Method for enhancing photoelectric composite catalytic reaction efficiency by self-adaptive multi-field coupling

Abstract

The invention belongs to the technical field of catalytic chemistry, and particularly relates to a method for enhancing photoelectric composite catalytic reaction efficiency through self-adaptive multi-field coupling. Specifically, firstly, a closed columnar reactor composed of an LED array, a reflection / lens system, a copper-nickel alloy cathode, a Ru-doped PbO2 anode, a Ti5O7 / Ti3O4 composite photocatalytic parallel-connection multi-tube plate and the like is built, a distributed silicon photodiode and a quantum dot fluorescent probe sensor are installed, a machine learning algorithm is utilized, and according to the reactant concentration and the by-product generation condition, the quantum dot fluorescent probe sensor and the quantum dot fluorescent probe sensor are analyzed. Light field conditions (wavelength, intensity and distribution) used by photocatalysis, electric field conditions (current, voltage and power line distribution) used by electrocatalysis and flow field conditions (flow velocity, flow rate and distribution) are dynamically adjusted to optimize the chemical reaction rate, and an AI model is combined to predict an optimal multi-field coupling setting scheme, so that the efficiency is improved, the energy consumption is reduced, and the cost is reduced. Meanwhile, the effect of enhancing the catalytic conversion reaction is achieved.

Description

1. Technical Field

[0001] The present invention belongs to the field of catalytic chemistry technology, and in particular, relates to a method for enhancing the efficiency of photoelectric composite catalytic reactions by adaptive multi-field coupling. The method focuses on solving the problems of the existing photoelectric composite reaction device, such as complex equipment structure, unclear mechanism, mutual interference of photoelectric reactions, low energy efficiency, and limited adjustability, so as to improve efficiency and reduce energy consumption, and at the same time achieve the effect of enhancing catalytic conversion reactions. 2. Background Technology

[0002] Photoelectric composite reactors have shown extensive application potential in the fields of environment, energy and chemical synthesis. In terms of environmental remediation, photoelectric composite catalytic technology is used to degrade organic pollutants in water and air. For example, the degradation of drug residues and difficult-to-degrade organic pollutants through photoelectric composite catalytic reactors has significantly improved efficiency compared with traditional photocatalysis. In the field of energy, photoelectric composite catalytic carbon dioxide reduction reactors can convert CO2 into fuels such as methane and methanol, and its conversion efficiency can reach more than 60% under certain conditions, providing important technical support for carbon neutrality. In chemical synthesis, photoelectric composite reactors are used to efficiently synthesize high-value-added chemicals. For example, CH bond activation can be achieved through photoelectric composite catalysis, and its reaction rate is more than 3 times higher than that of traditional methods. In addition, photoelectric composite catalytic reactors have also made significant progress in the fields of solar fuel preparation and water decomposition to produce hydrogen. These applications show that photoelectric composite reactors are of great value in solving environmental and energy problems.

[0003] Although photoelectric composite reactors have been widely used in many fields, their structural complexity and efficiency issues limit their further promotion. Existing reactors usually adopt a fixed structure design, such as a single-chamber or double-chamber reactor, and their electrode layout, light source position and reaction chamber shape are often not adjustable, resulting in limited optimization of reaction conditions. In addition, the multi-field coupling effects such as light, electricity, and heat inside the reactor are difficult to accurately control. For example, uneven light intensity distribution will lead to low local reaction efficiency, and excessive electric field intensity may induce side reactions. These problems make it difficult to break through the overall efficiency of the reactor. For example, the quantum efficiency of some photoelectric composite catalytic reactors is only 20%-30%, which is far below the theoretical value. At the same time, the scalability of existing reactors is poor, and it is difficult to adapt to the needs of reactions of different scales and types. For example, in scale-up experiments, the reaction efficiency often drops significantly. These shortcomings show that the existing photoelectric composite reactors have significant defects in structure and performance, and they urgently need to be improved through technological innovation.

[0004] In order to improve the defects of photoelectric composite reactors, domestic and foreign research teams and enterprises have actively explored structural design and performance optimization. For example, the team of Professor Feng Xinjian of Soochow University developed a three-phase photocatalytic reaction system. By introducing the gas-liquid-solid three-phase interface, the oxygen transmission efficiency and photogenerated carrier separation efficiency were significantly improved, and the quantum efficiency was increased by 5 times (J.Am.Chem.Soc., 2017, 139, 12402). Sun Yat-sen University (CN202010365016.8) proposed a new photoelectric composite catalytic reactor, which adopts an air diffusion cathode design and improves the pollutant degradation efficiency by in-situ generation of hydroxyl radicals (·OH). Its energy utilization rate is 40% higher than that of traditional reactors. In addition, the gas-solid phase photocatalytic reactor developed by Bofeilai Technology (CN202320652037.7) realizes full contact between the reaction gas and the catalyst through the gas circulation loop design, which increases the reaction efficiency by more than 30%. Sinopec (CN203525696U) has developed a fluidized bed photochemical reactor that significantly improves the reaction uniformity and stability by optimizing the distribution and flow pattern of photocatalysts. These innovations show that the performance of photoelectric composite reactors can be significantly improved through structural optimization and multi-field coupling regulation, providing an important reference for future technological development. But in general, the design and application of current photoelectric composite reactors are still focused on simple improvements to the types of traditional reactors, and there are still defects in two extremely important aspects:

[0005] 1) Improve the adaptability of the reactor so that the energy efficiency of light irradiation can accurately cover the points required for photocatalysis, and minimize the mutual interference between photoelectric reactions;

[0006] 2) In order to match the adaptability of the reactor, improvements are made to the electrocatalysis, photocatalysis and supporting photoelectric management systems. III. Summary of the invention

[0007] The present invention is made in view of the problems existing in the prior art, and aims to significantly improve the chemical reaction rate by optimizing the interaction between the four key physical fields of light, electricity, flow and heat. This method not only involves precision-designed hardware facilities, but also integrates advanced intelligent algorithms, so that the entire system can self-adjust according to real-time data to achieve optimal performance.

[0008] First, in order to match the adaptability of the reactor, the management systems such as electrocatalysis, photocatalysis and supporting photoelectric sensors are improved. The selected components need to have high activity, high stability and high adjustability to form a mutually matched technical system. Ru doping improves the crystal structure of PbO2, increases the number of surface active sites, and further promotes the occurrence of redox reactions, making the Ru-doped PbO2 anode more than three times more efficient than the traditional PbO2 anode; Ti5O7 / Ti3O4 each has unique physical and chemical properties. Mixing the two in a mass ratio of 4:1 can form an efficient photocatalytic layer, which not only inherits the good light absorption characteristics and stability of TiO2, but also greatly improves the separation efficiency of photogenerated carriers due to the synergistic effect between Ti5O7 / Ti3O4, and its quantum yield is about 20% higher than that of a single TiO2 material. In addition, the Ti2N3 on the surface layer plays a role in protecting the bottom layer, while also enhancing the light absorption capacity and catalytic effect; Furthermore, the combination of LED array and reflection / lens system ensures the optimization of lighting conditions. LED light sources are selected for their high energy efficiency, long life and wide wavelength range. The wavelength (300-800nm) and intensity (0.1-5W / cm 2 ), which can meet the needs of different catalysts. The reflector and lens system ensures that the light is evenly irradiated to the entire reaction area, avoiding the problem of uneven local illumination. This design is crucial to improving light utilization, especially in photocatalytic reactions, where uniform and high-intensity illumination is one of the key factors for achieving efficient reactions. Finally, distributed silicon photodiodes and quantum dot fluorescent probe sensors are installed to monitor key parameters in the reaction process in real time. These sensors can provide information about changes in reactant concentrations, by-product generation, and environmental conditions (such as temperature, pH value, and oxygen content), helping the intelligent control system to make accurate adjustment decisions, ensuring full coverage of the reaction area, and making data acquisition more detailed and reliable, which plays an irreplaceable role in optimizing multi-field coupling conditions and improving reaction efficiency.

[0009] The light field, electric field, flow field and thermal field are interrelated and key factors that jointly affect the reaction efficiency in the photoelectric composite reactor. The main input variables are as follows. The light field is mainly provided by an adjustable light source, and its intensity I (W / cm 2 ) and wavelength λ(nm) play a decisive role in the excitation of the catalyst; the electric field promotes the electron transfer process by applying voltage V(V) and current I(A) between the cathode and the anode; the flow field refers to the flow characteristics of the reactant liquid or gas, including flow velocity v(m / s) and flow rate Q(L / min); and the thermal field maintains the stability of the reaction environment by controlling the temperature T(℃). There is a complex nonlinear relationship between the above four fields. For example, the light intensity I and temperature T will directly affect the reaction rate constant k on the electrode surface, that is:

[0010] k=A·eE a / RT

[0011] Among them, A is the frequency factor, E a is the activation energy, and R is the ideal gas constant. In addition, the potential distribution in the electric field will also affect the particle mobility, thereby changing the flow field conditions. In order to achieve multi-field coupling, the present invention introduces machine learning algorithms such as enhanced random forest (RF++), and establishes a prediction model by learning a large amount of experimental data to guide the optimal configuration of each field parameter.

[0012] The RF++ algorithm is based on an integrated learning framework and gradually adds weak classifiers through the Boosting mechanism to improve overall performance. It can perform well when processing high-dimensional complex data and has good generalization capabilities. During the training process, historical experimental results are used as input features, covering reaction rate information under light, electric field and flow field conditions, to build a powerful prediction model. When actually running, the system collects current sensor data, including but not limited to light intensity, voltage, current, flow rate and other information, and inputs it into the RF++ model to obtain the most suitable operating parameter recommendations for given conditions.

[0013] In order to achieve "adaptive" control, the control system is developed in Python language, integrating RF++ model and other necessary functional modules. The software can not only dynamically adjust the light source output parameters such as wavelength, intensity and its distribution, but also optimize the voltage and current settings between the electrode plates, and even accurately control the flow rate and flow of the reaction liquid. For example, in terms of light field control, the PID controller maintains the set value stable, with an error of no more than ±0.05W / cm 2 In the electric field regulation, an overcurrent protection circuit is built in to prevent the equipment from being damaged due to abnormal conditions. More importantly, with the help of AI models to predict the optimal multi-field coupling solution, the system can continuously iterate and optimize, continuously improve the chemical reaction rate while reducing energy consumption, and ultimately achieve the goal of energy saving and consumption reduction.

[0014] In order to achieve the above object, the technical solution of the present invention is a method for enhancing the efficiency of photoelectric composite catalytic reaction by adaptive multi-field coupling, characterized in that the method specifically comprises the following steps:

[0015] Step 1: construct a closed cylindrical photoelectric composite reaction device with an aspect ratio of 3:1, a diameter of 20 cm and a height of 60 cm; the reaction chamber is equipped with an adjustable light source, supporting a wavelength range of 300 to 800 nm and an intensity adjustment range of 0.1 to 5 W / cm 2, and is equipped with an evenly distributed LED array to achieve optimized light distribution; the LED array consists of multiple ultraviolet LEDs, which ensure that the light is evenly irradiated to the entire reaction area through a reflector and lens system; the light source power supply system adopts constant current source control, and the current adjustment accuracy is ±0.1% to ensure stable light intensity output; the light source control method includes manual adjustment and automatic feedback control system, the latter is based on the built-in light sensor to monitor the light intensity changes in real time and dynamically adjust to the optimal value; the reactor is equipped with a feed port and a discharge port. The feed port is located on one side of the bottom of the reactor and is made of stainless steel with a quick connector to facilitate connection to an external feeding system. The inlet diameter is 10mm and the flow control range is 0.1-5L / min; the discharge port is located on the other side of the top and is also controlled by a stainless steel valve to ensure the product Smooth discharge; the dispersion system is composed of a porous distributor to prevent channeling, short circuit, open circuit and other problems caused by direct feeding; the reaction temperature is maintained by jacket cooling water circulation, the cooling water flow rate is 0.5-2L / min, the temperature control accuracy is ±0.5℃, and the operating temperature range is from room temperature to 100℃; the exhaust port is set at the top of the reactor, with a diameter of 8mm and a gas filter to prevent harmful gas from escaping; the electrical system includes power supply, cables and connectors; the electrocatalytic system consists of cathode, anode and supporting cables and power supply, the cathode is a metal wire mesh structure, made of copper-nickel alloy, with a pore size of 1mm and a thickness of 1mm; the anode is a porous mesh structure, made of a porous titanium substrate loaded with Ru-doped PbO2 active material, which is loaded by pulse electrodeposition, with a loading amount of 25mg / cm 2 The power supply is a DC regulated power supply with a voltage adjustment range of 0 to 10V and an accuracy of ±0.1%. The cable is made of high temperature resistant and corrosion resistant silicone rubber insulated wire, and the connection method is a waterproof plug and socket. The photocatalytic parallel multi-tube plate has a base made of highly transparent borosilicate glass in the form of a hollow tube with a length of 50cm, an outer diameter of 10mm, and an inner diameter of 8mm. The bottom layer is loaded with a layer of Si by ion sputtering with a thickness of 1μm, which plays a protective role. The middle layer is a Ti5O7 / Ti3O4 mixed (mass ratio of 4:1) nanoparticle coating, which is obtained by supersonic flame sintering and has a thickness of 1 to 2μm. It serves as the main photocatalytic active layer and has a certain thermal catalytic effect. The surface layer is Ti2N3, which is obtained by ion sputtering and has a thickness of 50nm. While enhancing the light absorption capacity and catalytic activity, it also protects the Ti5O7 / Ti3O4 mixed layer.

[0016] Step 2, during the catalytic reaction, the liquid flow enters the reactor from the bottom of the feed port, is evenly dispersed by a porous distributor, and then penetrates the cathode metal mesh plate, the photocatalytic parallel multi-tube plate, and the anode porous mesh plate layer by layer, flows out through the discharge port, and completes the reaction under the action of the light field and the electric field; during the reaction, the light field, the electric field, and the flow field can be automatically adjusted to achieve multi-field coupling; a distributed silicon photodiode and a quantum dot fluorescent probe are installed in the reaction chamber as photoelectric sensors for real-time monitoring of reactant concentration changes and by-product generation; the sensor density is at least 10 detection points per cubic centimeter to ensure full coverage of the reaction area; in addition, temperature, pH value, and oxygen content sensors are configured to assist in monitoring the reaction environment conditions, and form a data set of operating condition parameters and output corresponding product conversion rate, selectivity, and stability;

[0017] Step 3: Apply the enhanced random forest (RF++) algorithm for data processing and prediction to add a feature selection module, introduce the Boosting mechanism in ensemble learning, and improve the overall performance by gradually adding weak classifiers; the training data set is composed of historical experimental results, covering the reaction rate information under different lighting, electric field and flow field conditions;

[0018] Step 4: Build an intelligent control system based on Python language control software, integrate the RF++ model, and dynamically adjust the light field (wavelength, intensity, distribution), electric field (current, voltage, power line distribution), and flow field (flow velocity, flow rate, distribution) parameters according to sensor feedback;

[0019] Step 5: Based on the prediction results of the RF++ model, the intelligent control system automatically adjusts the light source output parameters so that the wavelength varies within the range of 300 to 800 nm and the intensity is within the range of 0.1 to 5 W / cm 2 The LED array layout is optimized to ensure uniform light distribution. The PID controller maintains the set value stable with an error of no more than ±0.05W / cm 2 ;

[0020] Step 6: The intelligent control system adjusts the voltage and current between the electrode plates according to the prediction model, changes the distribution of power lines, and optimizes the electrocatalytic effect; the control system has a built-in overcurrent protection circuit to prevent equipment damage and ensure safe operation;

[0021] Step 7, using a peristaltic pump and a Venturi effect nozzle to accurately control the flow rate (0.1-100 mL / min) and flow rate of the reaction liquid, combined with real-time monitoring by an ultrasonic flow meter to ensure that the flow field conditions meet the optimal solution; the control system supports a closed-loop feedback mechanism to automatically correct deviations and maintain set accuracy;

[0022] Step 8: Combine the AI ​​model to predict the optimal multi-field coupling setting plan, and through continuous iterative optimization, improve the chemical reaction rate, reduce energy consumption, and achieve the goal of energy saving and consumption reduction; the intelligent control system records the parameters after each optimization and forms a database for reference for subsequent improvements.

[0023] It is further defined that the photoelectric sensor described in step 2 also includes a Raman spectrometer and a Fourier transform infrared spectrometer (FTIR) for providing more detailed molecular structure information to assist in determining the reaction progress.

[0024] It is further defined that the machine learning algorithm described in step 3 can also be a gradient boosting tree (GBT++), which enhances the ability to capture nonlinear relationships and improves prediction accuracy by introducing feature interaction terms and regularization terms.

[0025] It is further defined that the light field condition control described in step 5 is further refined to use a microlens array to adjust the light distribution to ensure that the photocatalyst surface receives uniform and high-intensity light, thereby maximizing the photocatalytic efficiency.

[0026] It is further defined that the flow field condition control described in step 7 also involves the use of turbulence promoters and static mixers to improve the liquid flow characteristics, ensure sufficient contact between the reactants and the catalyst, and further increase the reaction rate.

[0027] The present invention is beneficial in that:

[0028] 1) Adopt a highly efficient closed columnar reactor composed of LED array, reflection / lens system, copper-nickel alloy cathode, Ru-doped PbO2 anode and Ti5O7 / Ti3O4 composite photocatalytic parallel multi-tube plate, install distributed silicon photodiodes and quantum dot fluorescent probe sensors, accurately control light, electricity, flow and thermal field to avoid energy loss.

[0029] 2) Through adaptive multi-field coupling technology, the synergy of the four physical fields of light, electricity, flow and heat is optimized. Using intelligent algorithms such as enhanced random forest (RF++), the system can automatically adjust the light source, electrode parameters and fluid flow conditions to ensure the best reaction environment. This not only increases the chemical reaction rate, but also enhances the selectivity of the target product.

[0030] 3) The built-in intelligent control system is developed based on Python language and integrates machine learning models such as RF++. It can analyze sensor data in real time and dynamically adjust various physical field parameters to ensure that the reaction process is always in the optimal state. IV. Description of the drawings

[0031] In order to more clearly illustrate the specific embodiments of the present invention, the drawings used in the description of the specific embodiments are briefly described below. Figure 1This is a schematic diagram of the structure of a closed cylindrical photoelectric composite reaction device, and the accompanying reference numerals are explained as follows:

[0032] A housing (1), an LED array and a reflector, a lens system (2), a feed port (3), a discharge port (4), a porous distributor (5), an exhaust port (6), a metal wire mesh cathode plate (7), a porous mesh anode plate (8), a photocatalytic parallel multi-tube plate (9), an electrocatalytic system controller (10), and a photocatalytic system controller (11).

[0033] Figure 2 The structure of the closed cylindrical photoelectric composite reaction device is schematically shown in cross-section. The reference numerals in the figure are the same as those in the appendix. Figure 1 .

[0034] Figure 3 These are the experimental results of testing the chlorine production performance of electrolyzing salt water in a closed cylindrical photoelectric composite reaction device. V. Specific implementation methods

[0035] The present invention is described in detail below with reference to the accompanying drawings and embodiments:

[0036] Embodiment 1:

[0037] Construction of a closed cylindrical photoelectric composite reactor: The reactor has an aspect ratio of 3:1, a diameter of 20 cm and a height of 60 cm. The reaction chamber is equipped with an adjustable light source, which supports a wavelength range of 300 to 800 nm and an intensity adjustment range of 0.1 to 5 W / cm 2, and is equipped with an evenly distributed LED array to achieve optimized light distribution; the LED array consists of multiple ultraviolet LEDs, which ensure that the light is evenly irradiated to the entire reaction area through a reflector and lens system; the light source power supply system adopts constant current source control, and the current adjustment accuracy is ±0.1% to ensure stable light intensity output; the light source control method includes manual adjustment and automatic feedback control system, the latter is based on the built-in light sensor to monitor the light intensity changes in real time and dynamically adjust to the optimal value; the reactor is equipped with a feed port and a discharge port. The feed port is located on one side of the bottom of the reactor and is made of stainless steel with a quick connector to facilitate connection to an external feeding system. The inlet diameter is 10mm and the flow control range is 0.1-5L / min; the discharge port is located on the other side of the top and is also controlled by a stainless steel valve to ensure the product Smooth discharge; the dispersion system is composed of a porous distributor to prevent channeling, short circuit, open circuit and other problems caused by direct feeding; the reaction temperature is maintained by jacket cooling water circulation, the cooling water flow rate is 0.5-2L / min, the temperature control accuracy is ±0.5℃, and the operating temperature range is from room temperature to 100℃; the exhaust port is set at the top of the reactor, with a diameter of 8mm and a gas filter to prevent harmful gas from escaping; the electrical system includes power supply, cables and connectors; the electrocatalytic system consists of cathode, anode and supporting cables and power supply, the cathode is a metal wire mesh structure, made of copper-nickel alloy, with a pore size of 1mm and a thickness of 1mm; the anode is a porous mesh structure, made of a porous titanium substrate loaded with Ru-doped PbO2 active material, which is loaded by pulse electrodeposition, with a loading amount of 25mg / cm 2 The power supply is a DC regulated power supply with a voltage adjustment range of 0 to 10V and an accuracy of ±0.1%. The cable is made of high-temperature resistant and corrosion-resistant silicone rubber insulated wire, and the connection method is a waterproof plug and socket. The photocatalytic parallel multi-tube plate has a base made of highly transparent borosilicate glass in the shape of a hollow tube with a length of 50cm, an outer diameter of 10mm, and an inner diameter of 8mm. The bottom layer is loaded with a layer of Si by ion sputtering with a thickness of 1μm, which plays a protective role. The middle layer is a nanoparticle coating of a Ti5O7 / Ti3O4 mixture (mass ratio of 4:1), which is obtained by supersonic flame sintering and has a thickness of 1 to 2μm. It serves as the main photocatalytic active layer and has a certain thermal catalytic effect. The surface layer is Ti2N3, which is obtained by ion sputtering and has a thickness of 50nm. While enhancing the light absorption capacity and catalytic activity, it also protects the Ti5O7 / Ti3O4 mixed layer.

[0038] During the catalytic reaction, the liquid flow enters the reactor from the feed inlet at the bottom, is evenly dispersed by a porous distributor, then penetrates the cathode metal mesh plate, the photocatalytic parallel multi-tube plate, and the anode porous mesh plate layer by layer, flows out through the discharge port, and completes the reaction under the action of the light field and the electric field. During the reaction, the light field, the electric field, and the flow field can be automatically adjusted to achieve multi-field coupling. Distributed silicon photodiodes and quantum dot fluorescent probes are installed in the reaction chamber as photoelectric sensors to monitor the concentration changes of reactants and the generation of by-products in real time. The sensor density is at least 10 detection points per cubic centimeter to ensure full coverage of the reaction area. In addition, temperature, pH value, and oxygen content sensors are configured to assist in monitoring the reaction environment conditions and form a data set of operating condition parameters and the conversion rate, selectivity, and stability of the corresponding output products.

[0039] The collected data is processed and predicted using the enhanced random forest (RF++) algorithm to add a feature selection module, introduce the Boosting mechanism in ensemble learning, and gradually increase weak classifiers to improve the overall performance (the machine learning algorithm can also be the gradient boosting tree (GBT++), which can enhance the ability to capture nonlinear relationships and improve prediction accuracy by introducing feature interaction terms and regularization terms); the training data set is composed of historical experimental results, covering reaction rate information under different illumination, electric field and flow field conditions; an intelligent control system is constructed based on Python language control software, integrating the RF++ model, and dynamically adjusting the light field (wavelength, intensity, distribution), electric field (current, voltage, power line distribution) and flow field (flow velocity, flow rate, distribution) parameters according to sensor feedback (the photoelectric sensor can also include a Raman spectrometer and a Fourier transform infrared spectrometer (FTIR) to provide more detailed molecular structure information to assist in judging the reaction process); according to the prediction results of the RF++ model, the intelligent control system automatically adjusts the light source output parameters so that the wavelength varies in the range of 300 to 800 nm and the intensity varies in the range of 0.1 to 5 W / cm 2 The LED array layout is optimized to ensure uniform light distribution. The PID controller maintains the set value stable with an error of no more than ±0.05W / cm 2; Light field condition control is further refined to use a microlens array to adjust the light distribution to ensure that the surface of the photocatalyst receives uniform and high-intensity light, thereby maximizing the photocatalytic efficiency; the intelligent control system adjusts the voltage and current between the electrode plates according to the prediction model, changes the distribution of power lines, and optimizes the electrocatalytic effect; the control system has a built-in overcurrent protection circuit to prevent equipment damage and ensure safe operation; the peristaltic pump and Venturi effect nozzle are used to accurately control the reaction liquid flow rate (0.1~100mL / min) and flow rate, combined with real-time monitoring by an ultrasonic flowmeter to ensure that the flow field conditions meet the optimal solution (flow field condition control also involves the use of turbulence promoters and static mixers to improve the liquid flow characteristics, ensure that the reactants are in full contact with the catalyst, and further increase the reaction rate); the control system supports a closed-loop feedback mechanism to automatically correct deviations and maintain set accuracy; combined with the AI ​​model to predict the best multi-field coupling setting solution, through continuous iterative optimization, the chemical reaction rate is improved, energy consumption is reduced, and the purpose of energy saving and consumption reduction is achieved; the intelligent control system records the parameters after each optimization and forms a database for subsequent improvement reference.

[0040] The structure diagram of the closed cylindrical photoelectric composite reaction device is shown in the attached figure. Figure 1 and attached Figure 2 shown.

[0041] Embodiment 2:

[0042] The closed cylindrical photoelectric composite reaction device constructed in Example 1 was used to test the chlorine production performance of catalytic salt water decomposition:

[0043] The initial saturated salt water inlet flow rate was set to 0.5 L / min, the residence time was 120 min, the electrocatalytic current was 2 A, the voltage was 3 V, and the light field intensity was 1 W / cm 2 Under this condition, the system was initially tested and the unit yield (active chlorine) was measured to be 0.08 g / L·min, the Faradaic efficiency was about 65%, and the unit energy consumption was 0.72 kWh / kg Cl2.

[0044] Furthermore, the enhanced random forest (RF++) algorithm was introduced for data processing and prediction, in order to improve performance through continuous iterative optimization. In the first iteration, the algorithm recommended adjusting the light field intensity to 1.51W / cm 2 , while increasing the electrocatalytic current to 2.57 A and keeping the voltage unchanged. After this adjustment, the unit yield increased to 0.12 g / L·min, the Faradaic efficiency reached 75%, and the unit energy consumption dropped to 0.62 kWh / kg Cl2, showing some improvement.

[0045] With more iterations, the system gradually approaches the optimal configuration. In the 8th iteration, the RF++ model recommends further optimizing the light field intensity to 2.7W / cm2 , the electrocatalytic current was set to 3.6A, and the voltage was fine-tuned to 3.2V. At this time, the water inlet flow rate was also adjusted to 0.68L / min to ensure the best flow field conditions. Finally, after the implementation of this series of optimization measures, the experimental results showed that the unit yield reached 0.16g / L·min, the Faraday efficiency was stabilized at 85%, and the unit energy consumption was reduced to 0.45kWh / kg Cl2, which was significantly improved compared with the data before optimization.

[0046] Specific parameter conditions and experimental data are shown in the attached Figure 3 shown.

[0047] Embodiment 3:

[0048] The closed cylindrical photoelectric composite reaction device constructed in Example 1 was used to perform a test on the degradation of simulated Congo red industrial wastewater:

[0049] The experiment was first tested with unoptimized parameters: water flow rate was 0.5 L / min, residence time was 120 min, electrocatalytic current was 2 A, voltage was 3 V, and light field intensity was 1 W / cm 2 Under these conditions, the system measured a Congo red removal rate of 62.5%, a COD removal rate of 34.7%, an ammonia nitrogen removal rate of 16.6%, and a unit energy consumption of 0.85 kWh / L.

[0050] The enhanced random forest (RF++) algorithm is introduced for data processing and prediction, in order to improve performance through iterative optimization. In the first iteration, the algorithm recommends adjusting the light field intensity to 1.75W / cm 2 , while increasing the electrocatalytic current to 2.25A and keeping the voltage unchanged. After this adjustment, the chroma removal rate increased to 78.5%, the COD removal rate reached 66.3%, the ammonia nitrogen removal rate increased to 42.5%, and the unit energy consumption dropped to 0.56kWh / L.

[0051] With more iterations, the system gradually approaches the optimal configuration. In the 12th iteration, the RF++ model recommends further optimizing the light field intensity to 2.25W / cm 2 , the electrocatalytic current was set to 3.17A, and the voltage was fine-tuned to 3.20V. At this time, the water flow rate was also adjusted to 0.685L / min to ensure the optimal flow field conditions. Finally, after the implementation of this series of optimization measures, the experimental results showed that the chroma removal rate reached 95.2%, the COD removal rate was stabilized at 88.7%, the ammonia nitrogen removal rate increased to 82.5%, and the unit energy consumption was reduced to 0.36kWh / L.

[0052] In addition, in order to verify whether the optimized parameters can maintain consistency in wastewater with different concentrations of Congo red, the above experiment was repeated at a series of different initial concentrations (50 mg / L, 100 mg / L, 200 mg / L). The results showed that in all test concentration ranges, the optimized parameters can maintain a high removal rate and low unit energy consumption, proving the stability and wide applicability of the method.

[0053] Embodiment 4:

[0054] The closed cylindrical photoelectric composite reaction device constructed in Example 1 was used to perform a test on the degradation of simulated nitrobenzene wastewater:

[0055] Initial experimental settings: water flow rate: 0.45 L / min, residence time: 130 min, electrocatalytic current: 2.2 A, voltage: 3.1 V, light field intensity: 1.2 W / cm 2 , a preliminary test was conducted on nitrobenzene wastewater under the above parameters. The results showed that the removal rate of nitrobenzene was 58.76%, the COD removal rate was 47.32%, the ammonia nitrogen removal rate was 36.45%, the chroma removal rate was 59.21%, and the toxicity (LC 50 ) increased from the initial 10.5mg / L to 25.6mg / L, with a unit energy consumption of 0.72kWh / L.

[0056] The enhanced random forest (RF++) algorithm is introduced for data processing and prediction. The first iteration recommends adjusting the light field intensity to 1.8W / cm 2 , while increasing the electrocatalytic current to 2.8A and keeping the voltage unchanged. After this adjustment, the nitrobenzene removal rate increased to 82.47%, COD removal rate reached 68.95%, ammonia nitrogen removal rate increased to 51.67%, chroma removal rate increased to 81.53%, toxicity (LC 50 ) was further increased from 25.6mg / L to 37.8mg / L, and the unit energy consumption was reduced to 0.63kWh / L.

[0057] With more iterations, the system gradually approaches the optimal configuration. In the 9th iteration, the RF++ model recommends further optimizing the light field intensity to 2.2W / cm 2 , the electrocatalytic current was set to 3.2A, and the voltage was fine-tuned to 3.4V. At this time, the water flow rate was also adjusted to 0.55L / min to ensure the best flow field conditions. Finally, after the implementation of this series of optimization measures, the experimental results showed that the nitrobenzene removal rate reached 94.56%, the COD removal rate was stable at 87.63%, the ammonia nitrogen removal rate increased to 72.34%, the chromaticity removal rate reached 92.89%, and the toxicity (LC 50) increased from 37.8mg / L to 50.2mg / L, and the unit energy consumption decreased to 0.48kWh / L.

[0058] Embodiment 5:

[0059] The closed cylindrical photoelectric composite reaction device constructed in Example 1 was used to perform a test on the degradation of simulated aniline industrial wastewater:

[0060] The initial experimental conditions were set as follows: water flow rate: 0.4 L / min, residence time: 140 min, electrocatalytic current: 2.3 A, voltage: 3.2 V, light field intensity: 1.3 W / cm 2 Under the above conditions, the system was preliminarily tested on aniline industrial wastewater. The results showed that the aniline removal rate was 56.89%, the COD removal rate was 48.74%, the ammonia nitrogen removal rate was 37.65%, the chroma removal rate was 60.12%, and the toxicity (LC 50 ) increased from the initial 8.5mg / L to 22.3mg / L, with a unit energy consumption of 0.75kWh / L.

[0061] The enhanced random forest (RF++) algorithm is introduced for data processing and prediction. The first iteration recommends adjusting the light field intensity to 1.9W / cm 2 , while increasing the electrocatalytic current to 2.9A and keeping the voltage unchanged. After this adjustment, the aniline removal rate increased to 83.45%, COD removal rate reached 70.23%, ammonia nitrogen removal rate increased to 52.17%, chroma removal rate increased to 82.47%, toxicity (LC 50 ) was further increased from 22.3mg / L to 36.5mg / L, and the unit energy consumption was reduced to 0.65kWh / L.

[0062] With more iterations, the system gradually approaches the optimal configuration. Specifically, in the 7th iteration, the RF++ model recommends further optimizing the light field intensity to 2.3W / cm 2 , the electrocatalytic current was set to 3.3A, and the voltage was fine-tuned to 3.5V. At this time, the water flow rate was also adjusted to 0.5L / min. Finally, after the implementation of this series of optimization measures, the experimental results showed that the aniline removal rate reached 95.12%, the COD removal rate was stabilized at 88.45%, the ammonia nitrogen removal rate increased to 73.26%, the chromaticity removal rate reached 93.67%, the toxicity (LC50) increased from 36.5mg / L to 51.8mg / L, and the unit energy consumption was reduced to 0.50kWh / L.

[0063] The specific implementation methods described above are only used to specifically illustrate the spirit of the present invention, and the protection scope of the present invention is not limited thereto. For those skilled in the art, other implementation methods can certainly be easily made by changing, replacing or modifying the technical contents disclosed in this specification, and these other implementation methods should all be covered within the protection scope of the present invention.

Claims

1. A method for enhancing the efficiency of photoelectric composite catalytic reaction by adaptive multi-field coupling, characterized in that The method specifically comprises the following steps: Step 1: construct a closed cylindrical photoelectric composite reaction device with an aspect ratio of 3:1, a diameter of 20 cm and a height of 60 cm; the reaction chamber is equipped with an adjustable light source, supporting a wavelength range of 300 to 800 nm and an intensity adjustment range of 0.1 to 5 W / cm 2 , and is equipped with an evenly distributed LED array to achieve optimized light distribution; the LED array consists of multiple ultraviolet LEDs, which ensure that the light is evenly irradiated to the entire reaction area through a reflector and lens system; the light source power supply system adopts constant current source control, and the current adjustment accuracy is ±0.1% to ensure stable light intensity output; the light source control method includes manual adjustment and automatic feedback control system, the latter is based on the built-in light sensor to monitor the light intensity changes in real time and dynamically adjust to the optimal value; the reactor is equipped with a feed port and a discharge port. The feed port is located on one side of the bottom of the reactor and is made of stainless steel with a quick connector to facilitate connection to an external feeding system. The inlet diameter is 10mm and the flow control range is 0.1-5L / min; the discharge port is located on the other side of the top and is also controlled by a stainless steel valve to ensure the product Smooth discharge; the dispersion system is composed of a porous distributor to prevent channeling, short circuit, open circuit and other problems caused by direct feeding; the reaction temperature is maintained by jacket cooling water circulation, the cooling water flow rate is 0.5-2L / min, the temperature control accuracy is ±0.5℃, and the operating temperature range is from room temperature to 100℃; the exhaust port is set at the top of the reactor, with a diameter of 8mm and a gas filter to prevent harmful gas from escaping; the electrical system includes power supply, cables and connectors; the electrocatalytic system consists of cathode, anode and supporting cables and power supply, the cathode is a metal wire mesh structure, made of copper-nickel alloy, with a pore size of 1mm and a thickness of 1mm; the anode is a porous mesh structure, made of a porous titanium substrate loaded with Ru-doped PbO2 active material, which is loaded by pulse electrodeposition, with a loading amount of 25mg / cm 2 The power supply is a DC regulated power supply with a voltage adjustment range of 0 to 10V and an accuracy of ±0.1%. The cable is made of high temperature resistant and corrosion resistant silicone rubber insulated wire, and the connection method is a waterproof plug and socket. The photocatalytic parallel multi-tube plate has a base made of highly transparent borosilicate glass in the form of a hollow tube with a length of 50cm, an outer diameter of 10mm, and an inner diameter of 8mm. The bottom layer is loaded with a layer of Si by ion sputtering with a thickness of 1μm, which plays a protective role. The middle layer is a Ti5O7 / Ti3O4 mixed (mass ratio of 4:1) nanoparticle coating, which is obtained by supersonic flame sintering and has a thickness of 1 to 2μm. It serves as the main photocatalytic active layer and has a certain thermal catalytic effect. The surface layer is Ti2N3, which is obtained by ion sputtering and has a thickness of 50nm. While enhancing the light absorption capacity and catalytic activity, it also protects the Ti5O7 / Ti3O4 mixed layer. Step 2, during the catalytic reaction, the liquid flow enters the reactor from the bottom of the feed port, is evenly dispersed by a porous distributor, and then penetrates the cathode metal mesh plate, the photocatalytic parallel multi-tube plate, and the anode porous mesh plate layer by layer, flows out through the discharge port, and completes the reaction under the action of the light field and the electric field; during the reaction, the light field, the electric field, and the flow field can be automatically adjusted to achieve multi-field coupling; a distributed silicon photodiode and a quantum dot fluorescent probe are installed in the reaction chamber as photoelectric sensors for real-time monitoring of reactant concentration changes and by-product generation; the sensor density is at least 10 detection points per cubic centimeter to ensure full coverage of the reaction area; in addition, temperature, pH value, and oxygen content sensors are configured to assist in monitoring the reaction environment conditions, and form a data set of operating condition parameters and output corresponding product conversion rate, selectivity, and stability; Step 3: Apply the enhanced random forest (RF++) algorithm for data processing and prediction to add a feature selection module, introduce the Boosting mechanism in ensemble learning, and improve the overall performance by gradually adding weak classifiers; the training data set is composed of historical experimental results, covering the reaction rate information under different lighting, electric field and flow field conditions; Step 4: Build an intelligent control system based on Python language control software, integrate the RF++ model, and dynamically adjust the light field (wavelength, intensity, distribution), electric field (current, voltage, power line distribution), and flow field (flow velocity, flow rate, distribution) parameters according to sensor feedback; Step 5: Based on the prediction results of the RF++ model, the intelligent control system automatically adjusts the light source output parameters so that the wavelength varies within the range of 300 to 800 nm and the intensity is within the range of 0.1 to 5 W / cm 2 The LED array layout is optimized to ensure uniform light distribution. The PID controller maintains the set value stable with an error of no more than ±0.05W / cm 2 ; Step 6: The intelligent control system adjusts the voltage and current between the electrode plates according to the prediction model, changes the distribution of power lines, and optimizes the electrocatalytic effect; the control system has a built-in overcurrent protection circuit to prevent equipment damage and ensure safe operation; Step 7, using a peristaltic pump and a Venturi effect nozzle to accurately control the flow rate (0.1-100 mL / min) and flow rate of the reaction liquid, combined with real-time monitoring by an ultrasonic flow meter to ensure that the flow field conditions meet the optimal solution; the control system supports a closed-loop feedback mechanism to automatically correct deviations and maintain set accuracy; Step 8: Combine the AI ​​model to predict the optimal multi-field coupling setting plan, and through continuous iterative optimization, improve the chemical reaction rate, reduce energy consumption, and achieve the goal of energy saving and consumption reduction; the intelligent control system records the parameters after each optimization and forms a database for reference for subsequent improvements.

2. A method for enhancing the efficiency of photoelectric composite catalytic reaction by adaptive multi-field coupling according to claim 1, characterized in that: The photoelectric sensor described in step 2 also includes a Raman spectrometer and a Fourier transform infrared spectrometer (FTIR) to provide more detailed molecular structure information to assist in determining the reaction progress.

3. According to the method of enhancing the efficiency of photoelectric composite catalytic reaction by adaptive multi-field coupling as described in claim 1, it is characterized in that: The machine learning algorithm described in step 3 can also be a gradient boosted tree (GBT++), which can enhance the ability to capture nonlinear relationships and improve prediction accuracy by introducing feature interaction terms and regularization terms.

4. According to the method of claim 1, wherein the method comprises: The light field condition control described in step 5 is further refined to use a microlens array to adjust the light distribution to ensure that the photocatalyst surface receives uniform and high-intensity light, thereby maximizing the photocatalytic efficiency.

5. The method of enhancing the efficiency of photoelectric composite catalytic reaction by adaptive multi-field coupling according to claim 1, characterized in that: The flow field condition control described in step 7 also involves the use of turbulence promoters and static mixers to improve the liquid flow characteristics, ensure sufficient contact between the reactants and the catalyst, and further increase the reaction rate.

Images