Post-combustion carbon capture mixing control system and method

Abstract

The invention provides a post-combustion carbon capture mixed control system and method, the system comprises a post-combustion carbon capture device and a control unit, the post-combustion carbon capture device comprises an absorption tower, a desorption tower, a lean and rich liquid heat exchanger and a reboiler, and the control unit comprises a distributed control module, a centralized control module and a ratio control module; the buffer tank stores and stabilizes the lean solution to enter the absorption tower; the reboiler heat source adjusting valve adjusts the steam supply amount according to the first optimization set value; the distributed control module is used for carrying out local stabilization control on the state of the lean solution in the buffer tank; the centralized control module performs optimization control on the overall performance of the post-combustion carbon capture device according to a pre-established target function to obtain a first optimization set value and a second optimization set value; and the ratio control module adjusts the flow ratio control of the lean solution and the flue gas according to a second optimization set value of the centralized control module, so that the stability of carbon capture control after combustion is improved.

Description

Technical Field

[0001] The present invention relates to the field of carbon capture control, and in particular to a post-combustion carbon capture hybrid control system and method. Background Art

[0002] With the increasing severity of global warming, reducing greenhouse gas emissions, particularly carbon dioxide (CO2), has become a widespread international consensus. Among the many emission reduction technologies, post-combustion carbon capture (PCC), particularly those based on amine chemical absorption, is relatively mature and offers high capture efficiency. A typical amine-based PCC system typically consists of an absorber and a desorber. Within the absorber, CO2-containing industrial flue gas and a lean solution undergo countercurrent contact, efficiently absorbing the CO2. The CO2-rich amine solution is then fed to the desorber, where it is regenerated at high temperatures, releasing high-purity CO2 gas to restore the lean solution, which is then recycled back to the absorber, forming a closed-loop system. To ensure the stable operation of this complex system, existing technologies primarily employ distributed PI / PID control or centralized model predictive control (MPC). PI / PID control is widely used due to its simple structure and ease of implementation. As an advanced control strategy, MPC utilizes dynamic models for online optimization, effectively handling the coupling effects and operational constraints of multivariable systems, and exhibits significant potential for improving system performance.

[0003] Although amine-based PCC technology and existing control strategies have achieved certain results in emission reduction, they still face severe challenges. On the one hand, the solution regeneration process in the desorption tower consumes a large amount of steam heat energy, which usually needs to be extracted from the upstream power plant, directly resulting in a significant decrease in the net power generation efficiency of the power plant, which constitutes the main part of the high operating costs of PCC technology; on the other hand, the system has a high degree of dynamic complexity and operational instability. The PCC device is closely coupled with the upstream power plant. The load fluctuations of the power plant will cause frequent changes in flue gas flow and CO2 concentration, which constitutes a continuous external disturbance to the PCC system. In addition, there is a strong material and energy coupling between the various units within the system. In particular, the operating condition fluctuations of the desorption tower (such as temperature and composition changes) will be directly transmitted to the absorption tower through the circulating lean solution, seriously affecting the absorption efficiency and even causing problems such as narrowing of the operating margin and system oscillation.

[0004] However, the existing control strategies have their own defects in dealing with these problems: the distributed PI / PID control cannot systematically deal with the strong coupling effect between variables, and it is difficult to achieve a good balance between ensuring high capture rate and achieving minimum energy consumption; while the fully centralized MPC is highly dependent on high-precision models, and the development and maintenance cost is high, and the centralized architecture may reduce the flexibility and robustness of the system, and it is too idealistic in dealing with sudden working conditions. More importantly, the existing technology generally fails to solve the core problem of the spread of the desorption tower disturbance to the absorption tower through the solution circulation, which seriously damages the stability of the entire system.

[0005] In summary, the existing PCC control strategy cannot achieve an ideal balance between simplicity, robustness and optimality, and cannot effectively deal with internal and external disturbances, especially cannot structurally suppress the internal propagation of key disturbances, which greatly limits the efficient, stable and energy-saving operation of PCC technology under dynamic working conditions, and is not conducive to improving the stability of PCC control.

[0006] In view of this problem, the present application provides a post-combustion carbon capture hybrid control system and method to solve the above problems. SUMMARY

[0007] The present application provides a post-combustion carbon capture hybrid control system and method to solve the above problems.

[0008] The first aspect of the present application provides a post-combustion carbon capture hybrid control system, comprising a post-combustion carbon capture device and a control unit in communication with the post-combustion carbon capture device, the post-combustion carbon capture device comprising: an absorption tower, a desorption tower, a lean-rich liquid heat exchanger, and a reboiler, the absorption tower and the desorption tower being connected by the lean-rich liquid heat exchanger for CO2 absorption and release, and the reboiler being connected with the desorption tower for supplying steam to the desorption tower; the post-combustion carbon capture device further comprises a buffer tank and a reboiler heat source regulating valve; the control unit comprises a distributed control module, a centralized control module, and a ratio control module; the buffer tank is used to store and stabilize the lean solution to be fed into the absorption tower; the reboiler heat source regulating valve is used to adjust the steam supply according to the first optimization set value; the distributed control module is used to locally stabilize the lean solution in the buffer tank; the centralized control module is used to optimize the global performance of the post-combustion carbon capture device according to the performance parameter measurement value of the post-combustion carbon capture device and the pre-established target function, to obtain the first optimization set value and the second optimization set value; and the ratio control module is used to adjust the flow ratio control of the lean solution and the flue gas according to the second optimization set value of the centralized control module, the inlet flue gas flow measurement value and the lean solution flow measurement value.

[0009] The second aspect of the present application provides a post-combustion carbon capture hybrid control method, which is realized based on the post-combustion carbon capture hybrid control system of the first aspect of the present application, and comprises the following steps: The distributed control module performs local stabilizing control on the lean solution state in the buffer tank; The centralized control module performs optimal control on the global performance of the post-combustion carbon capture device according to the performance parameter measurement values of the post-combustion carbon capture device and a pre-established target function, to obtain a first optimal set value and a second optimal set value; The reboiler heat source regulating valve adjusts the steam supply amount according to the first optimal set value; The ratio control module is used to adjust the flow ratio control of the lean solution and the flue gas according to the second optimal set value of the centralized control module, the inlet flue gas flow measurement value and the lean solution flow measurement value.

[0010] The technical scheme adopted by the present application has the following technical effects: 1. In the technical scheme of the present application, the buffer tank with multi-loop control is introduced, which constitutes an effective decoupling unit in terms of physics and control. It actively absorbs and eliminates a series of disturbances such as temperature, concentration and flow rate from the desorption tower side, prevents the propagation and amplification of these disturbances in the solution circulation loop, ensures that the absorption tower always works in stable and ideal conditions, improves the intrinsic stability and operability of the entire PCC system, controls the unstable lean solution in a stable state, realizes the functions of decoupling and eliminating disturbances, effectively solves the problem of low post-combustion carbon capture control stability caused by the prior art, and effectively improves the stability of post-combustion carbon capture control.

[0011] 2. In the technical scheme of the present application, the buffer tank is arranged on the lean solution circulation pipeline between the lean solution outlet of the lean-lean heat exchanger and the lean solution inlet of the absorption tower, is used to store the lean solution to be entered into the absorption tower, and maintains the stability of the liquid level, temperature and solution concentration of the lean solution to be entered into the absorption tower. The distributed control module comprises a liquid level control loop, a temperature control loop and a concentration control loop connected with the pipeline of the buffer tank. The liquid level control loop can maintain the constancy of the liquid level in the buffer tank. The temperature control loop can control the temperature of the lean solution in the buffer tank. The concentration control loop can compensate for the degradation loss of the solution and maintain the stability of the solution concentration. That is, the buffer tank control system comprises three independent PI control loops for controlling the temperature, concentration and liquid level, respectively. Residual deviations can be eliminated by adjusting the coolant flow regulating valve, the fresh amine solution supply regulating valve and the make-up water regulating valve, the unstable lean solution can be controlled in a constant temperature, constant concentration and constant amount state, and the functions of decoupling and eliminating disturbances can be realized.

[0012] 3、The preset constraint condition in the technical scheme of the present application includes hard constraint, soft constraint and input constraint, the hard constraint is that the predicted value of CO2 capture rate cannot be lower than the preset minimum environmental protection threshold at any time, the soft constraint is that the predicted value of unit CO2 capture energy consumption does not exceed the maximum energy consumption threshold, and the input constraint is that the manipulated variable satisfies the actual operable range, so that the control system is always operated in the mode close to the minimum energy consumption under the premise of meeting the capture rate requirement. Under dynamic disturbance, the EP value (energy performance index) of the present application can be stably controlled in the favorable interval of 3.1 MJ / kg CO2 or less, and the high unification of efficiency and economy is realized.

[0013] 4、The model predictive controller in the technical scheme of the present application directly sends the first optimized set value to the reboiler heat source regulating valve, realizes direct and optimized adjustment of the energy consumption of the desorption process, the ratio control module receives the current inlet flue gas flow as a feedforward signal, adjusts the speed of the lean solution circulating pump or the opening degree of the outlet valve, adjusts and controls the actual lean solution flow, so that the flow ratio between the actual lean solution flow and the current inlet flue gas flow can track the second optimized set value issued by the model predictive controller, and the CO2 capture rate can be stably kept near the target value when the flue gas flow is subjected to severe fluctuation of up to ±15%, the overshoot is small, and the adjustment time is short. This is much better than the traditional distributed PI control scheme, which often appears large capture rate fluctuation and longer recovery time under similar disturbance.

[0014] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical schemes in the embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below, and obviously, other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0016] Figure 1 It is the structural schematic diagram of the system of the embodiment one in the present application scheme; Figure 2 It is the method control strategy schematic diagram of the model predictive controller MPC in the system of the embodiment one in the present application scheme; Figure 3 It is the mixed control method control strategy schematic diagram in the system of the embodiment one in the present application scheme; Figure 4 It is the internal control strategy schematic diagram of the buffer tank in the system of the embodiment one in the present application scheme; Figure 5 It is the flow process schematic diagram of the embodiment two method in the present application scheme.

[0017] 1. Absorber; 2. Inlet flue gas flow transmitter; 3. Inlet flue gas concentration analyzer; 4. Inlet flue gas multiplication module; 5. Lean liquid concentration analyzer; 6. Lean liquid multiplication module; 7. Lean liquid flow transmitter; 8. Division module; 9. Ratio control module; 10. On-line concentration analyzer; 11. Liquid level sensor; 12. Make-up water control valve; 13. Liquid level controller; 14. Surge tank; 15. Make-up amine control valve; 16. Concentration controller; 17. Coolant flow control valve; 18. Temperature controller; 19. Desorber; 20. Absorber condenser; 21. CO2product multiplication module; 22. CO2capture rate calculation module; 23. Model predictive controller; 24. CO2product flow transmitter; 25. CO2product concentration analyzer; 26. Reboiler liquid level transmitter; 27. Reboiler heat source control valve; 28. Reboiler; 29. Reboiler heating coils; 30. Lean liquid circulation pump; 31. Lean / rich liquid heat exchanger; 32. Surge tank cooling coils; 33. Surge tank temperature transmitter; 34. Absorber inlet lean liquid control valve; 35. Rich liquid circulation pump. DETAILED DESCRIPTION

[0018] To clearly illustrate the technical features of the present application, the following detailed description and its accompanying drawings will be given to explain the present application in detail. The following disclosure provides many different embodiments or examples for implementing different structures of the present application. In order to simplify the disclosure of the present application, the components and settings of specific examples are described in the following. In addition, the present application can repeatedly refer to the numbers and / or letters in different examples. Such repetition is for the purpose of simplification and clarity, which itself does not indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. The present application omits the description of well-known components and processing techniques and processes to avoid unnecessary limitations on the present application.

[0019] Example One As Figure 1As shown, the present application provides a post-combustion carbon capture mixed control system, comprising a post-combustion carbon capture device and a control unit in communication connection with the post-combustion carbon capture device, the post-combustion carbon capture device comprising: an absorption tower 1, a desorption tower 19, a lean-rich liquid heat exchanger 31, a reboiler 28, the absorption tower 1 and the desorption tower 19 being connected through the lean-rich liquid heat exchanger 31 for CO2 absorption and release, and the reboiler 28 being connected with the desorption tower 19 for supplying steam to the desorption tower 19; the post-combustion carbon capture device further comprising: a buffer tank 14 and a reboiler heat source regulating valve 27, the control unit comprising a distributed control module, a centralized control module and a ratio control module 9, the buffer tank 14 being used for storing and stabilizing the lean solution to be entered into the absorption tower 1, and the reboiler heat source regulating valve 27 being used for adjusting the steam supply amount according to a first optimization set value; the distributed control module being used for locally stabilizing the lean solution in the buffer tank 14; the centralized control module being used for optimizing the global performance of the post-combustion carbon capture device according to the performance parameter measurement value of the post-combustion carbon capture device and a pre-established target function to obtain the first optimization set value and a second optimization set value; and the ratio control module 9 being used for adjusting the flow ratio control of the lean solution and the flue gas according to the second optimization set value of the centralized control module, the inlet flue gas flow measurement value and the lean solution flow measurement value.

[0020] The buffer tank 14 is arranged on the lean solution circulation pipeline between the lean solution outlet of the lean-rich liquid heat exchanger 31 and the lean solution inlet of the absorption tower 1, and is used for storing the lean solution to be entered into the absorption tower 1 and maintaining the stability of the liquid level, temperature and solution concentration of the lean solution to be entered into the absorption tower 1.

[0021] Referring to the accompanying drawings Figure 1 The system is described by taking a set of pilot-scale post-combustion carbon capture (PCC) demonstration devices using 30wt% concentration of monoethanolamine (MEA) aqueous solution as a standard chemical absorbent as an example, and the core target is to realize stable standard carbon dioxide capture rate, dynamic optimization of unit capture energy consumption and long-term stable operation of the whole process system under dynamic working conditions of dealing with large fluctuations of upstream flue gas load and components.

[0022] The post-combustion carbon capture device is composed of multiple unit devices, and the core unit of the system includes: an absorption tower 1 filled with 22-meter-high Mellapack 250Y type metal wire mesh structured packing inside, which aims to increase the gas-liquid contact area and ensure the realization of high capture rate, the tower body diameter is 1.5 meters, and the design is operated under the conditions of close to normal pressure and 320K inlet temperature which is most beneficial to the thermodynamics and kinetics of MEA absorbing CO2; a desorption tower 19 filled with 11-meter-high structured packing of the same type inside, the tower diameter is 1.3 meters, and the design is operated under the conditions of close to normal pressure and high temperature of about 380K to 395K, and the bottom of the tower is integrated with a tubular shell type reboiler 28, the steady-state design heat load is 2.1 MW, which is the most important energy input and consumption unit of the system, and its operating efficiency directly determines the economic feasibility of the entire PCC technology; the system is also equipped with a high-efficiency countercurrent tubular shell type lean-rich liquid heat exchanger 31, which is designed to achieve the maximum heat recovery between the rich and lean solutions, and the goal is to achieve the minimum temperature difference, which can reduce the steam consumption of the reboiler 28 and the total operating cost of the system; a vertical cylindrical buffer tank 14 has the function of attenuating high-frequency and medium-frequency disturbances from the desorption tower through its structure and control configuration, on the one hand, its volume is designed according to the maximum solution circulation amount and the expected disturbance period of the system, which can provide about 30 minutes of average residence time for the lean solution, physically forming a low-pass filter to attenuate the rapid fluctuations of the lean solution at the outlet of the desorption tower, on the other hand, the buffer tank control system (distributed control module) includes three independent PI control loops for controlling the temperature, concentration and liquid level of the lean solution in the buffer tank 14, which can control the unstable lean solution in a constant temperature, constant concentration and constant amount state, and realize the function of attenuating high-frequency and medium-frequency disturbances from the desorption tower; the lean-rich liquid heat exchanger 31 is a tubular shell heat exchanger used for heat recovery between the rich amine solution and the lean solution.

[0023] The specific flow path of the material flow path of the system is as follows: The CO2-containing inlet flue gas enters from the bottom of the absorption tower 1 and is countercurrently contacted with the lean solution flowing downward from the top of the tower, the CO2 in the flue gas is absorbed, and the purified tail gas is discharged from the top of the tower, and the lean solution absorbed with CO2 forms a rich solution at the bottom of the tower. The rich solution is pressurized by a rich solution circulating pump 35, preheated in a lean-rich liquid heat exchanger 31, and then sent to the top of a desorption tower 19. In the desorption tower 19, the rich solution is contacted with the rising water vapor, decomposes and releases CO2, and itself changes into high-temperature lean solution flowing out from the bottom; the released CO2 gas is cooled and dehydrated by a desorption tower condenser 20 and then output as a product. The high-temperature lean solution is pressurized by a lean solution circulating pump 30, enters the lean-rich liquid heat exchanger 31 to exchange heat with the rich solution in the shell side, and then enters the buffer tank 14 for treatment, and finally is sent back to the top of the absorption tower 1 through the absorption tower inlet lean solution regulating valve 34, completing the closed-loop circulation of the solution.

[0024] Specifically, the distributed control module includes: the liquid level control loop, temperature control loop, concentration control loop connected with buffer tank 14 pipeline; liquid level control loop includes liquid level sensor 11 (LT), liquid level controller 13 (LC), make-up water regulating valve 12, liquid level sensor 11 is used to transmit the measured liquid level signal in buffer tank 14 to liquid level controller 13, liquid level controller 13 is used to calculate the output liquid level signal according to the deviation of the measured liquid level signal in buffer tank 14 and the set liquid level value, adjust the opening of make-up water regulating valve 12, to maintain the constant liquid level in buffer tank 14; Temperature control loop includes temperature sensor 33 (TT), temperature controller 18 (TC), coolant flow regulating valve 17; temperature sensor 33 is used to transmit the measured solution temperature signal in buffer tank 14 to temperature controller 18, temperature controller 18 is used to calculate the output solution temperature signal according to the deviation of the measured solution temperature signal in buffer tank 14 and the set temperature value, adjust the opening of coolant regulating valve 17, to control the temperature of lean solution in buffer tank 14; Concentration control loop includes online concentration analyzer 10, concentration controller 16, make-up amine regulating valve 15 (fresh amine solution make-up regulating valve); online concentration analyzer 10 is used to transmit the measured solution concentration signal in buffer tank 14 to concentration controller 16, concentration controller 16 is used to calculate the output solution concentration signal according to the deviation of the measured solution concentration signal in buffer tank 14 and the set concentration value, adjust the opening of make-up amine regulating valve 15, to compensate for the solution degradation loss, maintain the stability of solution concentration.

[0025] Liquid level control loop: the liquid level in buffer tank 14 is continuously monitored by liquid level transmitter 11 (LT). When the measured value deviates from the pre-set liquid level value, a liquid level controller 13 (LC) will calculate the corresponding control action, and adjust the opening of a water make-up regulating valve 12 to accurately make up the water loss due to the system, so as to keep the liquid level at the set point.

[0026] Temperature control loop: the temperature of lean solution flowing into buffer tank 14 from lean-rich liquid heat exchanger 31 is continuously monitored by a temperature transmitter 33 (TT). When the measured value is higher than the pre-set temperature value most conducive to absorption reaction, a temperature controller 18 (TC) will adjust the flow of cooling water through the buffer tank cooling coil 32 in tank 14 to accurately cool and stabilize the lean solution at the target temperature.

[0027] Concentration control loop: the MEA concentration in lean solution is continuously monitored by an online concentration analyzer 10 (AT). When the measured value is lower than the pre-set optimal concentration value, a concentration controller 16 (AC) will start a make-up pump 15 (make-up amine regulating valve) to inject high-concentration fresh MEA solution to compensate for the solution degradation loss in system operation, and ensure the chemical activity of the solution.

[0028] The controller parameters (proportional gain P and integral time I) of the three distributed PI control loops are set by standard engineering tuning methods to achieve fast, accurate and overshoot-free or small-overshoot control of the respective targets.

[0029] The liquid level controller 13, the temperature controller 18, and the concentration controller 16 in the distributed control module are all proportional integral controllers. The set value input end of the ratio control module 9 (a feedforward ratio controller) is connected with the output end of the centralized control module, the process variable input end is connected with the inlet flue gas flow measurement point (the inlet of the absorption tower 1) and the lean solution flow measurement point (the lean solution circulating pipeline between the buffer tank 14 and the absorption tower 1), and the output end is connected with the lean solution circulating pump 30, which is used to perform the flow ratio control of the lean solution and the flue gas. The inlet flue gas flow measurement point includes the inlet flue gas flow collected by the inlet flue gas flow transmitter 2 and the inlet flue gas concentration collected by the inlet flue gas concentration analyzer 3, and the two are multiplied by the inlet flue gas multiplication operation module 4 to obtain the molar flow (G) of the inlet flue gas. The lean solution flow measurement point includes the lean solution flow collected by the lean solution flow transmitter 7 and the lean solution concentration collected by the lean solution concentration analyzer 5, and the two are multiplied by the lean solution multiplication operation module 6 to obtain the molar flow (L) of the lean solution. The ratio of the molar flow of the lean solution to the inlet flue gas can be obtained by the division operation module 8.

[0030] The centralized control module includes a model predictive controller 23, and the controlled variables of the model predictive controller include at least the CO2 capture rate of the purified flue gas at the outlet of the desorption tower 19 and the liquid phase temperature in the reboiler. The manipulated variables of the model predictive controller include the first optimized set value supplied to the reboiler and the second optimized set value supplied to the ratio control module. The first optimized set value is the set value of the reboiler heat load, and the second optimized set value is the set value of the ratio of the molar flow of the lean solution to the inlet flue gas.

[0031] The CO2 capture rate can be calculated by the CO2 capture rate calculation module 22, that is, the CO2 product flow collected by the CO2 product flow transmitter 24 at the outlet of the desorption tower 19 and the CO2 product concentration collected by the CO2 product concentration analyzer 25 at the outlet of the desorption tower 19 are multiplied to obtain the CO2 product capture molar flow. Then, according to the molar flow of the inlet flue gas obtained by multiplying the inlet flue gas flow and the inlet flue gas concentration at the inlet of the absorption tower 1, the CO2 capture rate is obtained according to the CO2 product capture molar flow and the molar flow of the inlet flue gas.

[0032] The reboiler liquid temperature in reboiler 28 can be acquired via reboiler liquid temperature transmitter 26. Reboiler heating coil 29 can be used to heat the heat source in reboiler 28, generating steam that is transported from reboiler 28 to desorption tower 19. Buffer tank cooling coil 32 can be used to cool the lean solution in buffer tank 14. An absorption tower inlet lean liquid regulating valve 32 is provided on the lean solution pipeline between buffer tank 14 and absorption tower 1 to regulate the flow of lean solution entering absorption tower 1.

[0033] Among them, Figures 2-3 As shown, in the model predictive controller 23 (MPC), the global performance of the post-combustion carbon capture device is optimized and controlled according to the performance parameter measurement values ​​of the post-combustion carbon capture device and the pre-established objective function, and the first optimized set value and the second optimized set value are obtained specifically including: S101, collecting actual measured values ​​of the controlled variable and the inlet flue gas flow rate as a disturbance variable during a preset sampling period; Specifically, the disturbance variable DV, i.e., the actual measured value of the inlet flue gas, is obtained; the actual measured values ​​of the controlled variables (CO2 capture rate, reboiler liquid phase temperature in the reboiler) are obtained; S102, at each control moment (current moment), based on the dynamic state space model and the current state variables, predict the impact trajectory of different manipulated variable sequences on the state variables in the future prediction time domain; The dynamic state space model includes an absorption tower model, a desorption tower model, a lean-rich liquid heat exchanger model, and a buffer tank model. The absorption tower model and the desorption tower model are both one-dimensional dynamic packed tower models, which include mass, component, and energy conservation equations. The lean-rich liquid heat exchanger is a one-dimensional countercurrent heat exchange model. The buffer tank model includes a heat balance equation and an amine solution concentration conservation equation. The centralized control module (MPC controller) relies on a built-in, high-precision dynamic state-space model based on first principles. This model accurately predicts the dynamic behavior of the PCC unit under different operating conditions. Specifically, both the absorption and desorption towers use a one-dimensional dynamic packed tower model, discretizing the tower axially into several control volume units. Within each unit, conservation equations for mass, composition, and energy are established for the gas phase (G) and liquid phase (L) to describe the mass transfer, absorption, and desorption processes of CO2.

[0034] Among them, the total mass conservation equation is: , Wherein, j represents the phase, gas phase is G, liquid phase is L; is the density of phase j (kg / m³); is the tower cross-sectional area (m²); is the linear velocity of phase j (m / s); is the tower axial coordinate (m); is the molar flux of component i (mol / m²·s); is the molar mass (kg / mol); the symbol ± indicates the positive or negative direction of the interphase mass transfer, gas→liquid in the absorption column and liquid→gas in the desorption column.

[0035] The component mass balance equation is: , where, is the mass concentration of component i in phase j (kg / m³); is the stoichiometric coefficient of component i in the chemical reaction; is the reaction rate (mol / m³·s), which is used to describe the reaction kinetics of CO2 and MEA in the amine solution.

[0036] The heat balance equation is: , where, is the specific heat at constant pressure (J / kg·K); is the temperature of phase j (K); is the interphase average heat transfer coefficient (W / m²·K); a is the specific surface area (m² / m³); is the reaction enthalpy (J / mol); is the latent heat of phase transition of component i (J / mol); is the temperature of the gas-liquid interface (K); The lean-rich liquid heat exchanger uses a simplified one-dimensional counterflow heat exchange model, considering the heat exchange process between the rich liquid and the lean liquid in the shell and tube sides, and ignoring phase transition and heat loss. The one-dimensional counterflow heat exchange model is: , , where, is the temperature of the hot fluid (K); is the temperature of the cold fluid (K); is the mass flow rate of the hot fluid (kg / s); is the mass flow rate of the cold fluid (kg / s); is the specific heat capacity of the hot fluid (J / kg·K); is the specific heat capacity of the cold fluid (J / kg·K); is the overall heat transfer coefficient of the lean-rich liquid heat exchanger (W / K); this equation is used to predict the outlet temperature of the heat exchanger, which serves as the boundary condition for the inlet conditions of the absorption and desorption columns.

[0037] The buffer tank 14 is a completely mixed vessel, and its model includes the heat balance equation: , where, is the liquid temperature in the buffer tank; is the lean liquid temperature entering the buffer tank; is the cooling water temperature; is the buffer tank volume; is the lean liquid density in the buffer tank; is the feed flow rate; 2 is the product of the heat transfer coefficient of the cooling coil and the cooling coil area; this equation is used to characterize the dynamic buffering process of high-frequency thermal disturbance.

[0038] The buffer tank 14 model also includes an amine solution concentration conservation equation: , where, is the amine concentration in the buffer tank; is the feed amine concentration; is the discharge amine concentration; is the volume flow rate of the feed amine; is the volume flow rate of the discharge amine; this equation reflects the real-time compensation and concentration control dynamics when the concentration fluctuates due to solution degradation.

[0039] The present application builds a high-fidelity dynamic model that can completely describe the material and energy transfer law of the PCC system through dynamic modeling of the above four core units (absorption tower, desorption tower, buffer tank, and lean- rich liquid heat exchanger). The high-fidelity dynamic model is the basis for the prediction and optimization of the model predictive controller (MPC). In the prediction process, the MPC first discretizes these nonlinear partial differential equations in space to form a high-dimensional dynamic state space model. At each control period, the MPC performs the following prediction steps: As shown in Figure 2 , first, the input parameters (the input parameters include disturbance measurement values, process measurement values, and external set values. The disturbance measurement value DV is the inlet flue gas flow rate, the process measurement value PV is the CO2 capture rate and the reboiler temperature, and the external set value SP is the set target capture rate and the target reboiler temperature) are obtained.

[0040] Second, the MPC substitutes the above inputs (the input parameters include disturbance measurement values, process measurement values, and external set values. The disturbance measurement value DV is the inlet flue gas flow rate, the process measurement value PV is the CO2 capture rate and the reboiler temperature, and the external set value SP is the set target capture rate and the target reboiler temperature) into its internal dynamic state space model, performs a forward simulation from the current time k, and calculates the evolution trajectory of all state variables in the future prediction time domain (P). Specifically, the state estimator first predicts the state, at each control time k, the state estimator uses the last optimal state estimate at the previous time k-1, the control input already implemented and the current measured disturbance to predict the state prior estimate at the current time k through the internal model (directly output as the state variable prediction); preferably, on this basis, error calculation and state correction can also be carried out, the controlled variable measurement value at the current time k is collected from each sensor and compared with the output prediction value calculated based on the state prior estimate to obtain the prediction error, and the state estimator corrects the state prior estimate through a gain matrix according to the error value to obtain the feedback-corrected state posterior estimate; finally, this corrected state is the optimal estimate of the real physical state of the current system, and this optimal estimate is the prediction value of the state variable generated based on the current value of the state variable, which also serves as an accurate initial condition for the MPC to predict the future trajectory at the next time, and finally this corrected state is the optimal estimate of the real physical state of the current system, which also serves as an accurate initial condition for the MPC to predict the future trajectory, through cyclic feedback, the state of the model is constantly updated, thereby ensuring the accuracy of the prediction and the robustness of the control. (The prediction process is based on the state balance equations in the dynamic state space model, and when the MPC controller predicts, the state balance equations are discretized and linearized in space to convert into linear difference equations to directly describe the change of the state variable with time. In the embodiment of the present application, the state variable specifically includes: the temperature of the liquid phase, the temperature of the gas phase, the concentration of each component, the temperature, the concentration and the liquid level height of the buffer tank in the absorption tower and the desorption tower (i.e. the parameters in the balance equation in the dynamic space model). Taking the calculation of the CO2 capture rate as an example, the CO2 capture rate should be the difference between the CO2 flow rate at the inlet flue gas and the outlet divided by the CO2 flow rate at the inlet; if the temperature of the reboiler is an element in the state vector x(k) and there is a functional relationship between the controlled variable capture rate, this functional relationship is abstracted as y(k)=Cx(k) to calculate the prediction value). The current values of all state variables are obtained through sensor measurement and state estimator, which are used as the initial conditions for simulation, the optimizer generates the reboiler heat load and L / G ratio at each time within the future control time domain (M), measures the main disturbance such as the current inlet flue gas flow rate, and assumes that it remains unchanged or is extrapolated according to the trend within the prediction time domain (P); Finally, the future prediction value of the controlled variable (CVs, i.e. y(k)) is calculated from the state variable trajectory x(k) obtained from the simulation, and this series of predicted trajectories of the controlled variables are then sent to the optimization solver to find the optimal control action of the manipulated variable.

[0041] S103, establishing a target function based on the dynamic state space model and the current state variable; The target function is specifically: , wherein J is a target function value, is a controlled variable prediction value at the time point; is a controlled variable set value at the time point; is a change amount of the manipulated variable; Q, R are weight matrices; M is a total number of time points; , , wherein, is a state variable prediction value at the time point, is a state variable measurement value at the -1 time point; is a manipulated variable actual value at the time point; is a controlled variable prediction value at the time point; is a disturbance variable measurement value at the -1 time point; F, B, C are system matrix, input matrix, output matrix respectively; E is an influence matrix of the disturbance variable on the state variable.

[0042] It should be noted that in the present scheme , represents a functional relationship between the controlled variable and the state variable at the current time point or the next time point, , , , represents a functional relationship between the state variable at the current time point or the next time point and the state variable, the manipulated variable, the disturbance variable at the previous time point or the current time point, i.e. the controlled variable prediction value at the current time point is obtained from the state variable prediction value at the current time point, and the state variable prediction value at the current time point is obtained from the state variable measurement value, the manipulated variable value, the disturbance variable measurement value at the previous time point; the controlled variable prediction value at the next time point is obtained from the state variable prediction value at the next time point, and the state variable prediction value at the next time point is obtained from the state variable measurement value, the manipulated variable value, the disturbance variable measurement value at the current time point.

[0043] the disturbance variable The system online measurement is used as the known feedforward input of the prediction model at each current control time k to participate in the prediction and optimization of future control effect, thereby effectively improving the robustness and capture rate stability of the system to high-frequency flue gas disturbance. The disturbance influence matrix E directly links the influence of external disturbance to the change of the internal state of the system, and the matrix is derived from the original state balance equation during modeling by linearization. For example, the change of the inlet flue gas flow (a component of the disturbance variable d) will directly affect the gas phase flow rate term in the mass conservation equation, and this influence is embodied in the specific element value of the matrix E in the final linear model. In the prediction process, MPC substitutes the real-time measured d(k) into the model, calculates the direct impact of the disturbance on the next state through the Ed(k) term, so that the controller can make a response in advance to achieve feedforward suppression of the disturbance.

[0044] The controlled variable is y, the manipulated variable is u, and the state variable is x. In the prediction process, at the current time k, the actual controlled variable value y(k) is measured, the state estimator predicts the current state variable x(k) using the state x(k-1) and the control input u(k-1) at the previous time, and then obtains the controlled variable y(k). By comparing the difference between the measured value y(k) 测 and the predicted y(k) 预 , the state estimator corrects the state variable to obtain the most accurate state estimate x(k); taking this as the initial condition for the next time, the value of x at the next time is obtained according to the state space model, so as to calculate the value of the controlled variable y at the next time. The optimizer finds the control sequence of the manipulated variable that minimizes the objective function.

[0045] S104, based on the preset constraint condition, the objective function is solved to find a set of optimal manipulated variable control sequences in the control time domain, which minimizes the deviation of the predicted controlled variable trajectory from the respective set value; The preset constraint condition includes hard constraint, soft constraint and input constraint. The hard constraint is that the predicted value of the CO2 capture rate at any time cannot be lower than the preset minimum environmental protection threshold. The soft constraint is that the predicted value of the unit CO2 capture energy consumption does not exceed the maximum energy consumption threshold. The input constraint is that the manipulated variable satisfies the actual operable range.

[0046] Specifically, the hard constraint is used to ensure that the environmental protection index of the system meets the minimum requirement, specifically: , Wherein, represents the predicted value of the CO2 capture rate at the kth step or the kth time in the prediction period, is the set CO2 capture rate threshold.

[0047] The economic soft constraints on unit CO2 capture energy consumption (EP) are as follows: , in, is the predicted value of unit CO2 capture energy consumption at the kth step or kth moment in the prediction period (MJ / kg CO2), which can be calculated by the reboiler heat load and capture capacity; is the optimization weight of the energy consumption target.

[0048] The input constraints that serve as the physical upper and lower limits of all manipulated variables are: , , in, is the reboiler heat load at the kth moment; is the lower limit of reboiler heat load; is the upper limit of reboiler heat load; is the molar flow ratio of lean solution to flue gas at the kth moment; is the minimum permissible value of the L / G ratio; is the maximum allowable value of the L / G ratio.

[0049] Solving the objective function is actually solving a constrained quadratic programming (QP) optimization problem online, which aims to find a series of optimal manipulated variables (MVs) within a preset control time domain (M) in the future - that is, the set values ​​of the reboiler heat load (Q_reboiler) and the lean solution to flue gas molar flow ratio (L / G). The objective function of this optimization problem aims to minimize the weighted square error between the predicted trajectory of the controlled variable and its set value, as well as the weighted square sum of the manipulated variable changes. During the optimization solution process, the optimizer must strictly meet all preset constraints, including input constraints, hard output constraints, and soft output constraints. The task of the optimizer is to find the optimal manipulated variable control sequence that minimizes the objective function among all manipulated variable control sequences that meet the constraints. The entire optimization solution process is implemented in the optimization solver, and the specific working principle is as follows: First, at each control time k, the optimization solver receives the external set value and the predicted trajectory input output by the state estimator after correction (correction is also possible), and performs optimization according to the objective function and constraints; secondly, during the optimization process, the optimization solver needs to find a set of incremental sequences of manipulated variable control actions in the future control time domain M. This set of sequences can minimize the value of the objective function, and all predicted trajectories and control actions caused by this set of manipulated variable control actions in the entire prediction time domain are carried out under the constraints. After the solution is completed, the entire optimal manipulated variable control sequence is output. Finally, according to the rolling time domain control principle, the system only applies the first control action in this optimal manipulated variable control sequence to the actual system operation, and the remaining control actions are discarded. The final output reboiler heat load set value and L / G ratio set value at the current moment are output. At the next control moment, data collection, state estimation, and optimization solution will be performed again to calculate and update the optimal manipulated variable control sequence.

[0050] S105: Output the first action in the optimal manipulated variable control sequence as the control instruction at the current moment.

[0051] After the solution is completed, the entire optimal manipulated variable control sequence is output. Finally, according to the rolling time domain control principle, the system only applies the first control action in this optimal manipulated variable control sequence to the actual system operation, and the remaining control actions are discarded. The final output reboiler heat load set value and L / G ratio set value (manipulated variable) at the current moment are output. At the next control moment, data collection, state estimation, and optimization solution will be performed again to calculate and update the optimal manipulated variable control sequence.

[0052] like Figure 3 As shown, the post-combustion carbon capture device also includes a temperature control module (which can be a separate temperature controller or executed by the temperature controller 18 in the temperature control loop of the distributed control module). The model predictive controller obtains a first optimized set value and a second optimized set value based on the inlet flue gas flow rate, CO2 capture rate, and reboiler temperature measurement value among the disturbance variables, and directly sends the first optimized set value (reboiler temperature set value) to the temperature control module. The temperature control module (reboiler temperature PI controller) controls and adjusts the reboiler heat source regulating valve based on the first optimized set value, adjusts the steam supply from the reboiler 28 to the desorption tower 19, and realizes direct and optimized regulation of the energy consumption of the desorption process. The model predictive controller sends the obtained second optimized set value (L / G set value) to the ratio control module. The ratio control module (feedforward ratio controller FFC) receives the current inlet flue gas flow rate as a feedforward signal, and adjusts and controls the actual lean solution flow rate by adjusting the speed of the lean solution circulation pump or the opening of the outlet valve, so that the flow ratio between the actual lean solution flow rate and the current inlet flue gas flow rate can track the second optimized set value issued by the model predictive controller.

[0053] Specifically, the optimal control instructions calculated by MPC are executed through a hierarchical instruction structure, which includes: (1) Direct control execution: The set value of the optimal reboiler heat load (Q_reboiler) calculated by MPC is directly sent to the bottom control loop of the regulating valve controlling the reboiler steam supply amount, to realize direct and optimized adjustment of the energy consumption of the desorption process.

[0054] (2) Cascade control execution: The set value of the optimal lean solution to flue gas molar flow ratio (L / G) calculated by MPC is sent to a ratio control module (FFC) as a dynamic target value, and the FFC simultaneously receives the current inlet flue gas flow (G_actual) from a flow meter (FT) as a feedforward signal. The core task of the FFC is to quickly adjust the speed of the lean solution circulating pump or the opening of its outlet valve, so as to accurately control the actual lean solution flow (L_actual), so that the actual flow ratio (L_actual / G_actual) can track the dynamic optimal set value issued by MPC in real time and accurately. This cascade structure can realize fast feedforward compensation and suppression of the most important external disturbance, i.e. flue gas flow.

[0055] As shown in Figure 4 , the liquid level controller 13, the temperature controller 18, and the concentration controller 16 control the corresponding make-up water regulating valve 12, the coolant flow regulating valve 17, and the make-up amine regulating valve 15 through the liquid level error / temperature error / concentration error between the actual measured liquid level / temperature / concentration in the buffer tank 14 and the corresponding liquid level set value / temperature set value / concentration set value in the control process, to realize fast, accurate, and no or small overshoot control of the respective targets.

[0056] Through the synergistic effect of the above structure and method, the specific embodiment of the present application constructs an advanced control system that deeply integrates local stabilization and global optimization, structural decoupling and algorithm synergy, aiming to fundamentally solve the problem of disturbance propagation in the PCC system, and to realize long-term stable, efficient and energy-saving operation of the device under dynamic changing conditions.

[0057] The embodiment of the present application can stabilize the CO2 capture rate near the target value when the flue gas flow fluctuates by up to ±15%, with small overshoot and short regulation time, through the direct and closed-loop control of the CO2 capture rate by MPC and the cascade cooperation with the PI ratio controller (ratio control module). Moreover, the optimization capability and constraint processing capability of the MPC for the energy performance index (EP) ensure that the system always operates in a way close to the lowest energy consumption under the premise of meeting the capture rate requirement. Under dynamic disturbance, the EP value can be stably controlled in the favorable range below 3.1 MJ / kg CO2, realizing the high unification of efficiency and economy; by using the constraint processing capability of the MPC, the minimum CO2 capture rate is set as a hard constraint, which ensures that the environmental protection emission index will not be broken regardless of the working condition, providing a solid guarantee for the safe and compliant operation of the PCC device.

[0058] The present application creatively introduces a buffer tank with multi-loop PI precision control, which constitutes an effective decoupling unit in terms of physics and control, actively absorbs and eliminates a series of disturbances such as temperature, concentration and flow from the side of the desorption tower, prevents the propagation and amplification of these disturbances in the solution circulation loop, and ensures that the absorption tower always works in stable and ideal conditions, improving the intrinsic stability and operability of the entire PCC system. Specifically, the decoupling and disturbance elimination effect of the buffer tank is first that its huge physical volume and long residence time constitute a low-pass filter, which can greatly attenuate and smooth the high-frequency disturbances such as temperature and concentration fluctuations from the side of the desorption tower. On the other hand, the buffer tank control system includes three independent PI control loops for controlling temperature, concentration and liquid level, respectively, can compare the smoothly changed solution state with a fixed optimal set value, and immediately adjust the coolant flow regulating valve, fresh amine solution supply regulating valve and make-up water regulating valve to eliminate the residual deviation, control the unstable lean solution in a constant temperature, constant concentration and constant amount state, and realize the decoupling and disturbance elimination effect.

[0059] The mixed architecture of the present application integrates the advantages of PI and MPC. The PI control of the buffer tank is simple, reliable, easy to implement and maintain, and conforms to the field operation habit. The MPC focuses on global and complex optimization tasks and handles the multivariable coupling and constraint problems it is best at. This "division of labor and cooperation" mode realizes global optimal control while maintaining the modularity and flexibility of the system, and has higher comprehensive performance and engineering application value than the completely centralized MPC or completely distributed PI.

[0060] In conclusion, the technical scheme of the present application introduces a buffer tank with multi-loop control, which constitutes an effective decoupling unit in terms of physics and control, actively absorbs and eliminates a series of disturbances such as temperature, concentration, flow rate from the side of the desorption tower, prevents the propagation and amplification of these disturbances in the solution circulation loop, ensures that the absorption tower always works in stable and ideal conditions, improves the inherent stability and operability of the entire PCC system, controls the unstable lean solution in a stable state, realizes the functions of decoupling and eliminating disturbances, effectively solves the problem of low stability of post-combustion carbon capture control caused by the prior art, and effectively improves the stability of post-combustion carbon capture control.

[0061] In the technical scheme of the present application, the buffer tank is arranged on the lean solution circulation pipeline between the lean solution outlet of the lean- rich solution heat exchanger and the lean solution inlet of the absorption tower, is used to store the lean solution to be introduced into the absorption tower, and maintains the stability of the liquid level, temperature and solution concentration of the lean solution to be introduced into the absorption tower; the distributed control module comprises a liquid level control loop, a temperature control loop and a concentration control loop connected with the pipeline of the buffer tank; the liquid level control loop can maintain the constancy of the liquid level in the buffer tank; the temperature control loop can control the temperature of the lean solution in the buffer tank; and the concentration control loop can compensate for the degradation loss of the solution and maintain the stability of the solution concentration, that is, the buffer tank control system comprises three independent PI control loops for controlling the temperature, concentration and liquid level, can eliminate residual deviation by adjusting the coolant flow regulating valve, fresh amine solution supply regulating valve and supply water regulating valve, controls the unstable lean solution in a constant temperature, constant concentration and constant amount state, realizes the functions of decoupling and eliminating disturbances, and has the advantages of simple structure, low cost, high stability and the like.

[0062] In the technical scheme of the present application, the preset constraint conditions include hard constraints, soft constraints and input constraints; the hard constraints are that the predicted value of the CO2 capture rate cannot be lower than the preset minimum environmental protection threshold at any time; the soft constraints are that the predicted value of the unit CO2 capture energy consumption does not exceed the maximum energy consumption threshold; and the input constraints are that the manipulated variables satisfy the actual operable range, so that the control system is always operated in a manner close to the minimum energy consumption under the premise of meeting the capture rate requirement. Under dynamic disturbance, the EP value (energy performance index) of the present application can be stably controlled in the favorable range of 3.1 MJ / kg CO2 or less, and the high unification of efficiency and economy is realized.

[0063] The model predictive controller of the technical solution of the present invention sends the first optimized set value directly to the reboiler heat source regulating valve, achieving direct and optimized regulation of the energy consumption of the desorption process. The ratio control module receives the current inlet flue gas flow as a feedforward signal and adjusts the actual lean solution flow by adjusting the speed of the lean solution circulation pump or the opening of the outlet valve. The flow ratio between the actual lean solution flow and the current inlet flue gas flow can track the second optimized set value issued by the model predictive controller. Even when the flue gas flow fluctuates violently by up to ±15%, the CO2 capture rate can still be stabilized near the target value, with small overshoot and short adjustment time. This is far superior to traditional distributed PI control schemes, which often experience large capture rate fluctuations and longer recovery times under similar disturbances.

[0064] Example 2 like Figure 5 As shown, the technical solution of the present invention further provides a post-combustion carbon capture hybrid control method, which is implemented based on a post-combustion carbon capture hybrid control system in Example 1, and includes: S1, a distributed control module performs local stabilization control on the state of the lean solution in the buffer tank; Local stability control, serving as the basic stability layer, is a distributed control module that implements continuous, independent local control of the buffer tank. The module consists of three parallel closed-loop control loops based on proportional-integral (PI) controllers: Liquid level control loop: A level transmitter (LT) continuously monitors the liquid level in the buffer tank. When the measured value deviates from the preset level, a level controller (LC) calculates the appropriate control action and adjusts the opening of a water supply valve to accurately replenish water lost from the system, thereby maintaining a constant liquid level at the set point.

[0065] Temperature Control Loop: A temperature transmitter (TT) continuously monitors the temperature of the lean solution flowing into the buffer tank from the lean-rich solution heat exchanger. When the measured value exceeds the preset temperature that is most conducive to the absorption reaction, a temperature controller (TC) adjusts the flow of cooling water through the tank's cooling coils to precisely cool the lean solution and stabilize it at the target temperature.

[0066] Concentration control loop: An online concentration analyzer (AT) continuously monitors the MEA concentration in the lean solution. When the measured value falls below the preset optimal concentration, a concentration controller (AC) activates a make-up pump to inject fresh, high-concentration MEA solution to compensate for solution degradation losses during system operation and ensure solution chemical activity.

[0067] The controller parameters (proportional gain P and integral time I) of the three distributed PI control loops are set by standard engineering tuning methods to achieve fast, accurate and no or small overshoot control of their respective targets.

[0068] S2, the centralized control module optimizes the global performance of the post-combustion carbon capture device according to the performance parameter measurement value of the post-combustion carbon capture device and a pre-established target function, to obtain a first optimization set value and a second optimization set value; System global performance optimization control, the centralized control module (model predictive controller MPC) executes its core optimization control algorithm with a fixed sampling time as the period. At each sampling time, its operation process is as follows: Information collection, the MPC collects key process information from sensors at various parts of the device, including: CO2 capture rate (CC) and reboiler liquid temperature (T_reboiler) two controlled variables (CVs) of the current actual value, at the same time, it also collects the measurement value of the inlet flue gas flow which is the main external disturbance variable (DV).

[0069] Prediction and optimization, the MPC uses its internal linear dynamic model to predict how the controlled variables will evolve in the future if no new control action is applied within a pre-set prediction horizon (P), and then it solves a constrained quadratic programming (QP) optimization problem online, aiming to find a series of optimal manipulated variables (MVs) - namely the reboiler heat load (Q_reboiler) and the set value of the molar flow ratio of lean solution to flue gas (L / G) - within a pre-set control horizon (M). During the optimization solving process, the optimizer must strictly satisfy all pre-set constraints, including input constraints, hard output constraints and soft output constraints. The task of the optimizer is to find the optimal manipulated variable control sequence that minimizes the objective function among all constraint-satisfying control sequences.

[0070] Control output: after the optimization solving is completed, the MPC only outputs the first action of the calculated optimal control sequence as the control instruction at the current time. At the next sampling time, it will repeat the above-mentioned "collection-prediction-optimization-output" process. This "rolling optimization" mechanism based on feedback correction enables the MPC to effectively deal with model mismatch and unmeasured disturbances.

[0071] S3, the reboiler heat source regulating valve adjusts the steam supply amount according to the first optimization set value; The model prediction controller obtains first optimization set value and second optimization set value according to the inlet flue gas flow, CO2 capture rate and reboiler temperature measurement in the disturbance variable, and the first optimization set value (reboiler temperature set value) is directly sent to the temperature control module, and the temperature control module (reboiler temperature PI controller) controls the adjusting reboiler heat source adjusting valve according to the first optimization set value, adjusts the steam supply of the reboiler to the desorption tower, and realizes direct and optimized adjustment of the energy consumption of the desorption process.

[0072] S4, the ratio control module adjusts the flow ratio control of the lean solution and the flue gas according to the second optimization set value of the centralized control module, the inlet flue gas flow measurement value and the lean solution flow measurement value.

[0073] The model prediction controller sends the obtained second optimization set value (L / G set value) to the ratio control module, and the ratio control module (feedforward ratio controller FFC) receives the current inlet flue gas flow as a feedforward signal, adjusts the speed of the lean solution circulating pump or the opening degree of the outlet valve, adjusts and controls the actual lean solution flow, so that the flow ratio between the actual lean solution flow and the current inlet flue gas flow can track the second optimization set value issued by the model prediction controller.

[0074] In the technical scheme of the present application, a buffer tank with multi-loop control is introduced, which constitutes an effective decoupling unit in physics and control, actively absorbs and eliminates a series of disturbances such as temperature, concentration and flow from the desorption tower side, prevents the propagation and amplification of these disturbances in the solution circulation loop, ensures that the absorption tower always works in stable and ideal conditions, improves the inherent stability and operability of the entire PCC system, controls the unstable lean solution in a stable state, realizes the functions of decoupling and eliminating disturbances, effectively solves the problem of low carbon capture control stability caused by the prior art, and effectively improves the stability of the carbon capture control after combustion.

[0075] In the technical solution of the present invention, the buffer tank is arranged on the lean solution circulation pipeline between the lean solution outlet of the lean-rich liquid heat exchanger and the lean solution inlet of the absorption tower, and is used to store the lean solution to be entered into the absorption tower and maintain the stability of the liquid level, temperature and solution concentration of the lean solution to be entered into the absorption tower; the distributed control module includes: a liquid level control loop, a temperature control loop and a concentration control loop connected to the buffer tank pipeline; the liquid level control loop can maintain the constant liquid level in the buffer tank; the temperature control loop can control the temperature of the lean solution in the buffer tank; the concentration control loop can compensate for the solution degradation loss and maintain the stability of the solution concentration, that is, the buffer tank control system includes three independent PI control loops, which respectively control the temperature, concentration and liquid level. The residual deviation can be eliminated by adjusting the coolant flow control valve, the fresh amine solution supply control valve and the supply water control valve, and the unstable lean solution can be controlled in a constant temperature, constant concentration and constant quantity state to achieve the effect of decoupling and eliminating disturbances.

[0076] The preset constraints in the technical solution of this invention include hard constraints, soft constraints, and input constraints. The hard constraint stipulates that the predicted CO2 capture rate must not fall below a preset minimum environmental threshold at any time; the soft constraint stipulates that the predicted energy consumption per unit of CO2 capture must not exceed a maximum energy consumption threshold; and the input constraint stipulates that the manipulated variables fall within their practical operable range. This ensures that the control system consistently operates at near-minimum energy consumption while meeting capture rate requirements. Under dynamic disturbances, the EP value (energy performance index) of this invention can be stably controlled within a favorable range below 3.1 MJ / kg CO2, achieving a high degree of efficiency and economic efficiency.

[0077] The model predictive controller of the technical solution of the present invention sends the first optimized set value directly to the reboiler heat source regulating valve, achieving direct and optimized regulation of the energy consumption of the desorption process. The ratio control module receives the current inlet flue gas flow as a feedforward signal and adjusts the actual lean solution flow by adjusting the speed of the lean solution circulation pump or the opening of the outlet valve. The flow ratio between the actual lean solution flow and the current inlet flue gas flow can track the second optimized set value issued by the model predictive controller. Even when the flue gas flow fluctuates violently by up to ±15%, the CO2 capture rate can still be stabilized near the target value, with small overshoot and short adjustment time. This is far superior to traditional distributed PI control schemes, which often experience large capture rate fluctuations and longer recovery times under similar disturbances.

[0078] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. A post-combustion carbon capture hybrid control system comprising a post-combustion carbon capture device and a control unit communicatively coupled to the post-combustion carbon capture device, the post-combustion carbon capture device comprising: The post-combustion carbon capture device further comprises a buffer tank and a reboiler heat source regulating valve, the control unit comprises a distributed control module, a centralized control module and a ratio control module, the buffer tank is used for storing and stabilizing the lean solution to be introduced into the absorption tower, the reboiler heat source regulating valve is used for adjusting the steam supply according to the first optimization set value, the distributed control module is used for locally stabilizing the lean solution in the buffer tank, the centralized control module is used for optimizing the global performance of the post-combustion carbon capture device according to the performance parameter measurement value of the post-combustion carbon capture device and the pre-established target function, to obtain the first optimization set value and the second optimization set value, and the ratio control module is used for adjusting the flow ratio control of the lean solution and the flue gas according to the second optimization set value of the centralized control module, the flue gas flow measurement value and the lean solution flow measurement value.

2. A post-combustion carbon capture hybrid control system according to claim 1, wherein, The buffer tank is arranged on the lean solution circulation pipeline between the lean solution outlet of the lean-lean heat exchanger and the lean solution inlet of the absorption tower, and is used for storing the lean solution to be introduced into the absorption tower and maintaining the stability of the liquid level, temperature and solution concentration of the lean solution to be introduced into the absorption tower.

3. A post-combustion carbon capture hybrid control system according to claim 2, wherein, The distributed control module comprises a liquid level control loop, a temperature control loop and a concentration control loop connected with the pipeline of the buffer tank, the liquid level control loop comprises a liquid level sensor, a liquid level controller and a makeup water regulating valve, the liquid level sensor is used for transmitting the measured liquid level signal in the buffer tank to the liquid level controller, the liquid level controller is used for calculating an output liquid level signal according to the deviation of the measured liquid level signal in the buffer tank and the set liquid level value, and adjusting the opening degree of the makeup water regulating valve to maintain the constant liquid level in the buffer tank; The temperature control loop comprises a temperature sensor, a temperature controller and a coolant flow regulating valve, the temperature sensor is used for transmitting the measured solution temperature signal in the buffer tank to the temperature controller, the temperature controller is used for calculating an output solution temperature signal according to the deviation of the measured solution temperature signal in the buffer tank and the set temperature value, and adjusting the opening degree of the coolant regulating valve to control the temperature of the lean solution in the buffer tank; The concentration control loop comprises an online concentration analyzer, a concentration controller and a makeup amine regulating valve, the online concentration analyzer is used for transmitting the measured solution concentration signal in the buffer tank to the concentration controller, the concentration controller is used for calculating an output solution concentration signal according to the deviation of the measured solution concentration signal in the buffer tank and the set concentration value, and adjusting the opening degree of the makeup amine regulating valve to compensate for the degradation loss of the solution and maintain the stability of the solution concentration.

4. The post-combustion carbon capture hybrid control system of claim 1, wherein, The centralized control module comprises a model predictive controller, controlled variables of the model predictive controller at least comprising: a measured CO2 capture rate at a desorption column purified flue gas outlet, a reboiler liquid phase temperature in a reboiler; manipulated variables of the model predictive controller comprising: a first optimized set value supplied to the reboiler, a second optimized set value provided to a ratio control module; wherein the first optimized set value is a set value of the reboiler heat load, and the second optimized set value is a set value of a ratio of the molar flow of the lean solution to the inlet flue gas.

5. A post-combustion carbon capture hybrid control system according to claim 4, wherein, The performance of the post-combustion carbon capture device is optimized and controlled according to the measured values of the performance parameters of the post-combustion carbon capture device and a pre-established target function, and the first optimized set value and the second optimized set value specifically comprise: In a preset sampling period, actual measured values of the controlled variables and the inlet flue gas flow as a disturbance variable are collected; At each control time, according to a dynamic state space model and a current state variable, the influence trajectory of different manipulated variable sequences on the state variable in a future prediction time domain is predicted; A target function is established based on the dynamic state space model and the current state variable; The target function is solved based on a preset constraint condition to find a set of optimal manipulated variable control sequences in the control time domain, and the manipulated variable control sequences minimize the deviation of the predicted controlled variable trajectory from the respective set value; The first action in the optimal manipulated variable control sequence is output as the control instruction at the current time.

6. A post-combustion carbon capture hybrid control system according to claim 5, wherein the pre-set The constraint condition comprises a hard constraint, a soft constraint and an input constraint, the hard constraint is that the predicted value of the CO2 capture rate cannot be lower than a preset minimum environmental protection threshold at any time, the soft constraint is that the predicted value of the unit CO2 capture energy consumption does not exceed a maximum energy consumption threshold, and the input constraint is that the manipulated variable satisfies an actual operable range.

7. A post-combustion carbon capture hybrid control system according to claim 5, wherein, The target function is specifically: ， wherein J is a target function value, is the controlled variable prediction value at the time instant; is the controlled variable set value at the time instant; is the manipulated variable change amount; Q, R are weight matrices; M is the total number of time instants; ， ， wherein, is the state variable prediction value at the time instant is the state variable measurement value at the time instant is the manipulated variable actual value at the time instant is the controlled variable prediction value at the time instant is the disturbance variable measurement value at the time instant is the manipulated variable actual value at the time instant is the controlled variable prediction value at the time instant is the disturbance variable measurement value at the time instant is the manipulated variable actual value at the time instant is the disturbance variable measurement value at the time instant F, B, C are system matrix, input matrix, output matrix respectively; E is the influence matrix of the disturbance variable on the state variable.

8. A post-combustion carbon capture hybrid control system according to claim 5, wherein, The dynamic state space model comprises an absorption tower model, a desorption tower model, a lean-rich liquid heat exchanger model and a buffer tank model, wherein the absorption tower model and the desorption tower model are one-dimensional dynamic packed tower models, the one-dimensional dynamic packed tower model comprises mass, component and energy conservation equations; the lean-rich liquid heat exchanger is a one-dimensional counterflow heat exchange model; and the buffer tank model comprises a heat balance equation and an amine solution concentration conservation equation. The total mass conservation equation is: ， where j denotes the phase, G for gas and L for liquid; is the j-phase density; is the column cross-sectional area; is the j-phase linear velocity; is the column axial coordinate; is the molar flux of component i; is the molar mass; the symbol ± indicates the positive or negative direction of mass transfer between phases; The component mass conservation equation is: ， wherein, is the mass concentration of component i in phase j; is the stoichiometric coefficient of component i in the chemical reaction; is the reaction rate; The heat balance equation is: ， wherein, Cp is the constant pressure specific heat capacity; Tj is the temperature of phase j; Uij is the interphase average heat transfer coefficient; a is the specific surface area; ΔH is the reaction enthalpy; Lpi is the latent heat of phase change of component i; Tgi is the temperature of the gas-liquid interface; The one-dimensional counterflow heat exchange model is: ， ， wherein, is the hot fluid temperature; is the cold fluid temperature; is the mass flow rate of the hot fluid; is the mass flow rate of the cold fluid; is the specific heat capacity of the hot fluid; is the specific heat capacity of the cold fluid; is the overall heat transfer coefficient of the lean-rich exchanger; The heat balance equation is: ， wherein, Tb is the liquid temperature in the buffer tank; Tin is the temperature of the lean liquid entering the buffer tank; Tw is the temperature of the cooling water; Vb is the volume of the buffer tank; Vb is the volume of the buffer tank; F is the feed flow rate; 2 is the product of the heat transfer coefficient of the cooling coil and the area of the cooling coil; The amine solution concentration conservation equation is: ， wherein, is the amine concentration in the buffer tank; is the amine concentration of the feed; is the amine concentration of the discharge; is the volumetric flow of the feed amine; is the volumetric flow of the discharge amine.

9. A post-combustion carbon capture hybrid control system according to claim 4, wherein, The post-combustion carbon capture device further comprises a temperature control module, the model predictive controller directly sends the first optimized set value to the temperature control module, the temperature control module controls a reboiler heat source adjusting valve according to the first optimized set value to realize direct and optimized adjustment of the desorption process energy consumption; the ratio control module receives the current inlet flue gas flow as a feedforward signal, adjusts and controls the actual lean solution flow by adjusting the rotation speed of a lean solution circulating pump or the opening degree of an outlet valve, so that the flow ratio between the actual lean solution flow and the current inlet flue gas flow can track the second optimized set value issued by the model predictive controller.

10. A hybrid control method for post-combustion carbon capture, characterized by, Based on the post-combustion carbon capture hybrid control system according to any one of claims 1-9, a method for controlling the post-combustion carbon capture device is realized, comprising: The distributed control module performs local stabilizing control on the lean solution state in the buffer tank; The centralized control module performs global performance optimization control on the post-combustion carbon capture device according to the performance parameter measurement value of the post-combustion carbon capture device and a pre-established target function, to obtain a first optimization set value and a second optimization set value; The reboiler heat source regulating valve adjusts the steam supply amount according to the first optimization set value; The ratio control module adjusts the flow ratio control of the lean solution and the flue gas according to the second optimization set value of the centralized control module, the inlet flue gas flow measurement value and the lean solution flow measurement value.

Images