of these solutions requires time. In this context, Carbon Capture, Utilization, and Storage (CCUS), particularly full flue gas carbon capture technology aimed at capturing over 90% of CO₂, has emerged as a crucial practical solution for achieving near-zero emissions from fossil fuels and ensuring a secure energy transition [1,2,3]. In China—which is the largest developing country with a coal-dominated power infrastructure committed to achieving its dual carbon goals—advancing full flue gas carbon capture technology not only aligns perfectly with its carbon neutrality objectives [4,5,6] but also mitigates electricity supply risks that could result from an overly abrupt phase-out of coal, underscoring its strategic significance.

Among various CCUS technological pathways, post-combustion capture based on chemical absorption demonstrates the highest technological maturity and represents a foundation for industrial applications [7,8,9,10]. Its core mechanism relies on the reversible absorption–desorption reaction of solvents to achieve CO₂ separation [11]. However, the primary challenge lies in its high energy consumption, with the solvent regeneration process accounting for more than 50% of the total capture energy [12]. This typically requires substantial steam (100–150 °C, 2.1–3.4 bar) to be extracted from steam turbines [13]. For instance, in a 220 MW unit, saturated steam consumption can reach 115 t/h [14], while auxiliary power consumption (e.g., for solvent pumps and compression equipment) is approximately 120 kW·h/tCO₂ [15], resulting in an approximately 10% reduction in net plant efficiency [16]. Although solvent improvements and thermal integration may potentially reduce regeneration energy below 2 GJ/tCO₂ [17], the high energy intensity directly constrains the power plant’s core competitiveness—operational flexibility, particularly its deep peak-shaving capability. This represents a critical bottleneck that requires urgent resolution.

Diversion steam from a steam turbine is the most common integration scheme for meeting the steam demand of a carbon capture system [18,19]. However, the selection of the extraction point is highly critical [20,21], with the crossover pipe between the intermediate-pressure (IP) and low-pressure (LP) cylinders usually considered a viable location [22]. Some studies have also explored alternative methods, such as installing a back-pressure turbine, to obtain the necessary steam [23]. Nevertheless, for new or retrofitted power plants, the available extraction flow is often insufficient to satisfy the substantial demand under full flue gas capture scenarios [24]. To enhance flexibility, a flexible operation strategy that decouples the absorption and desorption processes was proposed [25,26]. This approach involves storing rich solvent during off-peak hours and regenerating it during peak hours; theoretical studies have demonstrated its potential to improve the unit’s peak-shaving capability [27,28]. Furthermore, the integration of auxiliary energy sources, such as heat pumps and solar thermal energy, has been investigated to improve the overall system efficiency [29,30].

Although previous research has touched upon the interactions between carbon capture and thermal systems, a survey of the existing literature has revealed distinct knowledge gaps. First, most studies have focused on isolated processes or static analyses, lacking a holistic, dynamic mathematical model that captures the full process coupling of carbon, power, and heat. This limitation prevents systemic understanding of the underlying interaction mechanisms. Second, in the extreme scenario of full flue gas carbon capture in 1000 MW units, the profound impacts on operational security and peak-shaving capability, as well as the corresponding optimization pathways, have not been sufficiently elucidated or quantified. Against this backdrop, based on an engineering feasibility study for full flue gas capture at Unit #3 of the China Taizhou Power Plant, this research identifies a dual-element coupling convergence mechanism involving “steam” and “power generation” between carbon capture and the power generation system. This provides a novel perspective for understanding the systemic dynamic balance. Furthermore, a complete set of mathematical model equations was developed to enable a quantitative analysis oriented towards engineering applications. Consequently, innovative coupling modes are proposed by introducing “steam thermal storage” and “chemical solvent storage,” which effectively address peak-shaving challenges. By evaluating different coupling modes and key operational parameters, the framework can provide some supports for dispatch optimization, economic assessment, and definition of safe operating boundaries. These results would enhance the applicability of carbon capture systems in large-scale power plants and support more secure and flexible low-carbon operation. Collectively, this work provides new pathways for resolving security concerns and operational decision-making associated with full flue gas carbon capture in large-scale units.

2. Materials and Methods

2.1. Mechanism Characterization

Using the frameworks of “steam flow rate convergence” and “electrical power output convergence”. Figure 1 illustrates the complex interdependencies between carbon capture and power generation, encompassing energy exchange, load regulation, and self-sustaining operation. This “dual-element” coupling system, based on the two key elements of “steam” and “power”, integrates carbon capture with the power generation process. The core logic can be summarized as follows: carbon capture requires steam extracted from the turbine, which consequently lowers the power generation efficiency. Conversely, power load fluctuations alter the flue gas volume, which, in turn, affects the steam required for capture. This interdependence forms a nonlinear feedback loop between the carbon, power, and heat systems. The “dual-element” coupling convergence mechanism is characterized by (1) bidirectional energy constraints, arising from the inherent competition and synergy in energy flows between the carbon capture and power generation systems; (2) nonlinear dynamic properties, manifested as a nonlinear relationship between steam/power demand and unit load; and (3) a unique convergence mechanism by which system equilibrium is achieved by dynamically balancing steam flow and electrical power output.

2.2. Logic Flowchart

Figure 2 shows the control logic flowchart for the coupled power generation and carbon capture system. The process is driven by the requirement of supplying electricity to the grid. If this demand is not satisfied, the unit load is adjusted, changing the main steam flow rate. This change modifies the flue gas volume, leading to variations in the total amount of CO₂ entering the carbon capture system. The CO₂ capture volume is governed by the actual capture rate, while the steam and electricity demands of the system are influenced by the process technology level, which is characterized by the thermal consumption level (GJ/tCO₂). In self-sustaining mode, all utilities must be sourced from the thermal system, which directly or indirectly impairs the net power output. An iterative computational approach is essential for satisfying these constraints and achieving a convergent solution. The regeneration process of carbon capture systems necessitates a supply of low-grade steam. Although the IP cylinder exhaust is commonly used, it is prone to deficits in both volume and pressure parameters under full flue gas capture conditions, especially at low unit loads. To address this issue, extraction from the cold section of the double reheat system has emerged as a more reliable solution, as successfully demonstrated at the 1000 MW coal-fired unit of China Taizhou Power Plant. The direct coupling of the power generation unit with the carbon capture system can compromise grid stability. To mitigate this issue, two alternative coupling modes were explored: steam thermal storage (involving high-grade steam accumulation) and chemical solvent storage (involving CO₂-enriched solvent retention).

Figure 3 illustrates the fundamental configuration of the integrated thermal and chemical storage system. The steam accumulator is located on a bypass line of the main steam pipeline that delivers regeneration steam from the thermal power unit. It functions as a buffer by alternately absorbing and discharging steam, thereby addressing the steam supply shortfall encountered during full flue gas carbon capture operation at high unit loads. The solvent storage system is composed of lean and rich solvent tanks. Under conditions of inadequate steam extraction from the thermal system, the operation of the absorber is sustained, whereas regeneration is halted. The CO₂-rich solvent is stored, constituting a method of chemical energy storage, thereby meeting the carbon capture requirements for the entire unit. Under the conditions of sufficient steam extraction capacity from the thermal system, the steam supply can be increased to regenerate the excess stored rich solvent. Therefore, a lean solvent tank of an equivalent volume is necessary to maintain the mass balance of the absorption solvent circulation.

2.3. Mathematical Equations

(1)

We develop the mathematical model based on the aforementioned mechanisms and their logical relationships. The relationship between the main steam mass flow rate ($F$, t/h) and unit power output ($P$, MW) is as follows:

$$F=f_1(P)$$

(1)

(2)

The relationship between the captured CO₂ flow rate ($Q$, tCO₂/h), $P$, and carbon capture ratio (%) is as follows:

$$Q=f_2(P,k1-\beta )$$

(2)

The solvent storage impact rate, denoted by $\beta$ (%), is applicable only in chemical solvent storage mode. It is defined as the rate that characterizes the influence of solvent storage on the net CO₂ captured during regeneration. A positive value of $\beta$ indicates a net inflow (inflow > outflow) in the rich solvent tank, and a negative value indicates a net outflow (outflow > inflow).

(3)

Steam allocation balances extraction for power generation with consumption for carbon capture:

$$F\times \left(k3+\theta \times k4+\alpha \right)=\mu \times Q\times k2$$

(3)

The parameters in the model are defined as follows: $k2$ (GJ/tCO₂) denotes the specific thermal consumption level, reflecting the heat demand per unit of CO₂ captured. $k3$ (%) and $k4$ (%) quantify the extraction rates of the intermediate-pressure cylinder exhaust steam and secondary reheat cold-section steam, respectively. $\mu$ is the heat-to-steam conversion coefficient, representing the heat energy required per ton of steam generated for the capture process. $\theta$ is the volumetric ratio of the steam extracted from the secondary reheat cold section after being desuperheated and depressurized to the target parameters. The parameter $\alpha$ is the steam flow rate for the thermal storage system, defined as a percentage of the unit’s main steam flow. It is only applicable in storage mode, with $\alpha$ < 0 denoting absorption and $\alpha$ > 0 denoting discharge.

(4)

The dependence of $P1$ (post-extraction power output in MW) on $P$, $k3$, and $k4$ is as follows:

$$P1=f_3(P,k3,k4)$$

(4)

(5)

The electric demand $P2$ (MW) of the carbon capture system as a function of $Q$ and $k2$ is as follows:

$$P2=f_4(Q,k2)$$

(5)

(6)

The unit power output $P^{\prime }$ (MW) is a function of $P1$ and $P2$ as follows:

$$P^{\prime }=P1-P2$$

(6)

(7)

Determination of the usable capacity $V1$ (m³) of the thermal storage tank is as follows:

$$V1={\displaystyle \int \frac{F\times \alpha }{g1}}dt$$

(7)

The variable $dt$ (h) is the time interval, while $F\times \alpha \times dt$ (t) is the thermal storage capacity. The thermal energy storage per unit volume of saturated water $g1$ (t/m³) is defined as the mass of steam generated when a unit volume of saturated water is depressurized from the inlet steam pressure to the exhaust pressure. This parameter represents the volumetric energy storage capacity of a steam accumulator. It is evident that the thermal storage volume is influenced by numerous factors. The total required thermal storage capacity highly depends on the actual operating conditions of the unit, particularly its involvement in peak-shaving, which introduces considerable uncertainty. Therefore, a case-specific analysis is necessary. However, to minimize the required thermal storage volume, the influencing factor $g1$ must assume a larger value. Physically, this necessitates a greater pressure differential between the steam admission and exhaust parameters, indicating that higher-pressure steam should be supplied to the thermal storage unit.

(8)

The usable capacity $V2$ (m³) of the solvent storage tank is determined as follows:

$$V2={\displaystyle \int \frac{Q\times \beta }{g2}}dt$$

(8)

The variable $dt$ (h) is the time interval, while $Q\times \beta \times dt$ (tCO₂) is the CO₂ storage capacity. $g2$ (tCO₂/m³) is the CO₂ storage capacity of the rich solvent, defined as the mass of CO₂ absorbed from the absorber per unit volume of rich solvent. This parameter reflects the loading capacity of the solvent. It is evident that the solvent storage volume is influenced by multiple factors. The total required solvent storage capacity is also subject to the actual operational conditions of the unit (particularly its participation in peak-shaving), introducing similar uncertainties. However, to reduce the required solvent storage volume, the influencing factor $g2$ must reach a larger value. The underlying mechanism is that lean solvent with a higher $g2$ value can absorb more CO₂, resulting in rich solvent with a greater loading capacity. From this perspective, the CO₂ loading capacity of the lean solvent is a critical factor.

3. Results and Discussion

3.1. Engineering Model Equations

3.1.1. $F$ as a Function of $P$

To establish the relationship between the main steam flow rate and unit power output accurately, a model was developed using the thermodynamic simulation software EBSILON Professional V13.02. Using the 1000 MW unit as the object, the simulation results in

Table 1. Comparison between simulation and design value under different conditions.

Parameter	Load	Design Value	Simulation Value	Relative Error
Main steam temperature (°C)	100%THA	600	600	0.00%
	90%THA	600	600	0.00%
	75%THA	600	600	0.00%
	50%THA	600	600	0.00%
	40%THA	600	600	0.00%
	30%THA	600	600	0.00%
Reheat steam mass flow (kg/s)	100%THA	643.95	645.00	0.16%
	90%THA	568.63	575.35	1.17%
	75%THA	467.48	473.52	1.28%
	50%THA	309.01	311.01	0.64%
	40%THA	249.42	248.65	0.31%
	30%THA	190.77	186.30	2.40%
Reheat steam temperature (°C)	100%THA	600	600	0.00%
	90%THA	600	600	0.00%
	75%THA	600	600	0.00%
	50%THA	600	600	0.00%
	40%THA	600	600	0.00%
	30%THA	600	600	0.00%
Power output (MPa)	100%THA	1001.51	1000	0.00%
	90%THA	913.01	900	0.15%
	75%THA	758.94	750	1.19%
	50%THA	505.22	500	1.04%
	40%THA	406.82	400	1.71%
	30%THA	311.53	300	3.84%

show a typical deviation of less than 5% from the design values, ensuring the reliability of the model. In addition, multiple THA (Turbine Heat Acceptance) operating conditions were simulated using the EBSILON Professional software, with a primary focus on the linear influence of the main steam flow rate on power generation while downplaying the nonlinear effects of the turbine. Figure 4 shows a typical linear relationship between these two parameters. To simplify the model for practical applications, the final fitted equation for the main steam flow rate is $F=2.62\times P-100$, with the coefficient of determination R² approaching 1.

3.1.2. $Q$ as a Function of $P$ and $k1$

The substantial flue gas output from the 1000 MW unit necessitates two identical carbon capture systems to avoid oversized equipment, as determined by the engineering feasibility study. Figure 5 shows that flue gas flow, CO₂ concentration, and CO₂ flow rate all increase with rising unit generation. However, the relationships of flue gas flow versus unit load and CO₂ concentration versus unit load are not perfectly linear, exhibiting slight upward and downward deviations, respectively. As unit load increases, flue gas flow rises while CO₂ concentration changes only slightly. This occurs because load variations alter the excess air coefficient and combustion efficiency. At low loads, increased excess air dilutes CO₂, while at high loads, improved combustion efficiency offsets this dilution effect. Consequently, both the CO₂ mass flow rate and the captured CO₂ flow rate exhibit near-linear growth with unit load, as shown in Figure 5. The combined effect results in an approximately linear relationship between CO₂ flow rate and unit load (corresponding to THA operating conditions). Given the conversion relationship (the captured CO₂ flow rate $Q$ is derived from flue gas volume multiplied by CO₂ concentration), the final fitted equation for CO₂ capture is $Q=2\times k1\times (0.4P-10.3)$.

3.1.3. $P1$ as a Function of $P$, $k3$, and $k4$

To facilitate engineering applications, the impact of a single steam extraction on unit power generation is simplified to a linear relationship. This relationship, which holds except during ultra-low-load operation or rapid load changes, is consistent with the thermal energy conservation and turbine characteristics: a linear increase in extraction steam flow causes a linear decrease in main steam flow, leading to a linear reduction in generation capacity. Additionally, interaction terms are incorporated into the mathematical model to account for the nonlinear coupling effects when both the intermediate pressure exhaust extraction and secondary reheat cold section extraction operate simultaneously. The model structure and coefficient definitions are summarized in

Table 2. Mathematical relationship definitions: steam extraction impact on power generation.

Mathematical Relationship	P1 = P × (1 – A × k3 – B × k4 – C × k3 × k4)
P1, P	unit output after steam extraction (MW), unit power output (MW)
A	linear loss coefficient induced by intermediate-pressure exhaust steam extraction; A = 0 as low load condition
B	linear loss coefficient induced by secondary reheat cold section exhaust steam extraction
C	steam extraction interaction term coefficient; C = 0 as low load condition
k3	IP cylinder exhaust steam extraction ratio (%); (calculated as a decimal value)
k4	secondary reheat cold section steam extraction ratio (%); (calculated as a decimal value)

.

The impact of steam extraction on output power was quantified in an EBSILON Professional model with two extraction points (IP cylinder exhaust and secondary reheater cold section), using the extraction rates $k3$ and $k4$ as input variables. The fitting quality was evaluated using the coefficient of determination (R²), and an iterative optimization based on the gradient descent method was employed in the Origin software V10.5.36. The final model coefficients (

Table 3. Model results: steam extraction impact on power generation.

Mathematical Relationship	P1 = P × (1 − 0.53 × k3 − 0.61 × k4 − 0.1 × k3 × k4)
determination coefficient at THA	R2 = 0.982
determination coefficient at 90% THA	R2 = 0.983
determination coefficient at 80% THA	R2 = 0.989
determination coefficient at 70% THA	R2 = 0.993
determination coefficient at 60% THA and lower	R2 = 0.999

) demonstrate excellent performance (R² > 0.98) over a range of THA operating conditions. The fact that value A (0.53) is smaller than value B (0.61) aligns with fundamental thermodynamic principles. This is because the steam extracted from the cold section of the secondary reheat system possesses higher parameters and greater exergy, thereby exerting a more significant influence on power generation. The value of C = 0.1 indicates that the coupling interaction between the two extraction streams is non-negligible when they coexist, demonstrating the nonlinear effects of steam extraction on unit power generation. The mathematical model exhibited high predictive accuracy across all THA operating conditions (in Figure 6 and Figure 7, balls denote $P1$ and cyan area represents model surface). The close adherence of data points to the model surface, along with the minimal randomly distributed residuals and near-unity R² values, validate the precision of the model. The high consistency between the surface morphology and data distribution validates the capability of the model to effectively capture complex characteristics and maintain robust generalization. Under conditions of 70% THA and above, the mathematical model surface exhibited a slightly curved form, while at 60% and below, it exhibited a typically linear and planar state.

3.1.4. $P2$ as a Function of $Q$ and $k2$

The analysis of equipment energy consumption, conducted using Aspen Plus, is presented in Figure 8, which shows the variation in the unit electricity consumption with the captured CO₂ flow rate at different specific thermal consumption levels. It can be observed that, across different specific thermal consumption levels, the unit specific power penalty exhibits a consistent decreasing trend as the captured CO₂ flow rate increases. This indicates that the process technology level, represented by the specific thermal consumption, does not alter the fundamental electricity consumption trend. However, a lower specific thermal consumption is accompanied by a reduced unit-specific power penalty. For instance, at a specific thermal consumption level of 2.7 GJ/tCO₂, the unit specific power penalty under THA conditions is 231 kW·h/tCO₂, whereas at 1.8 GJ/tCO₂, it decreases to 215 kW·h/tCO₂. This observation can be attributed to the underlying principle of amine-based carbon capture. For a given technology level, a lower thermal consumption level is indicative of a superior solvent, lower solvent circulation rate, and less parasitic heat loss. Consequently, this leads to reduced shaft power requirements for auxiliary equipment and lower heating/cooling duties.

Similarly, to facilitate engineering applications and analysis, modeling and analysis were conducted on the unit-specific power penalty relative to the captured CO₂ flow rate under different thermal consumption levels. The results from the nonlinear $Pow2P2$ model ($M=a\times {(1+Q)}^b$) in

Table 4. Nonlinear modeling parameters.

Thermal Consumption Level GJ/tCO2	M = a × (1 + Q)b
	Coefficient a	Coefficient b	Determination Coefficient R2
1.8	627.1	0.183	0.976
2.1	651.0	0.184	0.979
2.4	676.9	0.187	0.971
2.7	696.4	0.188	0.969

indicate remarkable parameter stability: coefficient b ≈ 0.185 and goodness-of-fit R² ≈ 0.97 across all conditions. This parameter invariance suggests that the unit specific power penalty follows consistent nonlinear patterns regardless of the thermal consumption level, thereby revealing the fundamental nonlinear physical characteristics of the carbon capture system’s power demand. Assuming that coefficient b is constant, a polynomial relationship can be established between coefficient a and the thermal consumption level $k2$, given by $a=427+133\times k2-12.2\times k{2}^2$. This yields the final expression for the power demand $P2$ of the carbon capture system as a function of the captured CO₂ flow rate $Q$ and thermal consumption level $k2$: $P2=Q\times (427+133\times k2-12.2\times k{2}^2)\times {(1+Q)}^{0.185}$.

3.2. Direct Coupling Results of Thermal Power Unit and Carbon Capture System

3.2.1. Load-Dependent Performance of Carbon Capture on Power Output

Figure 9 shows the effect of the carbon capture system on the power generation of the thermal system under different operating conditions, considering a high carbon capture ratio of 90%. The unit power output varies synchronously with the load conditions. However, the impact of the carbon capture system on power generation does not follow this synchronous variation. A higher load corresponds to a lower power generation impact rate, indicating higher thermal system efficiency near the THA condition. In contrast, under low-load conditions, the generation impact rate increases significantly—rising from 25.7% at THA to 27.7% at 30% THA, an increase of 2%. However, the effects under both 50% and 75% THA conditions are nearly identical. This similarity is attributed to the insufficient pressure/temperature of the IP cylinder exhaust steam under low-load conditions, where only the cold section extraction of the secondary reheat steam is considered. At the same time, although the per-unit steam extraction from the secondary reheat cold section has a greater impact on power generation than the per-unit extraction of the IP cylinder exhaust steam, the higher initial parameters of the secondary reheat cold section steam allow it to provide more steam after desuperheating and pressure reduction. Consequently, the overall impacts of the two extraction methods on unit power generation are comparable. Further analysis indicated that neglecting the IP cylinder exhaust steam during low-load operation, while meeting high steam demands for carbon capture, risks excessive extraction from the secondary reheat cold section. This may induce detrimental effects on the turbine unit, including localized overheating and vibrations, which are fundamentally associated with the design specifications of the turbine. In this case study, the extraction rate from the secondary reheat cold section already reached 16% under low-load conditions. It is predicted that with increasing thermal consumption levels, the extraction flow rate will be greater, which will affect the maximum stable carbon capture ratio of the unit at low loads. Furthermore, due to the carbon capture system, the impact on power generation per ton of CO₂ captured decreases as the load increases. The specific power penalty reaches a maximum of 420.7 kW·h/tCO₂ at low load (30% THA) and a minimum of 366.7 kW·h/tCO₂ at high load (THA). This further demonstrates the high power penalty characteristics of the carbon capture system. From the perspective of reducing the specific power penalty, maintaining high-load operation is advantageous for full flue gas carbon capture units.

3.2.2. Mechanism of Carbon Capture Ratio Affecting Power Generation Performance

Figure 10 shows the effect of the carbon capture system on the thermal power unit’s generation under different capture rate conditions. As the carbon capture ratio decreased, both the thermal and power consumptions of the system decreased, with a more pronounced effect on the unit power output. As the carbon capture ratio decreased, the unit power output exhibited an almost linear increase. Specifically, the power output was only 743 MW at a 90% capture rate but approached 900 MW when the capture rate dropped to 30%. It can be observed that when the grid imposes high peak-shaving requirements on the power plant, the absence of an “energy storage” mode necessitates a reduction in the carbon capture ratio to prevent the capture system from affecting the peak power supply. As the carbon capture ratio increases, the impact on specific power penalty gradually decreases, dropping from 431.1 kW·h/tCO₂ at 30% capture rate to 366.7 kW·h/tCO₂ at 90% capture rate, a reduction of 64.4 kW·h/tCO₂ (14.94%). This indicates that to minimize specific power loss, the carbon capture system must be maintained at high capture rates.

3.2.3. Impact of Carbon Capture on Power Generation Under Different Thermal Consumption Levels

Figure 11 shows the variation in the unit power output with thermal consumption levels under high-load (THA) and low-load (30% THA) conditions at a 90% carbon capture ratio. The thermal consumption level represents the process technology level of the carbon capture system and directly determines its thermal and electrical energy requirements. As shown by the data, increased thermal consumption levels significantly diminish the power output capacity under all load conditions. At high unit load, the power output drops from 743 MW with low thermal consumption level (1.8 GJ/tCO₂) to only 677 MW with high thermal consumption level (2.7 GJ/tCO₂), representing an 8.9% reduction in generation capacity despite a 50% increase in thermal consumption value. Although the presence of the carbon capture system also reduces the power supply capacity of the unit under low-load conditions, it enhances the deep peak-shaving capability of the plant. Due to the carbon capture system, the unit power output is regulated to 217 MW under low thermal consumption level (1.8 GJ/tCO₂), while the deep peak-shaving capacity reaches 196 MW (<200 MW) under high thermal consumption level (2.7 GJ/tCO₂). This represents a further reduction of approximately one-third compared to the original low peak level of 300 MW. Figure 12 shows the variation in power loss per ton of CO₂ with thermal consumption levels under high-load (THA) and low-load (30% THA) operating conditions. The results demonstrate that high-load operation is advantageous in terms of specific power loss, regardless of the thermal consumption level. Comparing the 30% THA and THA cases, the average reduction reached 12.69%. When the thermal consumption level increased from 1.8 to 2.7 GJ/tCO₂, the specific power penalty increased by approximately 25.5% across different load levels. This corresponds to a coefficient where each 0.1 GJ/tCO₂ increase in thermal consumption level raises the specific power penalty by approximately 2.83%.

3.3. Comparative Analysis of Coupling Modes for Carbon Capture Integration

Figure 13 presents the impact of different coupling modes between the thermal unit and carbon capture system on the unit power output. The analysis is based on a conventional 1000 MW thermal unit, with a conventional net power output range of 300–1000 MW. The integration of a full flue gas carbon capture system reduces the unit’s maximum exportable power due to the parasitic load of the carbon capture equipment and steam consumption required for its operation. This total reduction is quantified as the composite power consumption of the capture system. When a full flue gas carbon capture unit incorporates thermal energy storage coupling technology and operates during peak power generation, the impact of steam consumption on electric draw from the carbon capture system is reduced. However, the electricity consumption of the equipment in the carbon capture system persists, resulting in an overall higher power supply capacity than that of a standard full flue gas capture unit. When the unit operates during power generation valleys, the peak-shaving effect of the thermal energy storage coupling can be further enhanced based on the full flue gas carbon capture, producing a “medium valley regulation depth effect.” The chemical solvent storage coupling mode for full flue gas capture units differs significantly from the thermal storage mode. During peak power generation, the carbon capture system can operate in “absorption-only” mode without regeneration, which substantially reduces equipment power consumption (because the primary power penalty of the capture system occurs in the CO₂ compression and liquefaction stage). This approach eliminates safety concerns regarding the peak power generation of the unit. Conversely, when the unit operates in power generation valleys, the carbon capture system can increase the regeneration demand. Compared to the thermal storage mode, this simultaneously creates additional power and thermal requirements, resulting in a “deep power generation effect with high valley regulation depth.” However, when comparing coupling modes, safety and operational constraints must be considered alongside flexibility. Steam thermal storage involves high-pressure steam storage and frequent charge–discharge cycles, which intensify thermal fatigue and maintenance demands on pressure vessels and valves. For chemical solvent storage, prolonged retention of carbon dioxide-rich solvents may accelerate solvent degradation and corrosion due to high CO₂ loading and oxygen ingress. Despite its superior decoupling capability, this approach may reduce absorption capacity while increasing solvent management and maintenance costs.

3.4. Operational Decision-Making Framework

The findings on how factors like load rate and thermal consumption level affect the specific power penalty, as revealed in Section 2.2, are critical for optimizing the operational scheduling, economic efficiency, and security of power plants.

3.4.1. Optimization of the Unit’s Role Positioning

The carbon capture system exhibits a lower impact on power generation at high unit loads than under low-load conditions, accompanied by a significant reduction in the specific power penalty. This negative correlation between load level and unit power consumption implies that, for grid dispatch, full flue gas capture units should be prioritized for high-load operation. Meanwhile, deep peak-shaving tasks should be increasingly allocated to flexible resources, such as energy storage systems. This operational strategy optimizes both economic efficiency and security of the entire grid.

3.4.2. Economic Break-Even of Carbon Capture System Operation

The specific power penalty of carbon capture directly determines its operational costs. The data from Section 2.2 reveal an inverse correlation between the capture rate and specific power penalty, along with a significant positive correlation with the thermal consumption level. The coefficients derived from this analysis provide critical inputs for calculating the dynamic break-even point of carbon capture operation. During periods of high electricity or low carbon prices, reducing the capture rate may prove more economical than maintaining full capture at high loads. Plant operators can leverage this model to develop dynamic economic optimization algorithms that evaluate electricity prices, carbon prices, and operating conditions in real-time, thereby making capture decisions that maximize profitability.

The thermal consumption level $k2$ is a critical coefficient determining economic viability. A case study demonstrated that a 50% increase in k can lead to an approximately 8.9% reduction in the unit’s net power output. This trend indicates that investing in high-efficiency solvents and optimizing process flows to lower $k2$ not only reduces steam consumption but also directly enhances grid revenue by mitigating the associated specific power penalty. Consequently, in project planning, priority should be given to technological alternatives with a lower $k2$. Furthermore, the additional revenue from reduced power penalty should be incorporated into the return on investment analysis to enable a more comprehensive valuation of new technologies.

In full flue gas carbon capture systems, the operating cost is primarily governed by electricity loss and solvent regeneration energy consumption. Electricity loss originates from both steam extraction–induced power output reduction and the auxiliary power demand of capture equipment. Economically, this loss represents an opportunity cost associated with reduced electricity export to the grid and can be quantified using the specific power penalty multiplied by the electricity price. Consequently, under high electricity price conditions, even moderate increases in specific power penalty can significantly deteriorate the economic performance of carbon capture, highlighting the importance of maintaining high-load operation or adopting flexible coupling modes during peak periods.

Solvent regeneration energy consumption, characterized by the specific thermal consumption level, affects operating cost both directly and indirectly. Higher thermal consumption increases the required steam extraction, which disrupts the turbine expansion process and further amplifies electricity loss through the “steam–power” coupling mechanism. Model results indicate that each 0.1 GJ/tCO₂ increase in thermal consumption leads to an approximately 2.83% rise in the specific power penalty, demonstrating a magnified economic impact rather than a simple linear addition. Therefore, reducing solvent regeneration energy demand not only lowers steam consumption but also effectively mitigates power output loss, making low-thermal-consumption solvents and optimized regeneration strategies critical for improving the economic viability of full-scale carbon capture systems.

3.4.3. Determination of the System Safety Operation Boundaries

To satisfy the steam demand under low-load conditions, it may be necessary to extract high-grade steam from the secondary reheat cold section, with an extraction rate potentially reaching 16%, which can readily trigger equipment safety concerns. This clearly defines the safe operational boundaries for a full-capture unit under the direct coupling mode. Failure to effectively manage the extraction parameters while indiscriminately pursuing a high capture rate at low loads compromises system security. Therefore, operational protocols must establish safety limits according to this trend.

Higher thermal consumption levels and intensive solvent regeneration accelerate solvent degradation, corrosion, and auxiliary equipment wear, thereby increasing maintenance costs and reducing system availability. Consequently, operating strategies must balance short-term economic benefits with long-term asset integrity. While storage coupling mitigates transient thermal loads and power constraints, it requires additional capital investment and increases maintenance complexity. Chemical solvent storage, though enhancing operational flexibility, necessitates stringent management of solvent degradation and corrosion risks. Developing operational strategies within safety and economic boundaries is critical for the reliable, sustainable deployment of large-scale carbon capture systems.

4. Conclusions

(1)

A dual-element coupling convergence mechanism of “steam” and “power generation” exists between the carbon capture and thermal power systems to achieve dynamic equilibrium. The specific thermal storage capacity of saturated water $g1$ and the CO₂ storage capacity of the rich solvent $g2$ are the key parameters reflecting the volumetric storage capacity in the steam thermal storage and chemical solvent storage coupling modes, respectively.

(2)

The higher thermal parameters of the secondary reheat cold section steam account for its greater linear impact on power generation, as evidenced by the coefficient comparison A = 0.53 < B = 0.61. The coefficient C = 0.1 reflects the coupling interaction between the two extraction streams and their nonlinear effects. The unit specific power penalties at different thermal consumption levels exhibits similar nonlinear variation patterns, reflecting the inherent nonlinear physical characteristics of the specific power penalty of the carbon capture system.

(3)

The direct coupling results between the thermal unit and carbon capture system show that under full flue gas carbon capture conditions, the power generation impact rate reaches 27.7% at 30% THA, which is 2% higher than that under the THA condition. The impacts under the 50% and 75% THA conditions were almost identical. The specific power penalty peaked at 420.7 kW·h/tCO₂ under 30% THA while reaching its minimum of 366.7 kW·h/tCO₂ under THA conditions. When the carbon capture ratio increased from 30% to 90%, the reduction in specific power penalty reached 64.4 kW·h/tCO₂ (14.94%). To minimize power loss, the carbon capture system should maintain a high capture rate.

(4)

Compared with conventional 1000 MW units, full flue gas carbon capture units exhibit reduced maximum power output capacity but generate significant deep peak-shaving effects. For instance, valley load levels can decrease by approximately 1/3 (based on a 300 MW baseline). Compared with standard full flue gas carbon capture units, the steam thermal storage coupling mode reduces the power consumption associated with steam usage, expands the peak power generation capacity, and further enhances the low-load operational capability. The chemical storage mode, which allows CO₂ absorption without regeneration during peak periods, minimizes the impact on peak power generation while maintaining full flue gas capture capacity. This ensures the highest grid security, enables increased regeneration demand during off-peak periods, and achieves an optimal deep peak-shaving performance.

(5)

In terms of dispatch, this study can guide the optimal load distribution for “carbon–power–heat” synergy in future. Economically, this aids in determining the optimal operating conditions and profit margins for carbon capture. Operational security defines the stable operating boundaries and guides the adoption of safer coupling modes. This study is applicable only to large units; recalibration may be required when applying to other scenarios. Future work should optimize real-time scheduling schemes and validate findings using large-scale CCUS operational data.