Field catalyst evaluation system and method based on surface detection

Abstract

The invention discloses an on-site catalyst evaluation system and method based on surface detection, and belongs to the technical field of flue gas treatment. According to the technical scheme, the field catalyst evaluation system based on surface detection comprises a gas inlet atmosphere modulation module, a reactor module, a gas detection module and a tail gas treatment module which are connected in sequence; the gas inlet atmosphere modulation module comprises a gas inlet flue, a gas inlet stop valve, a gas inlet adjusting valve, a flue gas mixer, a modulation gas pipeline, a modulation gas stop valve and a modulation gas mass flow meter; the high-density sensor array and the optical imaging unit are combined with a spatial interpolation algorithm to generate an efficiency distribution heat map, real-time monitoring of surface activity distribution of the catalyst is achieved, through parallel testing of multiple reactor components and machine learning life prediction, the evaluation precision and research and development efficiency of the catalyst are remarkably improved, and the research and development cost is lowered. The method has the advantages that the catalyst surface efficiency heterogeneity is captured in real time, multi-channel accelerated testing is supported, and the service life is accurately predicted.

Description

Technical Field

[0001] The present invention belongs to the technical field of flue gas treatment, and in particular relates to an on-site catalyst evaluation system and method based on surface detection. Background Art

[0002] With the advancement of ultra-low emission policies in my country's steel industry, CO emission control has become a key challenge in flue gas treatment. Flue gas generated by equipment such as sintering machines and coke ovens has a complex composition, containing high concentrations of dust, SO2, water vapor, and heavy metal particles, posing a serious threat to the activity and lifespan of catalysts. Currently, while precious metal catalysts (such as Pt / Al2O3) and transition metal catalysts (such as Fe-Mn oxide) exhibit good CO oxidation efficiency under laboratory conditions, in actual industrial scenarios, complex flue gas components lead to rapid catalyst deactivation, and their field service life is often significantly shortened compared to laboratory data.

[0003] Existing catalyst evaluation technologies face the following bottlenecks: single-point sampling leads to biased evaluations. Traditional testing equipment utilizes only a single-point gas analyzer at the reactor component outlet, making it impossible to obtain information on the spatial distribution of catalyst surface reaction efficiency. When catalyst activity is uneven due to localized dust accumulation, sulfur poisoning, or thermal sintering, single-point data will significantly underestimate the overall degree of deactivation and fail to pinpoint the deactivated region, hindering targeted catalyst optimization. Testing cycles are excessively long. Catalyst life evaluation requires months or even a year of continuous testing simulating actual operating conditions. Each change in catalyst formulation or type requires re-testing, resulting in lengthy R&D cycles and a struggle to meet the rapid selection needs of steel companies. The ability to test multiple catalysts in parallel is lacking. Existing systems are mostly based on a single reactor component, requiring different catalysts to be tested sequentially. This is inefficient and cannot guarantee consistent operating conditions across batches, hindering data comparability. Accelerated aging and life prediction technologies are inadequate. Existing methods accelerate catalyst aging by simply increasing pollutant concentrations, but lack the ability to simulate the synergistic effects of multiple flue gas components (such as SO₂, H₂O, and dust). This results in significant deviations between laboratory aging data and actual field attenuation patterns.

[0004] Literature and patent research indicate that while existing technologies have proposed multi-reactor component testing devices, they still rely on traditional single-point detection and fail to address the issue of spatial efficiency distribution analysis. Other approaches have attempted to incorporate optical imaging technology, but these techniques are only used to observe catalyst morphology and are not integrated with gas concentration data for analysis. Therefore, developing an on-site evaluation system that can capture catalyst surface efficiency heterogeneity in real time, support multi-channel accelerated testing, and accurately predict lifetime is key to promoting the industrial application of CO removal catalyst technology. To address these issues, existing technologies urgently need improvement. Summary of the Invention

[0005] The present invention provides an on-site catalyst evaluation system and method based on surface detection to solve at least one of the above technical problems.

[0006] The technical solution adopted in the present invention is: An on-site catalyst evaluation system based on surface detection includes an intake atmosphere modulation module, a reactor module, a gas detection module and an exhaust gas treatment module connected in sequence; The intake atmosphere modulation module includes an intake flue, an intake shut-off valve, an intake regulating valve, a flue gas mixer, a modulation gas pipeline, a modulation gas shut-off valve, a modulation gas mass flow meter, a flue gas heater, and a temperature sensor. The intake flue and the modulation gas pipeline are respectively connected to the flue gas mixer. After mixing, the flue gas is heated by the flue gas heater and the temperature is closed-loop controlled by the temperature sensor. The reactor module comprises at least two reactor components arranged in parallel, each inlet pipe is provided with a reaction gas mass flow meter and an inlet temperature meter, the outlet pipe is provided with an outlet temperature meter, and a catalyst is replaceably installed in the reactor component; The gas detection module includes a post-mixer measurement point located at the exhaust end of the flue gas mixer and a post-reactor component measurement point located at the outlet of the reactor component, and also includes a surface detection unit for detecting gas data discharged from the post-mixer measurement point and the post-reactor component measurement point. The surface detection unit includes a high-density sensor array and an optical imaging unit, and a local efficiency calculation unit connected to the two. The calculation unit generates a catalyst surface efficiency distribution heat map using a spatial interpolation algorithm. The exhaust gas treatment module includes an induced draft fan and a return air flue, which sends the tested flue gas back to the on-site flue. The return air flue is provided with a return air shut-off valve.

[0007] Preferably, the reactor module comprises 2-4 independently controlled reactor components, and the inlet valve and outlet valve of each reactor component can be opened and closed independently to adjust the number of reactor components involved in the test.

[0008] Preferably, the outer walls of the reactor components, the inlet pipe and the outlet pipe are covered with an insulation layer, the insulation layer material is ceramic fiber or rock wool, and the thickness is 30-50 mm.

[0009] Preferably, a modular catalyst tray is provided in the reactor component to support the rapid replacement of honeycomb, granular or coated catalysts, and the tray interface size is compatible with non-standard customization.

[0010] Preferably, the high-density sensor array is composed of micro-electrochemical sensors, which are arranged in a grid pattern along the cross-section of the reactor component, with a spacing of 5-10 mm between adjacent sensors; the optical imaging unit includes an infrared thermal imager and a visible light CCD camera, which obtains the catalyst surface temperature field and product deposition images through a sapphire observation window.

[0011] Preferably, the local efficiency calculation unit executes a multimodal data fusion algorithm to weightedly fuse the discrete concentration data of the sensor array with the continuous surface data of the optical imaging, and dynamically correct the overall conversion efficiency value of the catalyst.

[0012] Preferably, it also includes a machine learning life prediction module, which receives time series data of the efficiency distribution heat map, extracts local deactivation features through a convolutional neural network, and outputs a catalyst remaining life prediction value and a deactivation risk area marking map.

[0013] Preferably, the modulated gas pipeline is connected to the SO2, H2O vapor and dust generating device, and the accelerated aging gas with a concentration of 1.5-3 times that of the on-site flue gas is injected into the flue gas mixer through the modulated gas mass flow meter.

[0014] Preferably, a pressure difference sensor is provided between the induced draft fan and the return air duct. When the pressure difference between the on-site air intake point and the return air point is greater than the system resistance, the induced draft fan is automatically shut down and the natural pressure difference driving mode is enabled.

[0015] An on-site catalyst evaluation method based on surface detection, the specific steps are as follows: S1. Injecting modulated gas into the on-site flue gas to accelerate catalyst aging; S2, heating the mixed flue gas to 200-400°C and then distributing it to multiple reactor components; S3, synchronously collecting multi-dimensional data of the downstream cross section of the catalyst through a high-density sensor array and an optical imaging unit; S4. Use spatial interpolation algorithms to generate efficiency distribution heat maps and combine them with machine learning models to predict lifespan attenuation trends. S5. Screen the optimal catalyst formula and process parameters based on the parallel testing results of multiple reactor components.

[0016] Due to the adoption of the above technical solution, the beneficial effects achieved by the present invention are as follows: 1. After being thoroughly mixed with the modulating gas in a mixer, the on-site flue gas is heated to the set temperature and then enters the array of return air shut-off valves in the parallel reactor components. Each reactor component's return air shut-off valve independently operates a different catalyst sample. A sensor array at its outlet section monitors the spatial distribution of gas concentration in real time, while an optical imaging unit records the catalyst surface state. The calculation unit's return air shut-off valve fuses these two data sources and uses an interpolation algorithm to generate a heat map reflecting the conversion efficiency of each region of the catalyst surface. After testing, the flue gas returns to the original flue through the exhaust treatment module's return air shut-off valve to avoid secondary contamination.

[0017] This solution utilizes a parallel architecture with return gas shut-off valves for multiple reactor components to enable simultaneous comparative testing, ensuring consistent operating conditions. Furthermore, while existing accelerated aging methods only increase the concentration of a single pollutant, this solution utilizes a multi-component modulated gas injection device to more realistically simulate actual flue gas conditions.

[0018] Through the above technical solution, this application can obtain the conversion efficiency distribution of each area on the catalyst surface in real time, accurately locate the deactivated area, and provide spatially resolved data support for catalyst formulation optimization. The parallel testing structure of the return gas shut-off valve of multiple reactor components significantly shortens the evaluation cycle of multi-formulation catalysts and ensures the consistency of operating parameters through synchronous testing. The accelerated aging method of composite modulated gas is closer to the actual industrial environment, improving the correlation between laboratory test data and field service performance.

[0019] 2. During the test, any number of the 2-4 reactor component return air shut-off valves can be opened based on experimental requirements. For example, to compare the performance of three catalysts, three reactor component return air shut-off valves can be opened simultaneously while the remaining reactor component return air shut-off valves remain closed. Each reactor component return air shut-off valve maintains the same temperature, pressure, and gas flow conditions. The independent control of its inlet and outlet return air shut-off valves ensures a consistent test environment for each reactor component return air shut-off valve. After a reactor component return air shut-off valve completes its test cycle, its inlet and outlet return air shut-off valves can be individually closed for catalyst replacement, while the remaining reactor component return air shut-off valves remain in continuous operation.

[0020] This solution uses parallel testing of return gas shut-off valves for multiple reactor components to enable simultaneous life assessment of different catalysts under identical flue gas conditions, while avoiding energy losses caused by frequent shutdowns.

[0021] Through the above-mentioned technical solution, this application achieves simultaneous comparative testing of multiple catalyst formulations, eliminating differences in operating conditions between batch experiments and ensuring data comparability. During testing, the number of return air shut-off valves in the sewage reactor components can be increased or decreased at any time to meet the needs of experiments of varying scales, significantly shortening the catalyst selection cycle. If a return air shut-off valve in a reactor component experiences an abnormality, it can be isolated and maintained without affecting overall system operation, improving the stability and continuity of the testing process.

[0022] 3. When the mixed gas, after temperature control by the flue gas heater's return air shut-off valve, enters the reactor's return air shut-off valve, the insulation layer blocks heat dissipation through radiation and convection, keeping temperature fluctuations within the target reaction temperature range. When the flue gas temperature is within the catalytic reaction activity window of 200-400°C, the insulation layer effectively reduces the axial and radial temperature gradients within the reactor's return air shut-off valve, ensuring temperature uniformity across the catalyst bed. Especially when there are changes in wind speed or temperature differences between day and night in the field environment, the insulation layer can mitigate the external environment's impact on the thermal equilibrium state of the reactor's return air shut-off valve, ensuring that catalyst performance test results more accurately reflect its operating conditions in the actual flue.

[0023] This solution forms a stable temperature maintenance capability by precisely controlling the insulation material and thickness, thus eliminating the catalytic efficiency measurement error caused by temperature fluctuations.

[0024] Through the above-mentioned technical solution, this application solves the problem of difficult to stably control reaction temperatures in complex industrial environments, effectively avoiding deviations in catalyst activity assessments due to heat loss. The provision of an insulation layer enables the testing system to accurately simulate the thermal environment of the catalyst within a real flue, ensuring consistency between laboratory accelerated aging data and actual field attenuation patterns, providing a reliable thermodynamic parameter benchmark for catalyst formulation optimization.

[0025] 4. This application addresses the long experimental cycles and poor operating condition consistency caused by the need to test multiple catalyst formulations sequentially. Testers can quickly swap catalyst trays with different structures within the return air shut-off valve of the same reactor component, ensuring performance comparisons of different catalysts under the same operating conditions and improving data comparability. The tray interface's dimensional compatibility accommodates the non-standard shapes of customized catalysts from steel companies, avoiding test interruptions due to dimensional discrepancies.

[0026] 5. This solution achieves three-dimensional reconstruction of the catalyst surface efficiency distribution through the coordinated detection of a high-density gridded sensor array and multispectral imaging. Existing optical detection technologies only observe the macroscopic morphology of the catalyst and do not correlate analysis with gas concentration data. This solution, however, simultaneously acquires temperature fields, product deposition images, and local gas concentration data to establish a multidimensional failure pattern recognition model.

[0027] Through the above technical solution, this application solves the problem of catalyst efficiency evaluation deviation caused by single-point sampling. It realizes spatial interpolation calculation of local concentration field through gridded high-density sensor array, combines optical imaging data to identify temperature anomalies and product deposition areas, accurately locates the catalyst deactivation position and quantifies the degree of activity attenuation, and provides accurate data support for targeted optimization of catalyst formulation and regeneration process.

[0028] 6. The present invention organically combines the quantitative detection capability of electrochemical sensors with the spatial resolution capability of optical imaging through a multimodal data fusion algorithm, solving the evaluation bias problem caused by single-point sampling while avoiding the risk of misjudgment caused by dust obstruction or temperature drift in a single detection method.

[0029] Through the above-mentioned technical solution, this application can accurately identify the non-uniform distribution of catalyst surface activity, locate the spatial location of deactivated areas such as dust accumulation and sulfur poisoning, and provide data support for targeted cleaning or regeneration. Furthermore, a dynamic correction mechanism can eliminate the interference of local abnormal data on overall efficiency assessment, improve the reliability of the catalyst life prediction model, and shorten the field testing cycle.

[0030] 7. This approach fuses discrete concentration data from a high-density sensor array with continuous surface data from optical imaging to construct a spatiotemporal feature matrix recognizable by a convolutional neural network, thereby accurately quantifying the impact of local deactivation on overall lifespan. Based on the spatiotemporal evolution of catalyst surface efficiency, this approach can proactively identify localized deactivation areas during accelerated aging testing, shortening the experimental cycle by over 50% compared to traditional continuous testing methods. The resulting deactivation risk area marker map can be used directly to guide catalyst support structural optimization, such as adding an anti-sulfur coating to the marked areas or adjusting the honeycomb pore size distribution, thereby improving the catalyst's field suitability.

[0031] 8. By using surface detection technology to simultaneously obtain information on the full surface activity of the catalyst, combined with the parallel operation mode of the return air shut-off valves of multiple reactor components, it is possible to quickly compare the performance differences of multiple catalysts under the same flue gas conditions, and at the same time achieve dynamic prediction of life decay trends based on machine learning models. Through the above technical solutions, this application solves the problem of bias in the evaluation of local catalyst deactivation caused by single-point sampling, shortens the screening cycle of multiple catalyst formulations, improves the matching degree between accelerated aging experiments and actual on-site working conditions, and provides reliable technical support for the rapid selection and life management of catalysts in industrial flue gas treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Schematic diagram of the working principle of a specific embodiment of the present invention.

[0033] The drawings described herein are used to provide further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0034] In the attached figure: 1. Intake atmosphere modulation module; 2. Reactor module; 3. Gas detection module; 4. Exhaust gas treatment module; 101. Intake flue; 102. Intake shut-off valve; 103. Intake regulating valve; 104. Modulation gas pipeline; 105. Modulation gas shut-off valve; 106. Modulation gas mass flowmeter; 107. Flue gas mixer; 108. Flue gas heater; 109. Temperature sensor; 201. Inlet pipe; 202. Inlet valve; 203. Reaction gas mass flowmeter; 204. Reactor component; 205. Inlet temperature gauge; 206. Outlet temperature gauge; 207. Outlet pipe; 208. Outlet valve; 301. Measuring point after mixer; 302. Surface detection unit; 303. Calculation unit; 401. Induced draft fan; 402. Return air shut-off valve; 403. Return air flue. DETAILED DESCRIPTION

[0035] In order to more clearly illustrate the overall concept of the present invention, a detailed description is given below in an exemplary manner in conjunction with the accompanying drawings.

[0036] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0037] In the present invention, unless otherwise expressly specified or limited, terms such as "mounted," "connected," "connect," and "fixed" should be understood broadly. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection, electrical connection, or communication; direct connection or indirect connection through an intermediate medium; and internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0038] In the present invention, unless otherwise clearly specified and limited, a first feature "above" or "below" a second feature may be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. In the description of this specification, the description with reference to the terms "implementation method", "embodiment", "one embodiment", "example" or "specific example" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in an appropriate manner in any one or more embodiments or examples.

[0039] Those skilled in the art will understand that the performance evaluation of catalysts in flue gas treatment in the steel industry faces many challenges. The catalysts exhibit good activity in laboratory environments, but in actual industrial scenarios, they are rapidly deactivated due to complex flue gas components, and their service life is significantly shortened. Traditional evaluation devices use single-point gas analyzers, which cannot capture the spatial differences in the reaction efficiency of the catalyst surface. When the catalyst has localized dust accumulation or sulfur poisoning, the single-point sampling data will seriously underestimate the overall degree of deactivation and cannot locate the failure area. In addition, most existing systems have a single reactor component structure, and different catalysts need to be replaced and tested in sequence, resulting in a long R&D cycle and difficulty in meeting the needs of rapid selection.

[0040] To address these issues, a hybrid detection method combining a high-density sensor array with optical imaging is being considered to address the lack of visibility of catalyst surface efficiency distribution. Data fusion technology is used to reconstruct the catalyst surface activity distribution. To address the inefficiency of multi-catalyst testing, a parallel reactor component structure is proposed to enable parallel testing of different catalyst formulations. For accelerated aging, a modulated gas injection device with adjustable concentration is designed to simulate the synergistic effects of multiple components in actual flue gas.

[0041] Therefore, refer to Figure 1 This application proposes an on-site catalyst evaluation system based on surface detection, comprising an intake atmosphere modulation module 1, a reactor module 2, a gas detection module 3, and an exhaust gas treatment module 4 connected in sequence; wherein, the intake atmosphere modulation module 1 integrates the on-site flue gas and the modulation gas through the flue gas mixer 107, and delivers them to the reactor module 2 after heating and temperature control. The reactor module 2 is provided with multiple parallel reactor components 204, and each reactor component 204 is independently installed with a catalyst. The gas detection module 3 deploys a sensor array and an optical imaging unit at the outlet cross section of the reactor component 204, and generates a catalyst surface efficiency distribution map through the local efficiency calculation unit 303. The exhaust gas treatment module 4 returns the tested flue gas to the original flue.

[0042] Specifically, the air intake atmosphere modulation module 1 includes an air intake flue 101, an air intake shut-off valve 102, an air intake regulating valve 103, a flue gas mixer 107, a modulation gas pipeline 104, a modulation gas shut-off valve 105, a modulation gas mass flow meter 106, a flue gas heater 108 and a temperature sensor 109. The air intake flue 101 and the modulation gas pipeline 104 are respectively connected to the flue gas mixer 107. The mixed flue gas is heated by the flue gas heater 108 and the temperature is closed-loop controlled by the temperature sensor 109. The reactor module 2 includes at least two reactor components 204 arranged in parallel. Each inlet pipe 201 is provided with a reaction gas mass flow meter 203 and an inlet temperature meter 205, and the outlet pipe 207 is provided with an outlet temperature meter 206. The reactor component 2 04 is replaceably installed with a catalyst; the gas detection module 3 includes a mixer post-measuring point 301 arranged at the exhaust end of the flue gas mixer 107 and a reactor component 204 post-measuring point arranged at the outlet of the reactor component 204, and also includes a surface detection unit 302 for detecting the gas data discharged from the mixer post-measuring point 301 and the reactor component 204 post-measuring point. The surface detection unit 302 includes a high-density sensor array and an optical imaging unit, and a local efficiency calculation unit 303 connected to the two. The calculation unit 303 generates a catalyst surface efficiency distribution heat map through a spatial interpolation algorithm; the exhaust gas treatment module 4 includes an induced draft fan 401 and a return air flue 403, which sends the flue gas after the test back to the on-site flue, and the return air flue 403 is provided with a return air shut-off valve 402.

[0043] The flue gas mixer 107 in the intake atmosphere modulation module 1 is a device used to mix the on-site flue gas with the modulation gas. Specifically, it can be implemented using a multi-channel Venturi mixing structure to ensure thorough and uniform mixing of the gases. The flue gas heater 108 is a component used to raise the temperature of the mixed gas. Specifically, it can be implemented using a resistive heating tube, which works in conjunction with the temperature sensor 109 to form a closed-loop temperature control loop. The parallel reactor component 204 in the reactor module 2 represents multiple independently operating catalytic reaction units. Specifically, it can be implemented using a tubular reactor component 204 with a quick-release flange interface, supporting simultaneous testing of different catalysts. The high-density sensor array is a densely arranged gas concentration detection device. Specifically, it can be implemented using micro-electrochemical sensors that capture cross-sectional concentration distributions in a grid-like layout. The optical imaging unit is a device that obtains visual information about the catalyst surface. Specifically, it can be implemented using a sapphire observation window combined with a dual-spectral camera to simultaneously capture temperature fields and deposit images. The local efficiency calculation unit 303 is a computational module that processes multidimensional data. Specifically, it can be implemented using a spatial interpolation algorithm and a weighted fusion model to combine discrete concentration data with continuous surface data to generate an efficiency heat map.

[0044] Specifically, the on-site flue gas and the modulated gas are thoroughly mixed in a mixer, heated to a set temperature, and then fed into an array of parallel reactor components 204. Each reactor component 204 independently operates a different catalyst sample. A sensor array at its outlet section monitors the spatial distribution of gas concentration in real time, while an optical imaging unit records the catalyst surface conditions. The calculation unit 303 fuses these two data sources and uses an interpolation algorithm to generate a heat map reflecting the conversion efficiency of each region of the catalyst surface. After testing, the flue gas is returned to the original flue through the exhaust treatment module 4 to avoid secondary pollution.

[0045] Compared to existing technologies, traditional systems rely on single-point gas analyzers and are unable to identify uneven distribution of catalyst surface activity. This solution, however, utilizes cross-sectional multi-dimensional detection technology to accurately visualize efficiency heterogeneity. Existing devices often utilize a single reactor component 204, requiring serial testing of different catalysts. This solution utilizes a parallel architecture with multiple reactor components 204 to enable simultaneous comparative testing and ensure consistent operating conditions. Furthermore, while existing accelerated aging methods only increase the concentration of a single pollutant, this solution utilizes a multi-component modulated gas injection device to more realistically simulate actual flue gas conditions.

[0046] Through the above technical solution, this application can obtain the conversion efficiency distribution of each area on the catalyst surface in real time, accurately locate deactivated areas, and provide spatially resolved data support for catalyst formulation optimization. The parallel testing structure of multiple reactor components 204 significantly shortens the evaluation cycle of multi-formulation catalysts and ensures the consistency of operating parameters through synchronous testing. The accelerated aging method using composite modulated gas is more realistic for industrial environments, improving the correlation between laboratory test data and field service performance.

[0047] The present application further proposes that the reactor module 2 includes 2-4 independently controlled reactor components 204, and the inlet valve 202 and outlet valve 208 of each reactor component 204 can be opened and closed independently to adjust the number of reactor components 204 participating in the test.

[0048] Independently controlled reactor components 204 are sealed reaction vessels with independent fluid pathways and individually adjustable operating parameters. Specifically, this can be achieved using a pneumatic valve assembly controlled by a solenoid valve, ensuring that different catalyst testing conditions do not interfere with each other. Independently operable inlet and outlet valves 202 and 208 are independently operable, meaning that each reactor component 204 is equipped with independently operable shutoff devices at its inlet and outlet ends. Specifically, flanged ball valves or butterfly valves can be used. By adjusting the valve opening and closing states, a flexible number of reactor components 204 can be configured.

[0049] Specifically, during the test, any number of 2-4 reactor components 204 can be opened based on experimental requirements. For example, to compare the performance of three catalysts, three reactor components 204 can be opened simultaneously while the remaining reactor components 204 remain closed. Each reactor component 204 maintains identical temperature, pressure, and gas flow conditions. The independent control of its inlet valve 202 and outlet valve 208 ensures consistent testing environments across all reactor components 204. After a test cycle for a reactor component 204 is complete, its inlet and outlet valves 208 can be individually closed for catalyst replacement, while the remaining reactor components 204 remain in continuous operation.

[0050] Compared to existing technologies, the traditional single-reactor unit 204 configuration requires sequential replacement of different catalysts for testing. Each replacement requires system interruption and re-establishment of stable operating conditions, resulting in extended testing cycles and fluctuating operating conditions that affect data accuracy. This solution, by testing multiple reactor units 204 in parallel, allows for simultaneous lifespan assessments of different catalysts under identical flue gas conditions, while also avoiding energy losses caused by frequent downtime.

[0051] Through the above-described technical solution, this application enables simultaneous comparative testing of multiple catalyst formulations, eliminating differences in operating conditions between batches and ensuring data comparability. During testing, the number of wastewater reactor components 204 can be increased or decreased at any time to meet the needs of experiments of varying scales, significantly shortening the catalyst selection cycle. If a reactor component 204 experiences an anomaly, it can be isolated and maintained without affecting overall system operation, improving the stability and continuity of the testing process.

[0052] The present application further proposes that the outer walls of the reactor component 204 and the inlet and outlet pipes 207 are covered with an insulation layer, and the insulation layer material is ceramic fiber or rock wool with a thickness of 30-50 mm.

[0053] Among them, the insulation layer refers to the insulation structure covering the outer surface of the reactor component 204 and its connecting pipes, which can be realized by splicing and wrapping prefabricated fiber material modules to reduce heat loss during the reaction and maintain the temperature stability of the reaction system. Ceramic fiber or rock wool refers to an inorganic fiber material with high temperature resistance. It can be installed in the form of a needle-punched blanket or modular plate. Its low thermal conductivity is used to block the heat transfer path to prevent external ambient temperature fluctuations from interfering with the reaction temperature field. The thickness of 30-50mm refers to the laying size range of the insulation layer perpendicular to the pipe surface. It can be achieved by multi-layer superposition or single-layer customization. This size range can ensure the insulation effect while avoiding excessive increase in equipment volume.

[0054] Specifically, when the temperature-controlled mixed gas from the flue gas heater 108 enters the reactor component 204, the insulation layer blocks heat dissipation through radiation and convection, keeping temperature fluctuations within the target reaction temperature range. When the flue gas temperature is within the catalytic reaction activity window of 200-400°C, the insulation layer effectively reduces the axial and radial temperature gradients within the reactor component 204, ensuring temperature uniformity across the catalyst bed. Especially in field environments with fluctuating wind speeds or diurnal temperature differences, the insulation layer mitigates the external environment's impact on the thermal equilibrium state of the reactor component 204, ensuring that catalyst performance test results more accurately reflect the catalyst's operating conditions in the actual flue.

[0055] Compared to existing technologies, traditional catalyst testing equipment often uses exposed metal pipes or simple asbestos wrapping. This can easily lead to significant heat loss during the transmission of high-temperature flue gas, causing the actual catalyst bed temperature to deviate from the set value. This solution, by precisely controlling the insulation material and thickness, achieves stable temperature maintenance, eliminating catalytic efficiency measurement errors caused by temperature fluctuations.

[0056] Through the above-mentioned technical solution, this application solves the problem of difficult to stably control reaction temperatures in complex industrial environments, effectively avoiding deviations in catalyst activity assessments due to heat loss. The provision of an insulation layer enables the testing system to accurately simulate the thermal environment of the catalyst within a real flue, ensuring consistency between laboratory accelerated aging data and actual field attenuation patterns, providing a reliable thermodynamic parameter benchmark for catalyst formulation optimization.

[0057] The present application further proposes to provide a modular catalyst tray in the reactor component 204 to support the rapid replacement of honeycomb, granular or coated catalysts, and the tray interface size is compatible with non-standard customization.

[0058] The modular catalyst tray refers to a standardized load-bearing structure that can be detachably installed. Specifically, this can be achieved by connecting with positioning clips or flanges. Catalyst fixing slots of different shapes can be set inside the tray. Among them, quick replacement refers to an operating process that does not require tools for disassembly and assembly. Specifically, this can be achieved by using a standard pneumatic clamp in conjunction with a guide rail to form a sealed mating surface between the tray and the cavity of the reactor component 204. Among them, compatibility with non-standard customization means that the tray interface has a size adjustment function. Specifically, this can be achieved by using a replaceable adapter ring or elastic sealing ring structure. The inner diameter of the adapter ring can cover the size range of common industrial catalysts.

[0059] Specifically, the modular catalyst tray features pre-set fixed slots of varying configurations to match the regular pores of honeycomb catalysts, the stacked structure of granular catalysts, or the carrier surface of coated catalysts. Quick-release ports are located along the tray's edge. When the catalyst type needs to be changed, the tray is removed from reactor unit 204 and replaced with a tray adapted for the new catalyst structure. The tray's connection to reactor unit 204 is adjustable, allowing for customized catalyst installations beyond standard sizes by replacing adapter rings or adjusting the compression of the sealing ring.

[0060] Compared to existing technologies, conventional reactor component 204 uses fixed welding or bolting to install the catalyst, requiring complete disassembly of reactor component 204 for replacement. This solution physically separates the catalyst from reactor component 204 through modular trays, enabling testing of different catalyst types without requiring downtime and modification of the reactor component 204's main structure.

[0061] Through the above-mentioned technical solution, this application solves the problems of long experimental cycles and poor operating condition consistency caused by the need to test multiple catalyst formulations sequentially. Testers can quickly swap catalyst trays with different structures within the same reactor component 204, ensuring performance comparisons of different catalysts under the same operating conditions and improving data comparability. The tray interface's dimensional compatibility accommodates the non-standard shapes of customized catalysts required by steel companies, avoiding test interruptions due to dimensional discrepancies.

[0062] The present application further proposes that the high-density sensor array is composed of micro-electrochemical sensors, which are arranged in a grid pattern along the cross-section of the reactor component 204, with a spacing of 5-10 mm between adjacent sensors; the optical imaging unit includes an infrared thermal imager and a visible light CCD camera, which obtains the catalyst surface temperature field and product deposition images through a sapphire observation window.

[0063] The high-density sensor array refers to a gas concentration detection device covering the outlet cross-section of reactor component 204. Specifically, this can be achieved by densely arranging miniaturized electrochemical sensor elements. Each sensor unit independently measures changes in gas concentration in its corresponding area. This grid-like layout ensures spatial resolution across the entire reaction cross-section. A microelectrochemical sensor refers to a gas-sensing element based on a solid electrolyte. Specifically, it can be implemented using a miniaturized sensor chip manufactured using a microelectromechanical system (MEMS) process. High selectivity and rapid response characteristics are achieved through optimized sensitive electrode materials. A grid-like arrangement involves sensors forming a regular array of detection points within a two-dimensional plane at a fixed spacing. Specifically, laser positioning and installation can be used to achieve uniform distribution. The spacing between adjacent sensors must balance spatial resolution and equipment cost. The optical imaging unit refers to a multispectral image acquisition device. Specifically, it can be implemented by coaxially mounting an infrared thermal imager and a visible light camera. The infrared band is used to capture the temperature distribution on the catalyst surface, while the visible light band is used to identify morphological changes caused by product deposition. The sapphire observation window refers to a high-temperature-resistant, light-transmitting window. Specifically, it can be made of sapphire crystal processed into an optically flat lens and embedded in the wall of reactor component 204, allowing optical signals to pass through while maintaining airtightness.

[0064] Specifically, gas concentration detection at the outlet cross-section of reactor component 204 is achieved using a high-density sensor array. Microelectrochemical sensors form a regular detection grid along the cross-section, with the spacing between adjacent sensors controlled within a range of 5-10 mm. Multi-point simultaneous measurement is used to obtain reaction product concentration distribution data for each region downstream of the catalyst. Simultaneously, an infrared thermal imager continuously collects infrared radiation signals from the catalyst surface through a sapphire observation window and converts them into temperature distribution data. A visible light CCD camera simultaneously captures the optical reflectance characteristics of deposits on the catalyst surface and, combined with a multimodal data fusion algorithm, generates a catalyst activity distribution map. For example, when sulfide deposits appear in a localized area, the light reflectance at the corresponding location in the visible light image decreases significantly, and the temperature field data and gas concentration data for that area will exhibit abnormal fluctuations simultaneously.

[0065] Compared to existing technologies, traditional solutions only use a single-point gas analyzer at the outlet of reactor component 204, which fails to capture the spatial heterogeneity of catalyst surface activity distribution. This solution, however, achieves three-dimensional reconstruction of catalyst surface efficiency distribution through the coordinated detection of a high-density gridded sensor array and multispectral imaging. Existing optical detection technology only observes the macroscopic morphology of the catalyst and does not correlate analysis with gas concentration data. This solution, however, simultaneously acquires temperature fields, product deposition images, and local gas concentration data to establish a multidimensional failure pattern recognition model.

[0066] Through the above technical solution, this application solves the problem of catalyst efficiency evaluation deviation caused by single-point sampling. It realizes spatial interpolation calculation of local concentration field through gridded high-density sensor array, combines optical imaging data to identify temperature anomalies and product deposition areas, accurately locates the catalyst deactivation position and quantifies the degree of activity attenuation, and provides accurate data support for targeted optimization of catalyst formulation and regeneration process.

[0067] The present application further proposes that the local efficiency calculation unit 303 executes a multimodal data fusion algorithm to weightedly fuse the discrete concentration data of the sensor array with the continuous surface data of the optical imaging to dynamically correct the overall conversion efficiency value of the catalyst.

[0068] Multimodal data fusion algorithms are computational methods for spatiotemporal registration and feature extraction of heterogeneous data. These algorithms can be implemented using a combination of Kalman filtering and support vector regression to eliminate sensor noise and resolution differences in optical data. Discrete concentration data refers to gas concentration values at discrete spatial points in the downstream cross-section of the catalyst, collected by an array of microelectrochemical sensors. This can be achieved using a multi-point synchronous sampling circuit and is used to quantitatively reflect differences in local reaction efficiency. Continuous surface data from optical imaging refers to two-dimensional images of the catalyst surface temperature gradient and product deposition morphology, acquired by infrared thermal imagers and visible light CCD cameras. This can be achieved using an algorithm that maps image grayscale values to temperature radiation intensity, and is used to identify the distribution characteristics of physically deactivated regions. Weighted fusion assigns confidence weights to each data source based on sensor spacing and optical resolution. This can be achieved using an adaptive weight allocation model to balance the complementary advantages of discrete data accuracy and continuous data coverage. Dynamically correcting the overall catalyst conversion efficiency value refers to real-time updating of the catalyst's comprehensive performance indicators based on a fused efficiency distribution heat map. This can be achieved using a sliding time window and exponential smoothing to eliminate the influence of single-point sampling errors on the overall evaluation.

[0069] Specifically, during the operation of the catalyst, the sensor array collects discrete concentration data of the outlet cross-section of the reactor component 204 in a grid format, while the optical imaging unit obtains continuous images of the catalyst surface temperature field and sediment distribution. After the two data types are aligned in time and space, the confidence weight coefficient of each spatial position is calculated based on the sensor spacing and image resolution. For example, a higher weight is given to concentration data in areas with dense sensors, and a higher weight is given to optical data in areas with obvious image features. The weighted multi-source data is spatially interpolated to generate a two-dimensional efficiency distribution heat map, and the dynamically adjusted overall conversion efficiency value is calculated based on the integral calculation of the heat map. When an abnormal decrease in efficiency is detected in a local area, the algorithm automatically increases the weight of the data in that area in the overall calculation to ensure that the deactivation characteristics are accurately captured.

[0070] Compared to existing technologies, traditional methods rely solely on single-point gas analyzers or optical imaging techniques, failing to effectively integrate the advantages of discrete concentration measurements with continuous surface data. This invention utilizes a multimodal data fusion algorithm to organically combine the quantitative detection capabilities of electrochemical sensors with the spatial resolution capabilities of optical imaging. This solves the assessment bias caused by single-point sampling while also avoiding the risk of misjudgment caused by dust obstruction or temperature drift associated with single detection methods.

[0071] Through the above-mentioned technical solution, this application can accurately identify the non-uniform distribution of catalyst surface activity, locate the spatial location of deactivated areas such as dust accumulation and sulfur poisoning, and provide data support for targeted cleaning or regeneration. Furthermore, a dynamic correction mechanism can eliminate the interference of local abnormal data on overall efficiency assessment, improve the reliability of the catalyst life prediction model, and shorten the field testing cycle.

[0072] This application further proposes a machine learning life prediction module, which receives time series data of efficiency distribution heat map, extracts local deactivation features through convolutional neural network, and outputs the catalyst remaining life prediction value and deactivation risk area marking map.

[0073] The machine learning life prediction module refers to a catalyst life analysis unit built based on a supervised learning algorithm. Specifically, it can be implemented using a hybrid architecture of recurrent neural networks and convolutional neural networks, and is used to process data on the spatiotemporal evolution of catalyst surface efficiency. The time series data of the efficiency distribution heat map refers to a multidimensional dataset consisting of a two-dimensional heat map of efficiency values for different regions of the catalyst, arranged in chronological order. Specifically, it can be implemented using a sampling frequency of 1 to 24 hours, and is used to reflect the decay trajectory of catalyst activity over time. The convolutional neural network refers to a deep learning model with a convolution kernel structure, and can be implemented by stacking multiple two-dimensional convolutional layers with pooling layers. It is used to capture deactivation characteristics caused by sulfur poisoning or dust accumulation in localized areas of the catalyst surface. The remaining life prediction value refers to the remaining effective time for the catalyst to maintain the target conversion efficiency, calculated using a regression model. Specifically, it can be implemented by training a neural network using a mean squared error loss function, and is used to quantify the catalyst's service life under actual operating conditions. The deactivation risk area marker map refers to a visual image that identifies spatial locations on the catalyst surface where the probability of deactivation exceeds a threshold. Specifically, it can be implemented using a probabilistic threshold segmentation algorithm combined with morphological processing, and is used to guide catalyst formulation improvements or regeneration.

[0074] Specifically, the module continuously collects heat maps of catalyst efficiency distribution generated during the operation of reactor component 204 to construct a multidimensional dataset that captures spatial efficiency heterogeneity and temporal decay trends. After receiving the heat map sequence at the input layer, the convolutional neural network extracts local texture features by sliding the convolution kernel across the spatial dimension, identifying activity decay patterns caused by sulfur deposition or dust coverage. The network uses strided convolution operations in the temporal dimension to capture differences in the development rates of different deactivation mechanisms. The trained model can predict the overall remaining life of the catalyst based on current and historical efficiency distribution data and generate a binary image at the output layer that marks high-risk areas.

[0075] Compared to existing technologies, traditional catalyst lifespan assessment methods rely solely on concentration data from the outlet of a single reactor component 204, failing to capture the spatial heterogeneity of catalyst surface activity. Using a multi-reactor component 204 architecture with only a single-point gas analyzer, lifespan predictions fail to reflect the overall performance degradation caused by localized deactivation. This approach fuses discrete concentration data from a high-density sensor array with continuous surface data from optical imaging to construct a spatiotemporal feature matrix recognizable by a convolutional neural network, thereby accurately quantifying the impact of localized deactivation on overall lifespan.

[0076] Through this technical solution, the present application can identify localized deactivation areas in advance during accelerated aging testing based on the spatiotemporal evolution of catalyst surface efficiency, shortening the experimental cycle by over 50% compared to traditional continuous testing methods. The resulting deactivation risk area marker map can be used directly to guide catalyst support structure optimization, such as adding an anti-sulfur coating to the marked areas or adjusting the honeycomb pore size distribution, thereby improving the catalyst's field applicability.

[0077] The present application further proposes that the modulated gas pipeline 104 is connected to the SO2, H2O vapor and dust generating device, and the accelerated aging gas with a concentration of 1.5-3 times that of the on-site flue gas is injected into the flue gas mixer 107 through the modulated gas mass flow meter 106.

[0078] The SO2 generator refers to a device capable of generating sulfur dioxide gas. Specifically, this can be achieved using a chemical reaction device that burns sulfur or decomposes sulfites, and is used to simulate the corrosive sulfide environment in industrial flue gas. The H2O vapor generator refers to a device that produces a controllable amount of water vapor, and can be implemented using an electrically heated evaporator or steam generator. This device is used to replicate the effect of flue gas humidity on catalyst pore blockage. The dust generator refers to a device that produces solid particulate matter with a specific particle size distribution. Specifically, this can be implemented using an aerosol generator coupled with a graded screening system, and is used to simulate dust accumulation on the catalyst surface. The modulated gas mass flowmeter 106 is a metering device that precisely controls the delivery ratio of each gas component. Specifically, this can be implemented using a thermal or differential pressure flow sensor coupled with a PID controller, ensuring that the concentration gradients of each pollutant in the accelerated aging gas precisely match the preset ratios. Setting the accelerated aging gas concentration to 1.5-3 times the actual flue gas concentration proportionally increases the concentration of the pollutants under actual operating conditions. This can be achieved by adjusting the output power of each generator in conjunction with the set value of the reaction gas mass flowmeter 203, and is used to replicate the catalyst deactivation phenomenon under long-term exposure within a controllable timeframe.

[0079] Specifically, during the catalyst evaluation process, SO2, H2O vapor, and dust generators are activated simultaneously, generating corrosive gases, high-humidity steam, and dust particles, respectively, with typical characteristics of industrial flue gas. Based on a preset program and on-site flue gas measurement data, the modulated gas mass flowmeter 106 dynamically adjusts the flow rates of each component by a factor of 1.5-3, forming a mixed flow of accelerated aging gases. This mixed flow is introduced into the flue gas mixer 107 and thoroughly blended with the original flue gas, creating a test atmosphere with enhanced corrosion. By precisely controlling the concentration multiplication factor of each pollutant, the test cycle can be shortened under laboratory conditions while maintaining the actual corrosion pattern of the catalyst caused by the synergistic effects of multiple components.

[0080] Compared to existing technologies, traditional accelerated aging methods simply increase the concentration of a specific pollutant, failing to account for complex failure mechanisms such as sulfate crystallization caused by the coexistence of SO2 and H2O vapor, and capillary blockage caused by dust adsorption and humidity coupling. This approach, by synergizing multiple component generators, more realistically simulates the combined chemical and physical interactions of catalysts in actual industrial flue gas, significantly improving the consistency between laboratory aging data and field deactivation patterns.

[0081] Through the above technical solution, this application solves the problem of laboratory and field data deviation caused by the increase in the concentration of a single pollutant in the existing accelerated aging method. It can accurately reproduce the deactivation process of the catalyst under complex flue gas conditions within a controllable experimental cycle, and provide high-fidelity training data for the life prediction model, thereby shortening the catalyst R&D and verification cycle.

[0082] The present application further proposes setting a pressure difference sensor between the induced draft fan 401 and the return air duct 403. When the pressure difference between the on-site air intake point and the return air point is greater than the system resistance, the induced draft fan 401 is automatically shut down and the natural pressure difference driving mode is enabled.

[0083] Among them, the pressure difference sensor refers to a device used to monitor the pressure difference between the on-site flue gas intake point and the return gas point in real time. Specifically, it can be implemented by a capacitive or piezoresistive sensor. By detecting the pressure difference between the two ends, it is determined whether the natural pressure difference driving conditions are met. System resistance refers to the total flow resistance caused by pipeline friction, valve throttling and pressure drop of the reactor component 204 when the gas flows in the evaluation system. Specifically, the reference value can be obtained through experimental calibration or fluid mechanics calculation. The natural pressure difference driving mode refers to an operating mode that uses the pressure gradient existing in the on-site flue itself to maintain gas flow. When the pressure at the gas intake point is higher than the return gas point, flue gas circulation can be achieved without external power.

[0084] Specifically, the output signal of the differential pressure sensor is continuously monitored during the test. If the natural pressure differential in the flue gas exceeds the total system resistance, the control system immediately shuts off the power to the induced draft fan 401. The flue gas then relies on the flue gas's inherent pressure gradient to maintain its path through the evaluation system. If the natural pressure differential is insufficient to overcome the system resistance, the induced draft fan 401 restarts to compensate for the pressure differential shortfall. This process is seamlessly switched through a logic controller, ensuring continuous gas flow while avoiding energy waste.

[0085] In some specific embodiments, the differential pressure sensor can have a range of ±500 Pa to ±2000 Pa and should be installed at least two pipe diameters away from the inlet of the induced draft fan 401 to eliminate turbulent flow interference. In the natural pressure differential drive mode, flue gas flow regulation can be achieved by adjusting the opening of the inlet valve 202 of the reactor component 204, for example, controlling the valve opening within a range of 30%-70% to match pressure fluctuations.

[0086] Compared to existing technologies, traditional evaluation systems rely entirely on induced draft fans 401 to maintain gas flow. Even when a large natural pressure differential exists in the on-site flue, the fans continue to operate, wasting energy and increasing equipment wear. This solution dynamically identifies natural pressure differential conditions and activates power equipment only when necessary, reducing energy consumption and extending equipment life.

[0087] Through the above technical solution, this application effectively solves the problem of serious energy waste in traditional evaluation systems under flue pressure difference fluctuation conditions, realizes the adaptive matching of the test device operation mode with the actual on-site conditions, and improves the applicability and operation stability of the system in different industrial scenarios.

[0088] The present application further proposes an on-site catalyst evaluation method based on surface detection, and the specific steps are as follows: injecting modulated gas into the on-site flue gas to accelerate catalyst aging; heating the mixed flue gas and distributing it to multiple reactor components 204; synchronously collecting multi-dimensional data of the downstream cross-section of the catalyst through a high-density sensor array and an optical imaging unit; using a spatial interpolation algorithm to generate an efficiency distribution heat map, and combining a machine learning model to predict the life decay trend; and screening the optimal catalyst formula and process parameters based on the parallel test results of multiple reactor components 204.

[0089] Among them, the modulated gas refers to a mixed gas used to simulate complex flue gas components. Specifically, it can be achieved by using a combination of SO2, H2O vapor and dust. By increasing the concentration of pollutants, the deactivation phenomena such as sulfur poisoning and dust accumulation in the catalyst aging process are accelerated. Among them, the spatial interpolation algorithm refers to a mathematical method for reconstructing the continuous surface distribution based on discrete sensor data. Specifically, it can be implemented by using Kriging interpolation or radial basis function interpolation, and is used to expand the local concentration measurement value into the efficiency distribution of the entire surface of the catalyst. Among them, the machine learning model refers to a prediction algorithm for analyzing the dynamic changes of the efficiency distribution. Specifically, it can be implemented by using a convolutional neural network or a long short-term memory network. By extracting the local deactivation features in the heat map, a life decay trend prediction model is established. Among them, the parallel testing of multiple reactor components 204 refers to the simultaneous operation of multiple independently controlled reactor components 204. Specifically, it can be achieved by independently opening and closing the valves of each reactor component 204, supporting the synchronous performance comparison of different catalyst formulations or process parameters under the same flue gas conditions.

[0090] Specifically, on-site flue gas is first mixed with a modulated gas to create an accelerated aging environment. After heating, it is distributed to multiple parallel reactor components 204. Each reactor component 204 houses a different catalyst. A high-density sensor array collects cross-sectional concentration distribution data, while an optical imaging unit captures catalyst surface temperature and deposition images. Discrete concentration data and continuous image data are fused using a spatial interpolation algorithm to generate an efficiency distribution heat map reflecting the heterogeneity of catalyst surface activity. This heat map, when input into a machine learning model, can identify the evolution of localized deactivation regions and predict overall lifespan degradation trends. Ultimately, the optimal catalyst formulation and process parameter combination are selected using parallel testing data from multiple reactor components 204.

[0091] Compared to existing technologies, traditional methods rely on single-point sampling and testing of a single reactor component 204, failing to capture differences in catalyst surface efficiency distribution and only capable of testing one catalyst formulation at a time. This method uses surface detection technology to simultaneously acquire information on the full catalyst surface activity. Combined with the parallel operation of multiple reactor components 204, this method can rapidly compare the performance differences of multiple catalysts under the same flue gas conditions. It also enables dynamic prediction of catalyst life degradation trends based on machine learning models.

[0092] Through the above technical solution, this application solves the problem of deviation in the evaluation of local deactivation of catalysts caused by single-point sampling, shortens the screening cycle of multiple catalyst formulations, improves the matching degree between accelerated aging experiments and actual on-site working conditions, and provides reliable technical support for the rapid selection and life management of catalysts in industrial flue gas treatment.

[0093] Anything not described in the present invention can be achieved by adopting or drawing on existing technologies.

[0094] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0095] The foregoing is merely an embodiment of the present invention and is not intended to limit the present invention. It will be apparent to those skilled in the art that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of the present invention.

Claims

1. An on-site catalyst evaluation system based on surface detection, characterized in that: It comprises an intake atmosphere modulation module (1), a reactor module (2), a gas detection module (3) and an exhaust gas treatment module (4) which are connected in sequence; The air intake atmosphere modulation module (1) comprises an air intake flue (101), an air intake shut-off valve (102), an air intake regulating valve (103), a flue gas mixer (107), a modulation gas pipeline (104), a modulation gas shut-off valve (105), a modulation gas mass flow meter (106), a flue gas heater (108) and a temperature sensor (109). The air intake flue (101) and the modulation gas pipeline (104) are respectively connected to the flue gas mixer (107). After mixing, the flue gas is heated by the flue gas heater (108) and the temperature is closed-loop controlled by the temperature sensor (109). The reactor module (2) comprises at least two reactor components (204) arranged in parallel, each inlet pipe (201) is provided with a reaction gas mass flow meter (203) and an inlet temperature meter (205), the outlet pipe (207) is provided with an outlet temperature meter (206), and a catalyst is replaceably installed in the reactor component (204); The gas detection module (3) comprises a mixer post-measuring point (301) arranged at the exhaust end of the flue gas mixer (107) and a reactor component post-measuring point arranged at the outlet of the reactor component (204), and also comprises a surface detection unit (302) for detecting gas data discharged from the mixer post-measuring point (301) and the reactor component post-measuring point, wherein the surface detection unit (302) comprises a high-density sensor array and an optical imaging unit, and a local efficiency calculation unit (303) connected to the two, wherein the calculation unit (303) generates a catalyst surface efficiency distribution heat map through a spatial interpolation algorithm; The tail gas treatment module (4) comprises an induced draft fan (401) and a return air flue (403), which returns the tested flue gas to the on-site flue. The return air flue (403) is provided with a return air shut-off valve (402).

2. The on-site catalyst evaluation system based on surface detection according to claim 1, characterized in that: The reactor module (2) comprises 2-4 independently controlled reactor components (204), and the inlet valve (202) and outlet valve (208) of each reactor component (204) can be opened and closed individually to adjust the number of reactor components (204) participating in the test.

3. The on-site catalyst evaluation system based on surface detection according to claim 2, characterized in that: The outer walls of the reactor component (204), the inlet pipe (201) and the outlet pipe (207) are covered with an insulation layer. The insulation layer material is ceramic fiber or rock wool, and the thickness is 30-50 mm.

4. The on-site catalyst evaluation system based on surface detection according to claim 3, characterized in that: A modular catalyst tray is provided in the reactor component (204), supporting the rapid replacement of honeycomb, granular or coated catalysts, and the tray interface size is compatible with non-standard customization.

5. The on-site catalyst evaluation system based on surface detection according to claim 3, characterized in that: The high-density sensor array is composed of micro electrochemical sensors, which are arranged in a grid shape along the cross section of the reactor component (204), with the spacing between adjacent sensors being 5-10 mm; the optical imaging unit includes an infrared thermal imager and a visible light CCD camera, which obtains the catalyst surface temperature field and product deposition images through a sapphire observation window.

6. The on-site catalyst evaluation system based on surface detection according to claim 1, characterized in that: The local efficiency calculation unit (303) executes a multimodal data fusion algorithm to weightedly fuse the discrete concentration data of the sensor array with the continuous surface data of the optical imaging, and dynamically corrects the overall conversion efficiency value of the catalyst.

7. The on-site catalyst evaluation system based on surface detection according to claim 2, characterized in that: It also includes a machine learning life prediction module, which receives time series data of the efficiency distribution heat map, extracts local deactivation features through a convolutional neural network, and outputs a predicted value of the catalyst's remaining life and a deactivation risk area marking map.

8. The on-site catalyst evaluation system based on surface detection according to claim 1, characterized in that: The modulated gas pipeline (104) is connected to the SO2, H2O vapor and dust generating device, and injects accelerated aging gas with a concentration of 1.5-3 times that of the on-site flue gas into the flue gas mixer (107) through the modulated gas mass flow meter (106).

9. The on-site catalyst evaluation system based on surface detection according to claim 1, characterized in that: A pressure difference sensor is provided between the induced draft fan (401) and the return air duct (403). When the pressure difference between the on-site air intake point and the return air point is greater than the system resistance, the induced draft fan (401) is automatically shut down and the natural pressure difference driving mode is enabled.

10. A method for on-site catalyst evaluation based on surface detection, characterized in that: The specific steps are as follows: S1. Injecting modulated gas into the on-site flue gas to accelerate catalyst aging; S2, heating the mixed flue gas to 200-400° C. and then distributing it to a plurality of reactor components (204); S3, synchronously collecting multi-dimensional data of the downstream cross section of the catalyst through a high-density sensor array and an optical imaging unit; S4. Use spatial interpolation algorithms to generate efficiency distribution heat maps and combine them with machine learning models to predict lifespan attenuation trends. S5. Screening the optimal catalyst formula and process parameters based on the parallel test results of multiple reactor components (204).

Images