Industrial flue gas cascade carbon capture system and method based on multi-parameter characteristic analysis

Abstract

The invention provides an industrial flue gas cascade carbon capture system and method based on multi-parameter characteristic analysis. According to the industrial flue gas cascade carbon capture system, four-dimensional characteristic parameters of COconcentration, flow, temperature and impurity components in flue gas are monitored in real time, and a mixed decision model is constructed to dynamically generate 60 optimal combinations of four processes of membrane separation, adsorption, absorption and liquefaction; the system provided by the invention adopts a modular rapid recombination design, supports a process chain to complete switching within 30 minutes, and establishes a three-stage energy collaborative network of waste heat gradient utilization, cold energy circulation and pressure energy recovery, so that the trapping energy consumption is reduced by 38%-52%; according to the method, the concentration adaptability limitation of a single technology is broken through, the CO concentration range of 5%-95% is effectively covered, the treatment efficiency and economical efficiency of complex industrial flue gas are remarkably improved, and the method is suitable for meeting the carbon emission reduction requirements in the fields of electric power, steel, chemical engineering and the like.

Description

Technical Field

[0001] The present invention belongs to the technical field of carbon capture and storage, and specifically relates to an industrial flue gas cascade carbon capture system and method based on multi-parameter characteristic analysis. Background Art

[0002] Carbon capture, utilization and storage (CCUS) is an important means of large-scale low-carbon utilization of fossil energy. High energy consumption and high costs are the main obstacles to the large-scale promotion of CCUS technology. The cost of carbon capture accounts for 75% of its total chain cost. Reducing the cost of carbon capture is an international problem that needs to be urgently addressed in the development of CCUS technology.

[0003] Industrial flue gas is a major source of carbon emissions. Its complex composition and highly variable operating conditions pose significant challenges to carbon capture technology. Currently, the carbon capture technologies widely used in industry primarily include membrane separation (A), adsorption separation (B), absorption separation (C), and cryogenic liquefaction (D). However, significant technical bottlenecks remain in practical application. First, a single technology struggles to cover a wide range of CO2 concentrations (e.g., membrane separation efficiency drops by over 45% at concentrations <15%, while chemical absorption regeneration energy consumption increases 2.3-fold at concentrations >60%). The narrow adaptability of a single process technology to CO2 concentrations leads to a significant imbalance between energy consumption and efficiency. Second, traditional processes suffer from inefficient energy integration. Traditional cascade processes fail to achieve cascaded energy utilization across units (e.g., the desorption waste heat from the adsorption unit at 80-120°C is not thermally coupled with the regeneration steam from the absorption unit at 100-140°C). Furthermore, the residual pressure (0.3-0.8 MPa) on the permeate side of membrane separation is not recovered. Furthermore, the traditional process chain energy network design has defects. A cross-process thermodynamic coupling model has not been established (such as the cascade utilization of thermal energy in units B→C), and there is a lack of collaborative optimization algorithms for energy flow and material flow. Therefore, the traditional fixed process chain design is difficult to adapt to large fluctuations in flue gas flow, resulting in flow matching failure.

[0004] Furthermore, existing combined systems lack the ability to dynamically perceive flue gas characteristics. Static process combination designs are unable to reduce capture costs and efficiently integrate energy. They also cannot implement process switching under concentration grading, such as preferential adsorption enrichment in low-concentration areas, membrane separation and enrichment in medium-concentration areas, and low-temperature liquefaction and separation in high-concentration areas. Furthermore, they cannot modularly increase or decrease capacity to cope with flow fluctuations. Therefore, there is an urgent need to develop a new carbon capture system to efficiently capture carbon dioxide from industrial flue gas. Summary of the Invention

[0005] In response to some shortcomings in the existing technology, the present invention provides an industrial flue gas cascade carbon capture system and method based on multi-parameter characteristic analysis; the industrial flue gas cascade carbon capture system of the present invention monitors the four-dimensional characteristic parameters of CO2 concentration, flow, temperature and impurity components in the flue gas in real time, constructs a hybrid decision model to dynamically generate 60 optimized combinations of four types of processes: membrane separation, adsorption, absorption and liquefaction; the system of the present invention adopts a modular rapid reorganization design, supports the process chain to complete switching within 30 minutes, and establishes a three-level energy collaborative network of waste heat cascade utilization, cold energy circulation and pressure energy recovery, which reduces the capture energy consumption by 38% to 52%; the present invention breaks through the concentration adaptability limitation of a single technology, effectively covers the CO2 concentration range of 5% to 95%, significantly improves the treatment efficiency and economy of complex industrial flue gas, and is suitable for carbon emission reduction needs in the fields of electricity, steel, chemical industry, etc.

[0006] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0007] The present invention first provides an industrial flue gas cascade carbon capture system based on multi-parameter feature analysis, the industrial flue gas cascade carbon capture system includes a feature perception unit, an intelligent decision-making unit and a modular execution unit;

[0008] The characteristic sensing unit is used to monitor the CO2 concentration, gas flow, temperature and impurity components in the industrial flue gas in real time, and feed back the monitored characteristic parameters to the intelligent decision-making unit;

[0009] The intelligent decision-making unit has a built-in hybrid decision-making model, which dynamically generates a set of carbon capture process combinations based on characteristic parameters and calculates the applicability index of each combination. The carbon capture process includes a combination of at least two processes selected from membrane separation, adsorption, absorption, and liquefaction.

[0010] The modular execution unit includes a standardized process module and a rapid reorganization interface, and the modular execution unit is assembled according to the results generated by the intelligent decision-making unit; the standardized process module includes a membrane separation unit, an adsorption unit, an absorption unit and a liquefaction unit.

[0011] Preferably, the feature sensing unit includes a quantum cascade laser spectrometer, an ultrasonic flowmeter, a micro gas chromatograph, an infrared thermal imager, a CO2 concentration sensor, a temperature sensor and a pressure sensor.

[0012] Preferably, the hybrid decision model is constructed based on industrial flue gas characteristics, and the hybrid decision model includes:

[0013] (1) A dynamic decision library of 60 process combinations is formed based on the four-dimensional characteristic matrix of “concentration-flow-impurity-temperature”;

[0014] (2) Establish a multi-objective optimization function through dynamic optimization control and real-time weight distribution;

[0015] (3) Evaluate the process combination effectiveness through fuzzy comprehensive evaluation method and calculate the applicability index of each process combination;

[0016] (4) The operator may select an execution plan based on preset rules.

[0017] Preferably, in step (1), the process of forming the dynamic decision library includes:

[0018] (a) Concentration classification:

[0019] Low concentration range: CO2<15%, treated by membrane separation unit and / or adsorption unit;

[0020] Medium concentration range: 15% ≤ CO2 ≤ 40%, treated by a combination of at least two of the following: membrane separation unit, adsorption unit and absorption unit;

[0021] High concentration range: CO2>40%, using absorption unit and / or liquefaction unit for treatment;

[0022] (b) Flow correction factor:

[0023] When flow rate Q≥20,000Nm 3 / h, parallel membrane separation units are mandatory to reduce pressure drop;

[0024] When Q≤1,000Nm 3 / h, activate the intermittent operation mode of the liquefaction unit;

[0025] (c) Temperature compensation mechanism:

[0026] When the industrial flue gas temperature T≥150℃, select membrane separation unit and / or adsorption unit for treatment;

[0027] When T≤50℃, use the absorption unit and / or liquefaction unit for treatment;

[0028] Based on the above classification rules, a dynamic decision library containing 60 process combinations is formed.

[0029] Preferably, in step (2), the general formula of the multi-objective function is:

[0030]

[0031] Where: αi is the process weight coefficient (0≤α i ≤1), Ei is the unit energy consumption of each process (GJ / t CO2), η is the switching cost factor (0.02-0.05GJ / time), n switchis the number of switching modules;

[0032] The objective function weights (α, β) directly affect the economic ranking of process combinations.

[0033] Preferably, the multi-objective function is optimized as follows:

[0034] min E total =α·E A +β·E B +γ·E C +δ·E D +η·E switch

[0035] The constraints are: C out ≥95%; △P≤15kPa; T regeneration ≤140℃;

[0036] Among them, A is the membrane separation unit, B is the adsorption unit, C is the absorption unit, and D is the liquefaction unit;

[0037] α / β / γ / δ are process energy consumption weight coefficients, which are dynamically adjusted according to the flue gas characteristics; E total is the total energy consumption of the system; E A 、E B 、E c 、E D are the unit energy consumption of process A, process B, process C, and process D respectively; η is the equipment switching cost factor; E switch The switching energy consumption caused by the process chain reorganization (positively correlated with the number of modules); C out is the CO2 product gas purity constraint; ΔP is the system pressure drop constraint; T regeneration Solvent regeneration temperature constraint (to prevent thermal degradation).

[0038] Preferably, the formula for the combined energy efficiency evaluation is:

[0039]

[0040]

[0041]

[0042] In the formula, S i is the applicability index of the i-th process combination; w1, w2, w3 are weight coefficients (default values ​​are 0.4 / 0.4 / 0.2); E i is the specific energy consumption of combination i; η CO2 is the CO2 capture rate; C OPEX is the annualized operating cost,

[0043] That is: S i=0.4·(1 / E i )+0.4·η CO2 +0.2·(1 / C OPEX ).

[0044] Preferably, the modular execution units are assembled according to the rules of the energy coordination network:

[0045] When both adsorption unit and absorption unit are included, the desorption waste heat of the adsorption unit is supplied to the regeneration tower of the absorption module through the heat exchanger;

[0046] When both a liquefaction unit and an adsorption unit are included, the cold energy of the expander of the liquefaction unit can be used to pre-cool the inlet airflow of the adsorption unit;

[0047] When a membrane separation unit is included, the residual pressure on the permeate side drives the turbine generator set.

[0048] The present invention also provides a method for cascade carbon capture of industrial flue gas, which is implemented based on the above-mentioned cascade carbon capture system for industrial flue gas and comprises:

[0049] (1) The industrial flue gas is passed into the feature sensing unit, and the four-dimensional feature parameters of the industrial flue gas are collected in real time through the feature sensing unit: CO2 concentration, flow rate, temperature and impurity vector, and a four-dimensional feature matrix of "concentration-flow rate-impurity-temperature" is constructed;

[0050] (2) Input the four-dimensional characteristic matrix of “concentration-flow-impurity-temperature” into the intelligent decision-making unit, generate a set of candidate process combinations by the hybrid decision model of the intelligent decision-making unit, and calculate the applicability index of each combination;

[0051] (3) According to the combination with the highest applicability index, the modular units are driven to complete physical reorganization, perform energy collaborative scheduling, and obtain modular execution units;

[0052] (4) The industrial flue gas is passed into the modular execution unit for carbon capture.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] (1) Compared with the existing single carbon capture process with narrow concentration adaptability and poor impurity tolerance, the adaptability of the industrial flue gas cascade carbon capture method of the present invention is significantly improved. The technical effect is: full concentration coverage: through the four processes of dynamic combination membrane separation (A), adsorption (B), absorption (C), and liquefaction (D), it effectively covers industrial flue gas scenarios with CO2 concentrations of 5% to 95% (traditional technology only covers 15% to 60%).

[0055] (2) The flow adaptability of the industrial flue gas cascade carbon capture method of the present invention is enhanced, and it supports a flow fluctuation range of ±40% (traditional systems only have ±20%). Through modular parallel design, when the flow rate is >20,000Nm 3 / h when the pressure drop is reduced to ≤8kPa (the pressure drop of traditional fixed design is ≥15kPa).

[0056] (3) The industrial flue gas cascade carbon capture method of the present invention has the advantage of a breakthrough in impurity tolerance, with its H2S tolerance increased to 200ppm (the adsorption capacity of traditional adsorbents decreases by 40% at 50ppm); the dust load allowable value reaches 50mg / Nm 3 (The traditional membrane system is 20mg / Nm 3 Cleaning is required daily).

[0057] (4) The industrial flue gas cascade carbon capture method of the present invention solves the defect of the existing technology of serious waste of waste heat / cold energy / pressure energy. The process energy consumption efficiency of the present invention is reduced, and the system specific energy consumption is optimized through the energy synergy network (waste heat cascade utilization, cold energy circulation, pressure energy recovery). Through cascade utilization, the waste heat utilization rate is improved. The energy consumption of the refrigeration system is reduced by the cold energy circulation efficiency enhancement. In addition, the dynamic response capability of the present invention has a great breakthrough, the process switching time is optimized, and the operation and maintenance efficiency is improved. The parameter adjustment response and adaptive control can respond quickly to ensure that the system is always in a quasi-optimal state.

[0058] (5) The modularized industrial flue gas cascade carbon capture method of the present invention reduces costs and increases efficiency through intelligent operation and maintenance. The modular design of the equipment reduces initial investment and saves energy consumption in operation and maintenance costs, effectively solving the problem of high investment and long payback period in existing technologies. In addition, the technical compatibility and scalability of the present invention are strong, the process combination is highly flexible, and it supports dynamic configuration of 60 process sequences (traditional systems only have 3 to 5 fixed ones), adapting to future technological iterations and supporting seamless integration with the park energy management system.

[0059] Through the three-level collaborative innovation of "feature perception-intelligent decision-making-dynamic execution", this invention has achieved comprehensive breakthroughs in process adaptability, energy efficiency, economy and other dimensions, providing a scalable and highly flexible solution for industrial carbon capture, and its comprehensive performance indicators have reached the international leading level.

[0060] (6) The present invention uses a data fusion module to calibrate multi-sensor data, which can eliminate measurement deviations between the QCL spectrometer and the μGC (such as CO2 concentration errors caused by temperature drift). Furthermore, through characteristic parameter correlation modeling, a four-dimensional correlation matrix of CO2 concentration, flow rate, temperature, and impurity composition is established to extract the main controlling factors and support the weight allocation of the decision-making unit. Furthermore, it can also identify sensor failures or sudden changes in operating conditions in real time, isolate abnormal data, and trigger the system self-repair protocol.

[0061] The decision-making unit in the system described in the present invention includes two algorithms: a multi-objective optimization function and a combined energy efficiency evaluation. A collaborative decision-making mechanism is adopted. On the one hand, a parallel-screening architecture, i.e., multi-objective optimization, is used to generate a Pareto solution set (60 process combinations). The combined energy efficiency evaluation is used to further screen the solutions. Ultimately, the execution solution is selected by the operator or preset rules (such as the minimum carbon tax cost). On the other hand, through adaptive weight adjustment, during peak electricity price periods, the operating cost weight in the objective function gives priority to low-power consumption processes (such as adsorption over liquefaction). BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a schematic diagram of the system architecture.

[0063] Figure 2 It is a flow chart of the hybrid decision model.

[0064] Figure 3 It is a detailed structural diagram of the modular execution unit.

[0065] Figure 4 It is the energy collaborative network topology diagram.

[0066] Figure 5 It is a dynamic control timing diagram. DETAILED DESCRIPTION

[0067] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the protection scope of the present invention is not limited thereto.

[0068] Example 1: Construction of an industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis

[0069] The schematic diagram of the industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis of the present invention is as follows: Figure 1 As shown, the industrial flue gas cascade carbon capture system based on multi-parameter feature analysis includes a feature perception unit, an intelligent decision-making unit and a modular execution unit;

[0070] The characteristic sensing unit is used to monitor the CO2 concentration, gas flow, temperature and impurity components in the industrial flue gas in real time, and feed back the monitored characteristic parameters to the intelligent decision-making unit;

[0071] The intelligent decision-making unit has a built-in hybrid decision-making model, which dynamically generates a set of carbon capture process combinations based on characteristic parameters and calculates the applicability index of each combination. The carbon capture process includes a combination of at least two processes selected from membrane separation, adsorption, absorption, and liquefaction.

[0072] The modular execution unit includes a standardized process module and a rapid reorganization interface, and the modular execution unit is assembled according to the results generated by the intelligent decision-making unit; the standardized process module includes a membrane separation unit, an adsorption unit, an absorption unit and a liquefaction unit.

[0073] Preferably, the feature sensing unit includes a quantum cascade laser spectrometer, an ultrasonic flowmeter, a micro gas chromatograph and an infrared thermal imager.

[0074] In some embodiments, the hybrid decision model is constructed based on industrial flue gas characteristics, and the hybrid decision model includes:

[0075] (1) A dynamic decision library of 60 process combinations is formed based on the four-dimensional characteristic matrix of “concentration-flow-impurity-temperature”;

[0076] (6) Establish a multi-objective optimization function through dynamic optimization control and real-time weight distribution;

[0077] (7) The process combination performance is evaluated by fuzzy comprehensive evaluation method, and the applicability index of each process combination is calculated.

[0078] In some embodiments, the process of forming the dynamic decision library includes:

[0079] (1) Concentration classification:

[0080] Low concentration range: CO2<15%, treated by membrane separation unit and / or adsorption unit;

[0081] Medium concentration range: 15% ≤ CO2 ≤ 40%, treated by a combination of at least two of the following: membrane separation unit, adsorption unit and absorption unit;

[0082] High concentration range: CO2>40%, using absorption unit and / or liquefaction unit for treatment;

[0083] (2) Flow correction factor:

[0084] When flow rate Q≥20,000Nm 3 / h, parallel membrane separation units are mandatory to reduce pressure drop;

[0085] When Q≤1,000Nm 3 / h, activate the intermittent operation mode of the liquefaction unit;

[0086] (3) Temperature compensation mechanism:

[0087] When the industrial flue gas temperature T≥150℃, select membrane separation unit and / or adsorption unit for treatment;

[0088] When T≤50℃, use the absorption unit and / or liquefaction unit for treatment;

[0089] Based on the above classification rules, a dynamic decision library containing 60 process combinations is formed.

[0090] In some embodiments, the general formula of the multi-objective function is:

[0091]

[0092] Where: αi is the process weight coefficient (0≤α i ≤1), Ei is the unit energy consumption of each process (GJ / t CO2), η is the switching cost factor (0.02-0.05GJ / time), n switch is the number of switching modules.

[0093] In some embodiments, the multi-objective optimization function is:

[0094] min E total =α·E A +β·E B +γ·E C +δ·E D +η·E switch

[0095] The constraints are: C out 95%; △PP≤15kPa; T regeneration ≤140℃;

[0096] Among them, A is the membrane separation unit, B is the adsorption unit, C is the absorption unit, and D is the liquefaction unit;

[0097] α / β / γ / δ are process energy consumption weight coefficients, which are dynamically adjusted according to the flue gas characteristics; E total is the total energy consumption of the system; E A 、E B 、E c 、E D are the unit energy consumption of process A, process B, process C, and process D respectively; η is the equipment switching cost factor; E switch The switching energy consumption caused by the process chain reorganization (positively correlated with the number of modules); C out is the CO2 product gas purity constraint; ΔP is the system pressure drop constraint; T regeneration Solvent regeneration temperature constraint (to prevent thermal degradation).

[0098] In some embodiments, the formula for the combined energy efficiency evaluation is:

[0099]

[0100] Normalization processing:

[0101]

[0102] In the formula, S i is the applicability index of the i-th process combination; w1, w2, w3 are weight coefficients (default values ​​are 0.4 / 0.4 / 0.2); E i is the specific energy consumption of combination i; η CO2 is the CO2 capture rate; C OPEX is the annualized operating cost,

[0103] That is: S i =0.4·(1 / E i )+0.4·η CO2 +0.2·(1 / C OPEX ).

[0104] In some embodiments, modular execution units are assembled according to the rules of the energy coordination network:

[0105] When both adsorption unit and absorption unit are included, the desorption waste heat of the adsorption unit is supplied to the regeneration tower of the absorption module through the heat exchanger;

[0106] When both a liquefaction unit and an adsorption unit are included, the cold energy of the expander of the liquefaction unit can be used to pre-cool the inlet airflow of the adsorption unit;

[0107] When a membrane separation unit is included, the residual pressure on the permeate side drives the turbine generator set.

[0108] This embodiment also provides a method for carbon capture using the above-mentioned industrial flue gas cascade carbon capture system, the specific steps of which include:

[0109] (1) The industrial flue gas is passed into the feature sensing unit, and the four-dimensional feature parameters of the industrial flue gas are collected in real time through the feature sensing unit: CO2 concentration, flow rate, temperature and impurity vector, and a four-dimensional feature matrix of "concentration-flow rate-impurity-temperature" is constructed;

[0110] (2) Input the four-dimensional characteristic matrix of “concentration-flow-impurity-temperature” into the intelligent decision-making unit, generate a set of candidate process combinations by the hybrid decision model of the intelligent decision-making unit, and calculate the applicability index of each combination;

[0111] (3) According to the combination with the highest applicability index, the modular units are driven to complete physical reorganization, perform energy collaborative scheduling, and obtain modular execution units;

[0112] (4) The industrial flue gas is passed into the modular execution unit for carbon capture.

[0113] Figure 2This is a flowchart of a hybrid decision-making model used to optimize carbon capture from industrial flue gas. The flowchart is divided into five main layers: input layer, rule screening layer, neural network prediction layer, dynamic optimization layer, and output layer.

[0114] Figure 3 This is a detailed diagram of the modular execution unit, showing the specific construction and component arrangement of the unit in the carbon capture process. The detailed diagram is divided into three main parts: the membrane separation module (A), the adsorption module (B), and the quick-connect interface.

[0115] Figure 4 This is a topological diagram of the energy synergy network. It shows that the network consists of three main energy recovery pathways: a thermal coupling pathway, a cooling energy circulation pathway, and a pressure energy recovery pathway. Each pathway involves specific energy conversion and transfer processes to ensure efficient energy utilization in the system.

[0116] Figure 5 It is a dynamic control timing diagram. From the diagram, we can see the timeline of the entire control process, and the entire process is divided into three main stages: data acquisition stage, decision calculation stage and execution stage.

[0117] Example 2: Treatment of low-concentration flue gas from a coal-fired power plant

[0118] This embodiment uses the industrial flue gas cascade carbon capture system described in Example 1 to treat low-concentration flue gas from a coal-fired power plant. The specific steps are as follows:

[0119] (1) The low-concentration flue gas from a coal-fired power plant was passed into the characteristic sensing unit of the industrial flue gas cascade carbon capture system to obtain the four-dimensional characteristic parameters of the flue gas: CO2 concentration of 12.5%, flow rate of 38,000 Nm 3 / h, temperature 130℃, impurities NOx 65ppm, dust 8mg / Nm 3 .

[0120] (2) The feature sensing unit detects low concentration and high flow characteristics, triggering the rule engine to select the adsorption pre-enrichment strategy. The intelligent decision-making unit generates the optimized combination of adsorption (B) → membrane separation (A) → absorption (C)

[0121] Multi-objective optimization function

[0122] The optimization function can be expressed as:

[0123] F=w1·E+w2·C+w3·(1-η CO2 )

[0124] in:

[0125] F is the comprehensive optimization goal;

[0126] w1, w2, and w3 are weight coefficients, which respectively indicate the importance of energy consumption, cost, and CO2 capture rate in the optimization objectives;

[0127] E is the specific energy consumption, which indicates the energy required to treat unit mass of CO2;

[0128] C is the annual operating cost;

[0129] η CO2 is the CO2 capture rate.

[0130] Constraints:

[0131] The constraints that need to be met during the optimization process include:

[0132] Energy consumption constraint: E≤E max

[0133] Cost constraint: C≤C max

[0134] CO2 capture rate constraint: η CO2 ≥η min

[0135] Among them, E max , C max , η min are the maximum acceptable and minimum required values ​​for energy consumption, cost and CO2 capture rate respectively. The calculation process of the applicable index is as follows:

[0136] Collect data on energy consumption, cost and CO2 capture rate for each process combination.

[0137] Specific energy consumption: E = 1.9GJ / tCO2

[0138] Annual operating cost: C = 2.85 million yuan

[0139] CO2 capture rate: η CO2 =96.2%

[0140] The collected data is normalized to facilitate comparison between different dimensions. The normalization formula is as follows: E′=EE min / (E max -E min )

[0141] C′=CC min / (C max -C min )

[0142] η′ CO2 =η CO2 -η min / (η max -η min )

[0143] Among them, E min , C min , η min and E max , C max , η max are the minimum and maximum values ​​of energy consumption, cost and CO2 capture rate respectively. Based on the normalized data, the applicability index of each process combination is calculated. The calculation formula of the applicability index can be expressed as: S = w1·E′+w2·C′+w3·(1-η′ CO2 )

[0144] Among them, S is the applicable index; w1, w2, w3 are weight coefficients, which are set according to actual needs.

[0145] Assume that the weight coefficients are w1 = 0.4, w2 = 0.4, and w3 = 0.2, and the minimum and maximum values ​​of energy consumption, cost, and CO2 capture rate are known. Based on the collected data, normalization processing and applicable index calculation are performed:

[0146] Normalization processing:

[0147] E′=1.9-E min / (E max -E min )

[0148] C′=285-C min / (C max -C min )

[0149] η′ CO2 =96.2-η min / (η max -η min )

[0150] Applicable index calculation:

[0151] S=0.4·E′+0.4·C′+0.2·(1-η′ CO2 )

[0152] Through the above steps, the applicability index of the combination of adsorption (B) → membrane separation (A) → absorption (C) can be calculated and compared with other combinations, and finally the combination with the highest applicability index is selected as the optimal combination.

[0153] (3) The physical reorganization is completed by driving the modular units according to the combination with the highest applicability index. Module B is a MOF-303 adsorption tower with a loading capacity of 3.5t, an adsorption pressure of 0.8MPa, and a cycle of 8min; Module A is a polyimide membrane assembly with an operating pressure of 2.8MPa and a membrane area of ​​600m 2Module C is MDEA-PZ composite solvent, with a circulation volume of 12m 3 / h, regeneration temperature 115℃.

[0154] (4) Energy coordination is implemented. The desorption waste heat (105°C) from module B is used to preheat the feed to the regeneration tower in module C, saving 23% of steam consumption. The residual pressure (0.6 MPa) on the permeate side of module A drives the turbine to generate electricity, with a power output of 42 kW.

[0155] (5) The industrial flue gas is passed into the modular execution unit for carbon capture, with a CO2 capture rate of 96.2%, a specific energy consumption of 1.9 GJ / tCO2 (44% lower than the traditional MEA process), and an annual operating cost of 2.85 million yuan (saving 35%).

[0156] Example 3: Treatment of sulfur-containing flue gas from a refinery

[0157] This example uses the industrial flue gas cascade carbon capture system described in Example 1 to treat sulfur-containing flue gas from an oil refinery. The specific steps are as follows:

[0158] (1) The sulfur-containing flue gas from the refinery was passed into the characteristic sensing unit of the industrial flue gas cascade carbon capture system to obtain the four-dimensional characteristic parameters of the flue gas: CO2 concentration of 18%, flow rate of 9,200 Nm 3 / h, temperature is 85℃, impurity is H2S180ppm, and pressure is 0.3MPa.

[0159] (2) The feature sensing unit detects that H2S exceeds the standard, triggering the liquefaction desulfurization priority strategy. The intelligent decision-making unit generates a combined path: liquefaction (D) → adsorption (B) → membrane separation (A).

[0160] Collect data on energy consumption, cost, CO2 capture rate, H2S removal rate, and solvent consumption for each process combination. For the liquefaction (D) → adsorption (B) → membrane separation (A) combination, the collected data are:

[0161] Energy consumption: E D , E B , E A

[0162] Cost: C D , C B , C A

[0163] CO2 capture rate: η CO2

[0164] H2S removal rate: η H2S

[0165] Solvent consumption: S

[0166] The collected data are normalized to eliminate the influence of different dimensions:

[0167] E′=EE min / (E max -E min )

[0168] C′=CC min / (C max -C min )

[0169] η′ CO2 =η CO2 -η min / (η max -η min )

[0170] η′ H2S =η H2S -η min / (η max , H2S -η min,H2S) )

[0171] S′=SS min / (S max -S min )

[0172] Among them, E min , C min , η min and E max , C max , η max are the minimum and maximum values ​​of energy consumption, cost and CO2 capture rate respectively. min ,C min ,η min,CO2 ,η min,H2S ,S min and E max ,C max ,η max,CO2 ,η max,H2S ,S max They are the minimum and maximum values ​​of energy consumption, cost, CO2 capture rate, H2S removal rate and solvent consumption, respectively.

[0173] According to the normalized data and the set weight coefficient, the applicability index of each process combination is calculated:

[0174] Si=w1·E′ i +w2·C′ i +w3·(1-η′ CO2,i )+w4·(1-η′ H2S,i )+w5·S′ i

[0175] Among them, S i is the applicability index of the i-th process combination.

[0176] Assume that the weight coefficients are w1 = 0.3, w2 = 0.2, w3 = 0.2, w4 = 0.2, and w5 = 0.1, and the minimum and maximum values ​​of energy consumption, cost, CO2 capture rate, H2S removal rate, and solvent consumption are known. Based on the collected data, normalization processing and applicable index calculation are performed:

[0177] E′=EE min / (E max -E min )

[0178] C′=CC min / (C max -C min )

[0179] η′ CO2 =η CO2 -η min / (η max -η min )

[0180] η′ H2S =η H2S -η min / (η max,H2 S -η min,H2S) )

[0181] S′=SS min / (S max -S min )

[0182] Applicable index calculation:

[0183] S=0.3·E′+0.2·C′+0.2·(1-η′ CO2 )+0.2·(1-η′ H2S )+0.1·S′

[0184] Through the above steps, the applicability index of the liquefaction (D) → adsorption (B) → membrane separation (A) combination can be calculated and compared with other combinations, and finally the combination with the highest applicability index is selected as the optimal combination.

[0185] (3) The modular units are physically reassembled based on the combination with the highest applicability index. Module D is a three-stage compression refrigeration unit (evaporation temperature -40°C) with a desulfurization efficiency of 99.5%. Module B is a copper-loaded molecular sieve with an adsorption temperature of -15°C and a breakthrough time of 45 minutes. Module A is a carbon molecular sieve membrane with a selectivity of CO2 / CH4 = 50.

[0186] (4) Energy coordination is implemented. The cooling energy (-45°C) of the expander in module D is used to pre-cool the intake air in module B to -10°C, reducing refrigeration energy consumption by 28%. The desorbed exhaust gas (0.2MPa) from module B drives the Venturi ejector, improving the negative pressure stability of the system.

[0187] (5) The industrial flue gas is passed into the modular execution unit for carbon capture, with a CO2 purity of 99.7%, H2S residue of <0.5ppm, and solvent consumption of 0.08kg / tCO2 (62% lower than the traditional process).

[0188] Example 4: Biomass gasification gas treatment

[0189] This embodiment uses the industrial flue gas cascade carbon capture system described in Example 1 to process biomass gasification gas. The specific steps are as follows:

[0190] (1) The biomass gasification gas is introduced into the characteristic sensing unit of the industrial flue gas cascade carbon capture system to obtain the four-dimensional characteristic parameters of the biomass gasification gas: CO2 concentration of 6.8%, flow rate of 1,200 Nm 3 / h, temperature is -20℃, impurity is CH438%, and pressure fluctuation is ±20%.

[0191] (2) The feature sensing unit detects that low concentration and high CH4 content trigger a resource recovery strategy. The intelligent decision-making unit generates a combined path: liquefaction (D) → membrane separation (A) → adsorption (B).

[0192] The constraints that need to be met during the optimization process include:

[0193] Energy consumption constraint: E≤E max

[0194] Cost constraint: C≤C max

[0195] CH4 recovery rate constraint: η CH4 ≥η CH4,min

[0196] Pressure fluctuation adaptive response time constraint: T s ≤T s,max

[0197] Normalize each target value:

[0198] E′=EE min / (E max -E min )

[0199] C′=CC min / (C max -C min )

[0200] η′ CH4 =η CH4 -η CH4,min / (η CH4,max -η CH4,min )

[0201] T′ s =T s -T s,min / (T s,max -T s,min )

[0202] Calculate the applicability index

[0203] According to the normalized target value and weight coefficient, the applicability index S of each process combination is calculated:

[0204] S=w1·E′+w2·C′+w3·(1-η′ CH4 )+w4·T′ s

[0205] (3) Physical reorganization is completed according to the modular units driven by the combination with the highest applicability index: Module D is cascade refrigeration (propane + CO2 working fluid), with a CH4 liquefaction rate of 92%, Module A is a double-layer composite membrane (PDMS / PES), with a CH4 recovery rate of 91%, and Module B is a KMF-1 adsorbent with a working pressure of 0.15 MPa.

[0206] (4) Energy coordination is implemented, with liquefaction cooling energy (-50°C) used to cool the membrane modules, improving separation selectivity by 15%. The CH4 compression heat (85°C) is reused for adsorbent regeneration.

[0207] (5) The industrial flue gas is passed into the modular execution unit for carbon capture. The implementation effect is: CH4 recovery rate is 93.5%, CO2 capture cost is ￥152 / t (41% lower than the single process), and the pressure fluctuation adaptive response time is ≤2 minutes.

[0208] Example 5: Treatment of high-dust flue gas in a steel plant

[0209] This example uses the industrial flue gas cascade carbon capture system described in Example 1 to treat high-dust flue gas from a steel plant. The specific steps are as follows:

[0210] (1) The high-dust flue gas from the steel plant was introduced into the characteristic sensing unit of the industrial flue gas cascade carbon capture system to obtain the four-dimensional characteristic parameters of the high-dust flue gas from the steel plant: CO2 concentration of 22%, flow rate of 15,000 Nm 3 / h, temperature 180℃, impurity dust 42mg / Nm 3 , humidity 8%.

[0211] (2) The feature sensing unit detects that the dust load triggers the pretreatment mechanism. The intelligent decision-making unit generates a combined path: electrostatic precipitation → membrane separation (A) → adsorption (B) → absorption (C).

[0212] Constraints

[0213] The constraints that need to be met during the optimization process include:

[0214] Energy consumption constraint: E≤E max

[0215] Cost constraint: C≤C max

[0216] CO2 capture rate constraint: η CO2 ≥η min,CO2

[0217] System pressure drop constraint: ΔP≤ΔPmax

[0218] Module switching time constraint: T s ≤T max,s

[0219] Determine the weight coefficient: Determine the weight coefficient w1, w2, w3, w4, w5w1, w2, w3, w4, w5 of each target based on actual needs and priorities. Normalization: Normalize each target value to facilitate comparison and calculation.

[0220] E′=EE min / (E max -E min )

[0221] C′=CC min / (C max -C min )

[0222] η′ CO2 =η CO2 -η CO2,min / (η CO2,max -η CO2,min )

[0223] Δ′P=ΔP-ΔP min / (ΔP max -ΔP min )

[0224] T′ s =T s -T s,min / (T s,max -T s,min )

[0225] Calculate the applicability index: Calculate the applicability index SS of each process combination based on the normalized target value and weight coefficient.

[0226] S=w1·E′+w2·C′+w3·(1-η′ CO2 )+w4·ΔP′+w5·T′ s

[0227] (3) Physical reorganization was achieved by driving modular units based on the combination with the highest applicability index: Module A: Ceramic membrane dust removal unit (filtration accuracy 0.5 μm), pressure drop ≤ 3 kPa. Module B: Moisture-resistant MOF-303, loading height 1.5 m, adsorption capacity 5.2 mmol / g. Module C: Phase-change AMP solvent, regeneration energy consumption 2.3 GJ / tCO2.

[0228] (4) Implement energy coordination and dispatch, and use the flue gas waste heat (180℃) for solvent regeneration tower insulation, reducing external steam supply by 15%. The residual pressure power generation on the permeate side is connected to the plant grid, with an annual power supply of 1.8×10 5 kWh.

[0229] (5) The industrial flue gas is passed into the modular execution unit for carbon capture. The implementation effect is: the dust removal rate is 99.9% (the outlet concentration is <0.1mg / Nm 3 ), system pressure drop: 12kPa (traditional process ≥22kPa), module switching time: 25 minutes (including dust removal unit activation).

[0230] In summary, the industrial flue gas cascade carbon capture system described in the present invention monitors the four-dimensional characteristic parameters of CO2 concentration, flow rate, temperature and impurity components in the flue gas in real time, and constructs a hybrid decision-making model to dynamically generate 60 optimized combinations of four types of processes: membrane separation, adsorption, absorption, and liquefaction; the system described in the present invention adopts a modular rapid reorganization design, supports the process chain to complete switching within 30 minutes, and establishes a three-level energy collaborative network of waste heat cascade utilization, cold energy circulation and pressure energy recovery, which reduces the capture energy consumption by 38% to 52%; the present invention breaks through the concentration adaptability limitation of a single technology, effectively covers the CO2 concentration range of 5% to 95%, significantly improves the treatment efficiency and economy of complex industrial flue gas, and is suitable for carbon emission reduction needs in the fields of electricity, steel, chemical industry, etc.

[0231] The embodiments described are preferred implementations of the present invention, but the present invention is not limited to the above implementations. Any obvious improvements, substitutions or modifications that can be made by those skilled in the art without departing from the essence of the present invention are within the scope of protection of the present invention.

Claims

1. An industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis, characterized in that: The industrial flue gas cascade carbon capture system includes a feature sensing unit, an intelligent decision-making unit and a modular execution unit; The characteristic sensing unit is used to monitor the CO2 concentration, gas flow, temperature and impurity components in the industrial flue gas in real time, and feed back the monitored characteristic parameters to the intelligent decision-making unit; The intelligent decision-making unit has a built-in hybrid decision-making model, which dynamically generates a set of carbon capture process combinations based on characteristic parameters and calculates the applicability index of each combination. The carbon capture process includes a combination of at least two processes selected from membrane separation, adsorption, absorption, and liquefaction. The modular execution unit includes a standardized process module and a rapid reorganization interface, and the modular execution unit is assembled according to the results generated by the intelligent decision-making unit; the standardized process module includes a membrane separation unit, an adsorption unit, an absorption unit and a liquefaction unit.

2. The industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis according to claim 1 is characterized in that: The feature sensing unit includes a quantum cascade laser spectrometer, an ultrasonic flowmeter, a micro gas chromatograph, an infrared thermal imager, a CO2 concentration sensor, a temperature sensor and a pressure sensor.

3. The industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis according to claim 1 is characterized in that: The hybrid decision model is constructed based on the characteristics of industrial flue gas, and the hybrid decision model includes: (1) A dynamic decision library of 60 process combinations is formed based on the four-dimensional characteristic matrix of "concentration-flow-impurity-temperature"; (2) Establish a multi-objective optimization function through dynamic optimization control and real-time weight distribution; (3) Evaluate the process combination effectiveness through fuzzy comprehensive evaluation method and calculate the applicability index of each process combination; (4) The operator may select an execution plan based on preset rules.

4. The industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis according to claim 3 is characterized in that: In step (1), the process of forming the dynamic decision library includes: (a) Concentration classification: Low concentration range: CO2<15%, treated by membrane separation unit and / or adsorption unit; Medium concentration range: 15% ≤ CO2 ≤ 40%, treated by a combination of at least two of the following: membrane separation unit, adsorption unit and absorption unit; High concentration range: CO2>40%, using absorption unit and / or liquefaction unit for treatment; (b) Flow correction factor: When flow rate Q≥20,000Nm 3 / h, parallel membrane separation units are mandatory to reduce pressure drop; When Q≤1,000Nm 3 / h, activate the intermittent operation mode of the liquefaction unit; (c) Temperature compensation mechanism: When the industrial flue gas temperature T≥150℃, select membrane separation unit and / or adsorption unit for treatment; When T≤50℃, use the absorption unit and / or liquefaction unit for treatment; Based on the above classification rules, a dynamic decision library containing 60 process combinations is formed.

5. The industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis according to claim 3 is characterized in that: In step (2), the general formula of the multi-objective function is: Where: αi is the process weight coefficient (0≤α i ≤1), Ei is the unit energy consumption of each process (GJ / tCO2), η is the switching cost factor (0.02-0.05GJ / time), n switch is the number of switching modules.

6. The industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis according to claim 5 is characterized in that: The multi-objective function optimization is: min E total =α·E A +β·E B +γ·E C +δ·E D +η·E switch The constraints are: C out ≥95%; ΔP≤15kPa; T regeneration ≤140℃; Among them, A is the membrane separation unit, B is the adsorption unit, C is the absorption unit, and D is the liquefaction unit; α / β / γ / δ are process energy consumption weight coefficients, which are dynamically adjusted according to the flue gas characteristics; E total is the total energy consumption of the system; E A 、E B 、E c 、E D are the unit energy consumption of process A, process B, process C, and process D respectively; η is the equipment switching cost factor; E switch The switching energy consumption caused by the reorganization of the process chain; C out is the CO2 product gas purity constraint; ΔP is the system pressure drop constraint; T regeneration is the solvent regeneration temperature constraint.

7. The industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis according to claim 3 is characterized in that: In step (3), the formula for the combined energy efficiency evaluation is: Normalization processing: In the formula, S i is the applicability index of the i-th process combination; w1, w2, w3 are weight coefficients (default values ​​are 0.4 / 0.4 / 0.2); E i is the specific energy consumption of combination i; η CO 2 is the CO2 capture rate; C OPEX is the annualized operating cost, That is: S i =0.4·(1 / E i )+0.4·η CO2 +0.2·(1 / C OPEX ).

8. The industrial flue gas cascade carbon capture system based on multi-parameter characteristic analysis according to claim 1 is characterized in that: Modular execution units are assembled according to the rules of the energy coordination network: When both adsorption unit and absorption unit are included, the desorption waste heat of the adsorption unit is supplied to the regeneration tower of the absorption module through the heat exchanger; When both a liquefaction unit and an adsorption unit are included, the cold energy of the expander of the liquefaction unit can be used to pre-cool the inlet airflow of the adsorption unit; When a membrane separation unit is included, the residual pressure on the permeate side drives the turbine generator set.

9. A cascade carbon capture method for industrial flue gas, characterized in that: The industrial flue gas cascade carbon capture method is implemented based on the above-mentioned industrial flue gas cascade carbon capture system, and includes: (1) The industrial flue gas is passed into the feature sensing unit, and the four-dimensional feature parameters of the industrial flue gas are collected in real time through the feature sensing unit: CO2 concentration, flow rate, temperature and impurity vector, and a four-dimensional feature matrix of "concentration-flow rate-impurity-temperature" is constructed; (2) Input the four-dimensional characteristic matrix of "concentration-flow-impurity-temperature" into the intelligent decision-making unit, and generate a set of candidate process combinations by the hybrid decision model of the intelligent decision-making unit, and calculate the applicability index of each combination; (3) According to the combination with the highest applicability index, the modular units are driven to complete physical reorganization, perform energy collaborative scheduling, and obtain modular execution units; (4) The industrial flue gas is passed into the modular execution unit for carbon capture.

Images