Low-temperature solid-state ysZ electrochemical pump and ai4ai federation edge agent carbon capture catalytic conversion integrated system

Abstract

The application discloses a low-temperature modified YSZ electrochemical pump and an AI4AI federal edge agent integrated system, and belongs to the cross field of CCUS solid-state electrochemistry and federal intelligent control. The application adopts lattice doping modification YSZ composite carbonate electrolyte to realize 60-80 DEG C low-temperature solvent-free capture, and one-step output of 99.99% high-purity CO2; the federal encryption AI4AI base and the three-level edge agent are matched, local data is encrypted and shared, dynamic matching capture and catalytic load, electrolyte aging prediction, and formic acid and methanol products can be switched. The system has low comprehensive energy consumption of 0.72-0.78 GJ / tCO2, covers 3%-15% full-concentration industrial flue gas, and solves the industry defects of traditional pure YSZ high temperature, high energy consumption and inability to resource linkage.

Description

Technical Field

[0001] This invention belongs to the fields of solvent-free solid-state electrochemical carbon capture, CO2 electrocatalytic resource utilization, and federated edge multi-agent collaborative technology, specifically CCUS (Carbon Capture, Utilization and Storage). It targets the one-step solvent-free preparation of electronic-grade CO2 from low-concentration flue gas in steel, cement, and aluminum plants, and the conversion of formic acid / methanol. The hardware is a solid-state electrochemical stacked pump, eliminating wet solvents and spiral wound membrane modules. It innovatively integrates a low-temperature modified YSZ solid electrolyte electrochemical pump, a full-link catalytic conversion unit, and an AI4AI (Artificial Intelligence for Agent Interaction) federated encrypted peer-to-peer collaborative unsupervised base and a three-level edge agent (artificial intelligence agent) cluster architecture. This technology has no overlap with the wet ionic liquid and electrochemical phase change membrane technologies. Background Technology

[0002] CCUS (Carbon Capture, Utilization and Storage) is a core supporting technology for realizing the high-value utilization of carbon resources in the industrial field. Solid-state electrochemical carbon capture has advantages such as no need for organic absorption solvents, simplified process flow and high product purity, and is the direction of the next generation of decarbonization technology.

[0003] Existing publicly available solid-state electrochemical capture systems mainly use unmodified commercial pure YSZ electrolyte, with operating temperatures generally exceeding 500℃. The investment in supporting high-temperature heating equipment is high, heat loss is huge, and overall energy consumption remains high. Conventional electrochemical devices are only suitable for high-concentration CO2 gas sources and cannot directly process 3%~15% industrial low-concentration flue gas. Furthermore, the capture and catalytic conversion processes are arranged separately, resulting in a large amount of venting loss due to the imbalance between the supply and demand of high-purity CO2.

[0004] Meanwhile, traditional intelligent control solutions have several inherent technical defects: ① Conventional machine learning models rely on massive manually labeled working condition datasets, resulting in long debugging cycles for new projects and making it impossible to share optimization experience across production lines; ② Without a local data privacy protection architecture, all raw production data from the entire plant must be uploaded to the cloud, posing a risk of industrial data leakage; ③ The lack of a long-term simulation mechanism for solid electrolyte aging, relying solely on periodic shutdowns for testing, easily leads to sudden equipment failures and shutdowns; ④ The inability to dynamically adjust the capture and catalytic load in conjunction with chemical market prices and grid time-of-use electricity prices results in low overall returns for carbon resource recovery projects.

[0005] It is evident that existing technologies cannot achieve the integrated closed-loop operation of low-temperature modified electrolyte solvent-free capture, high-purity CO2 in-situ catalytic conversion, and federal privacy-protected full-domain intelligent control, thus hindering the industrial-scale promotion and high-value resource utilization of solid-state electrochemical CCUS. Summary of the Invention

[0006] 3.1 Purpose of the Invention Our original low-temperature modified YSZ composite carbonate solid electrolyte electrochemical pump, unlike commercial high-temperature pure YSZ, can operate stably at 60~80℃ and directly extract ≥99.99% electronic-grade CO2 without solvents; Build a three-level edge agent cluster to coordinate the entire operating status of electrochemical pump arrays, CO2 buffers, and electrocatalytic reactors; It is equipped with a self-developed AI4AI (intelligent agent mutual collaboration framework) federated encrypted peer-to-peer collaborative unsupervised base, which retains the original data locally at the edge, only encrypts the model to communicate, and takes into account both experience sharing and industrial data privacy. Dynamic matching of front-end CO2 production and back-end catalytic conversion load reduces CO2 venting loss to below 1%; digital twin predicts the aging of modified YSZ electrolyte and provides early warning for component replacement. By independently switching the product synthesis route based on the market prices of formic acid and methanol, the project's average annual comprehensive income can be improved. The system's overall capture energy consumption is reduced to 0.72~0.78 GJ / tCO2 (gigajoules per ton of carbon dioxide, a unit of energy consumption measurement), which is suitable for cost reduction through off-peak electricity storage and long-term iterative needs for carbon resource utilization.

[0007] 3.2 Complete Integrated Technical Solution Module 1: Low-Temperature Solid-State Modified YSZ Composite Carbonate Electrochemical Pump YSZ (Yttria-Stabilized Zirconia; this invention uses a lattice-doped modified YSZ matrix, which differs from the commercially available unmodified high-temperature pure YSZ. By controlling the yttria doping ratio and grain size, it is co-fired with multi-element carbonates, overcoming the limitation that pure YSZ only has ion-conducting activity above 500℃) composite carbonate sintered body as a solid electrolyte; operating temperature 60~80℃; stacked modular design, single module annual production of 1000 tons of CO2; suitable for flue gas CO2 concentration of 3%~15%, current efficiency ≥85%, CO2 purity ≥99.99%, electrolyte service life ≥9.6 years.

[0008] Module 2: End-to-End Hardware Unit Flue gas pretreatment → stacked electrochemical pump array → high-purity CO2 buffer pressure stabilizing tank → Pd-Cu (palladium-copper bimetallic) electrocatalytic reactor → waste heat recovery unit; the entire production line is equipped with multiple types of sensors to collect real-time data on temperature, electric field strength, buffer pressure, and liquid phase product concentration.

[0009] Module 3: Level 3 Edge Agent + AI4AI Federated Encryption Collaboration Base Agent hierarchy: Level 1 global model coordination agent, Level 2 capture / caching / catalysis segmentation agent, and Level 3 single electrochemical module execution agent; AI4AI Federation Base Core: An unsupervised framework for encrypted peer-to-peer interaction among four edge nodes; local raw production data is not transmitted externally, only encrypted model weights are shared across production lines; a modified YSZ electrolyte aging digital twin model is constructed, which can provide early warning of component replacement 3 months in advance; automatic energy storage and carbon capture during off-peak hours at night, and concentrated catalytic conversion during the day; Closed-loop process: Flue gas is introduced into each electrochemical module, and the bottom agent differentiates and controls the electric field and temperature → the buffer pressure agent calculates the gas volume reserve in real time → the global big model combines time-of-use electricity price and chemical market price to issue product production load instructions → the catalytic unit matches the CO2 supply to synthesize formic acid or methanol in a targeted manner → the whole process data is fed back to the federated base iterative aging prediction model.

[0010] 3.3 Core Innovation Protection Points A modified YSZ composite carbonate solid electrolyte electrochemical pump integrated sintering preparation process adapted to low temperature conditions of 60~80℃; Stacked electrochemical modules with independent and differentiated electric field temperature control architecture for each module; A dynamic coupling matching algorithm between front-end capture and back-end electrocatalytic conversion load; A digital twin model for extrapolating the conductivity of modified YSZ solid electrolyte over aging; The logic of price-linked products in the chemical market and their self-regulating switching; An electrochemical Joule waste heat recovery and scheduling mechanism reduces external heating to near zero. AI4AI Federation-Encrypted Four-Edge Node Peer-to-Peer Unsupervised Collaboration Base for Local Original Data Privacy Protection; An unsupervised clustering adaptive adaptation framework for multi-condition flue gas components; Low-cost off-peak electricity at night, energy storage for carbon capture, and daytime catalytic operation mode; Single-module fault online isolation and uninterrupted production operation strategy; Temperature-electric field co-optimization algorithm, stable current efficiency ≥85%; Dispatch module for the combustion and utilization of O2 (oxygen) byproducts from electrolysis in the plant area; Mechanism for transferring and adapting historical control experience in production line expansion; 3%~8% extremely low concentration flue gas adaptive boosting control logic; An adaptive compensation algorithm for electrolyte constant temperature in all four seasons. Automatic closed-loop module for generating equipment maintenance and repair work orders; Cloud-based industry regulation strategy remote iteration architecture.

[0011] 3.4 Beneficial Technical Effects Energy consumption: The overall energy consumption of the system is 0.72~0.78 GJ / tCO2, which is far lower than that of commercial pure YSZ high temperature system; Lifespan: The lifespan of the modified YSZ composite electrolyte is extended from 8 years in conventional high-temperature systems to 9.6 years, reducing equipment depreciation costs by 20%. Benefits: By switching target products in line with chemical market prices, the average annual comprehensive economic benefits increase by 27%; Losses: High-purity CO2 venting losses were reduced to 0.8%, and the overall utilization rate of carbon raw materials was significantly improved; Stability: Unplanned equipment downtime is reduced by 90%, and the system's continuous operation reliability is 99.85%; Adaptability: Full coverage of industrial flue gas from 3% to 15%, adapting to the mainstream industrial development direction of carbon resource utilization. Attached Figure Description

[0012] Figure 1. Logical overview of the Triple Intelligence Architecture (Large Model + Agent + Federated AI4AI Foundation); Figure 2. Cross-sectional view of the stacked module structure of the low-temperature modified YSZ electrochemical pump; Figure 3. Complete process flow of capture-buffering-catalytic conversion; Figure 4. Aging digital twin projection curve of modified YSZ composite electrolyte; Figure 5. Time series comparison of dynamic matching of supply and demand load. Detailed Implementation

[0013] Test platform: a standardized 500-ton / year CO2 integrated pilot system, configured with 12 electrochemical modules, with flue gas CO2 concentration dynamically fluctuating from 5% to 14%; the test cycle is 180 days, and except for the electrolyte type and federated AI architecture, the other equipment, flue gas and process conditions are completely unified.

[0014] Comparative Example 1: Commercial unmodified pure YSZ with artificially fixed parameters The commonly used unmodified pure YSZ electrolyte is used, the operating temperature is 520℃, and the parameters of the collection and catalysis processes are manually fixed. Actual measurements: Comprehensive energy consumption 3.1 GJ / tCO2, CO2 venting loss 8.7%; after 156 days of operation, the electrolyte failed and caused a sudden shutdown of 16 hours, and it can only produce formic acid.

[0015] Comparative Example 2: Modified YSZ + Single BP (Back Propagation) Segmented Neural Network The electrolyte is the modified YSZ of this invention, but it only uses a segmented independent neural network and has no federated encryption cooperative architecture; Actual test results showed that optimization experience could not be shared across production lines, supply and demand imbalances led to high losses from unloaded loads, and products could not be switched independently.

[0016] Example 1: The complete integrated system of Federal AI4AI modified YSZ of the present invention The model converged and stabilized in the first 25 days, with average operating losses of 0.8% and average current efficiency of 86.2%. The digital twin predicted electrolyte aging in advance to prevent sudden shutdowns. The production paths for formic acid and methanol were switched according to monthly chemical prices, resulting in an overall increase in revenue of 26%.

[0017] Example 2: Verification of Off-Peak Electricity Storage Mode By activating off-peak electricity storage strategies at night, overall electricity costs have been reduced by 29%.

[0018] Example 3: Module Expansion and Adaptation Verification Four new electrochemical modules were added in the later stages of operation, and the system completed adaptive operation in 5 days with no reduction in capacity or separation efficiency.

Claims

1. A low-temperature solid-state YSZ electrochemical pump and AI4AI federated edge agent integrated carbon capture and catalytic conversion system, characterized in that, include: Modified YSZ composite carbonate low-temperature solid-state electrochemical stacked pump array, CO2 buffer voltage regulator unit, Pd-Cu electrocatalytic conversion reactor, waste heat recovery unit, three-level hierarchical edge agent cluster, AI4AI federated encrypted peer-to-peer unsupervised collaborative base; The modified YSZ is a lattice-doped yttrium-stabilized zirconium oxide matrix, which is sintered with multi-component carbonates to form a low-temperature solid electrolyte. The electrochemical pump operates at 60~80℃, directly extracting ≥99.99% high-purity CO2 from 3%~15% industrial flue gas without solvents. The three-level agent includes a top-level global overall planning agent, a second-level trapping / caching / catalytic segment agent, and a bottom-level single electrochemical module execution agent. The AI4AI federated base adopts four edge node encrypted interaction, with original data retained locally. Only the encrypted model weights are shared across production lines, constructing a digital twin of the modified YSZ composite electrolyte aging, dynamically matching CO2 production and catalytic conversion load, and autonomously switching product synthesis paths based on the market prices of formic acid and methanol, realizing intelligent closed-loop operation of the entire chain of solvent-free trapping, high-purity purification, and high-value chemical conversion.

2. The system according to claim 1, characterized in that, The modified YSZ composite carbonate solid electrolyte is prepared by adjusting the yttrium oxide doping amount and grain size, and then co-firing it with multi-element carbonates. A single module can process 1,000 tons of CO2 per year, with a current efficiency of ≥85% and a service life of ≥9.6 years.

3. The system according to claim 1, characterized in that, The AI4AI Federal Base can provide early warnings of aging of modified YSZ composite solid electrolytes up to 3 months in advance and automatically generate equipment maintenance work orders.

4. The system according to claim 1, characterized in that, Configure a nighttime off-peak electricity storage and carbon capture operation mode to reduce overall electricity costs by utilizing low-priced electricity.

5. The system according to claim 1, characterized in that, Faults in a single component of the electrochemical module can be isolated online, eliminating the need for a complete system shutdown for maintenance.

6. A method for integrated carbon capture and catalytic control based on the system described in any one of claims 1 to 5, comprising the following steps: S1 Industrial flue gas is fed into a low-temperature electrochemical pump array, and the bottom agent differentiates and controls the electric field and temperature of each module to achieve selective CO2 extraction; S2 High-purity CO2 is fed into the buffer tank, and the buffer agent calculates the amount of gas stored in real time; The S3 global model combines the time-of-use electricity price of the power grid and the market price of chemical products to issue product production load instructions; The S4 catalytic unit is used to synthesize formic acid or methanol in a targeted manner by matching the CO2 supply. The S5 full-process time-series data is fed back to the AI4AI federated base, and the YSZ composite electrolyte aging model and global control strategy are modified by unsupervised clustering and iterative modification, and the system operates continuously in a closed loop.

Images