Intelligent dispersion system and method for preparing high-performance nano-photocatalytic water-based coatings

Abstract

This invention provides an intelligent dispersion system for preparing high-performance nano-photocatalytic water-based coatings and its application method. The intelligent dispersion system comprises an ultrasonic disperser, a microfluidizer, an online laser particle size analyzer, a conductivity sensor, a pH sensor, an electric valve, an intelligent predictive control and process monitoring system, and a low-temperature control device. By utilizing ultrasonic-microfluidic synergistic dispersion, online monitoring, and AI-powered dynamic control technology, the system addresses the performance degradation of photocatalysts during coating production, achieving efficient, low-cost, and stable nano-dispersion. The system is adaptable to multiple catalysts and can be widely used in industrial-grade coating production.

Description

Technical Field

[0001] The present invention relates to the technical field of preparation of photocatalytic water-based coatings and improvement of catalytic performance, and in particular to an intelligent photocatalytic nano-dispersion system that combines ultrasonic and microfluidic collaborative dispersion, real-time monitoring and intelligent regulation. Background Art

[0002] Common photocatalysts such as TiO2, BiVO4, and ZnO are widely used in coatings, environmental protection, and energy fields due to their excellent photocatalytic properties. By combining photocatalysts with coatings, coatings can be made to have the function of degrading NOx. Photocatalysts generate electron-hole pairs through light excitation, which further generate active oxygen species, which can efficiently oxidize NOx to nitrate (NO3 - ) and other harmless products. This type of photocatalytic coating can be applied to surfaces such as building exteriors and road guardrails, continuously purifying the environment under sunlight and effectively alleviating urban air pollution. With tightening global environmental regulations and the advancement of the "dual carbon" goals, the development of highly efficient photocatalytic coatings has become a key technology for achieving green buildings and improving air quality, and is of urgent importance for promoting sustainable urban development.

[0003] During the coating production process, photocatalysts are prone to agglomeration due to high surface energy, electrostatic adsorption, and mechanical shear forces, resulting in insufficient exposure of active sites and a significant decrease in photocatalytic efficiency. Agglomerated catalysts also lead to poor coating stability and a significantly shortened shelf life, further increasing costs and wasting catalyst raw materials during coating production due to the need to add additional catalysts to compensate for efficiency losses. Existing dispersion technologies such as mechanical stirring and single ultrasonic treatment have problems such as high energy consumption, uneven dispersion, and lack of real-time regulation, making it difficult to meet the needs of industrial-scale production. Therefore, it is necessary to develop a nano-dispersion system for water-based coatings with catalytic functions in order to better solve the problem of performance degradation caused by agglomeration of photocatalysts after they are added to the coating, improve dispersion efficiency and coating performance, save production raw materials, and reduce production costs. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent photocatalytic nano-dispersion system that combines ultrasonic and microfluidic collaborative dispersion, real-time monitoring and intelligent regulation to solve the problem of decreased catalytic performance of photocatalysts in coating production, improve photocatalytic efficiency and coating stability, realize online monitoring and dynamic intelligent regulation of the catalytic performance of catalysts in coatings, and maintain the stability and controllability of catalyst performance in coating production through ultrasonic-microfluidic collaborative dispersion.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] An intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coatings, comprising a material processing unit, an online monitoring unit, an execution control unit, a computer and a piping system; the computer is connected to the material processing unit, the online monitoring unit and the execution unit in a communication manner;

[0007] The material processing unit includes an ultrasonic disperser and a micro jet homogenizer;

[0008] The online monitoring unit includes an online laser particle size analyzer, a conductivity sensor and a pH sensor;

[0009] The execution control unit includes two three-way valves with the same structure: a No. 1 electric valve and a No. 2 electric valve. The No. 1 electric valve is used to switch between the internal circulation and discharge modes; the No. 2 electric valve is used to switch between the internal circulation and introduction modes.

[0010] The computer includes an intelligent predictive control and process monitoring system for real-time data collection, processing, and adjustment;

[0011] The pipeline system includes a No. 1 delivery pipe, a No. 2 delivery pipe, a No. 3 delivery pipe and a No. 4 delivery pipe;

[0012] The No. 2 electric valve is connected to the ultrasonic disperser through the No. 1 delivery pipe, the ultrasonic disperser is connected to the microjet homogenizer through the No. 2 delivery pipe, the microjet homogenizer is connected in series with the conductivity sensor, pH sensor and online laser particle size analyzer through the No. 3 delivery pipe to the No. 1 electric valve, and the No. 1 electric valve is connected to the No. 2 electric valve through the No. 4 delivery pipe.

[0013] Preferably, the intelligent predictive control and process monitoring system includes a real-time data acquisition module, a parameter identification module, a communication module and a safety protection mechanism, and connects the material processing unit, the online monitoring unit and the execution unit through an industrial communication protocol to achieve closed-loop control;

[0014] The specific implementation of the intelligent predictive control and process monitoring system is as follows:

[0015] The real-time data acquisition module uses an online laser particle size analyzer, a conductivity sensor, and a pH sensor to synchronously collect slurry particle size distribution (including D10 / D50 / D90 values), conductivity, and pH value data at a frequency of ≥1 Hz. The data is transmitted to a computer in real time via the Modbus-RTU / TCP industrial bus protocol with a transmission delay of ≤20ms. The collected raw data is pre-processed by sliding window mean filtering and z-score normalization to eliminate noise and dimensional differences.

[0016] The AI ​​prediction model is a time series prediction model built based on a long short-term memory (LSTM) network. The input layer of the model receives the time series process parameters in the standardized data set, including particle size distribution, conductivity, pH value, and the current operating parameters of the ultrasonic disperser and the microjet homogenizer. The hidden layer adopts a two-layer LSTM structure to capture the dynamic correlation between process parameters. The output layer generates optimal control instructions for the ultrasonic frequency adjustment amount, the microjet pressure adjustment amount, and the electric valve opening.

[0017] The AI ​​prediction model is trained using a supervised learning method. The training data set is derived from more than 10,000 sets of valid data records accumulated during historical production processes. The records need to cover the dispersion process parameters and corresponding performance indicators of a variety of typical nano-photocatalysts such as TiO2, BiVO4, ZnO and their composite materials. The training process uses mean square error as the loss function and adopts the Adam optimizer for parameter optimization. After the AI ​​prediction model is trained, it must pass a 5-fold cross-validation to ensure that its prediction accuracy on an independent test set is not less than 93% before it can be deployed and applied.

[0018] The anomaly detection module of the AI ​​prediction model establishes a parameter fluctuation warning interval by setting the 3σ principle. When three consecutive sampling values ​​exceed the interval, the reflow mechanism is triggered. The web interactive interface allows engineers to manually set parameter thresholds, view real-time production curves, and export historical process reports. It has a multi-level permission management function, realizing hierarchical control by administrators and operators.

[0019] The safety protection mechanism includes a multi-level interlocking protection module, a hardware redundancy module, an emergency power supply module and a physical isolation module;

[0020] The multi-level interlocking protection module is started as follows:

[0021] When the online laser particle size analyzer (14) detects that D90>300nm exceeds the limit continuously, the first level protection is triggered and the micro jet pressure is automatically reduced to the reference value of 500bar;

[0022] When the conductivity sensor (15) detects a fluctuation of >15% and the pH exceeds the range of 6.5-9.0, the secondary protection is triggered, the feed valve is immediately closed and the internal circulation reflux is started;

[0023] When the temperature sensor detects that the material temperature is ≥32°C, the third level protection is triggered, the low temperature control device is forced to start cooling and the ultrasonic output is suspended;

[0024] The hardware redundancy module includes a conductivity sensor (15), a pH sensor (12), and an online laser particle size analyzer (14), all of which adopt a dual-probe redundancy design and automatically switch to the backup signal when the difference between the master and slave probe data is greater than 5%;

[0025] The electric valve is equipped with dual control signals, which will automatically switch to the 4-20mA analog backup channel when the main control signal fails;

[0026] The emergency power supply module includes a UPS uninterruptible power supply, which maintains the system's minimum power consumption for ≥30 minutes when the main power is interrupted, giving priority to sensor data storage and valve safety position switching;

[0027] The physical isolation module includes isolating the strong electric control circuit and the weak electric signal circuit through a photoelectric coupler to prevent malfunction caused by electromagnetic interference; the control instruction transmission adopts CRC-16 check, and automatically retransmits when the error rate is greater than 1‰.

[0028] Preferably, the intelligent predictive control includes an AI prediction model, an anomaly detection module and a Web interactive interface;

[0029] The conductivity sensor, pH sensor and online laser particle size analyzer collect the particle size distribution, pH value and conductivity parameters of the water-based paint in real time. After data cleaning, the data is input into a pre-trained AI prediction model. The AI ​​prediction model generates and executes dynamic adjustment of ultrasonic frequency, micro-jet pressure and valve opening based on historical data, and controls the discharge opening when the parameters are qualified and automatically generates a process report;

[0030] The process report is a PDF format document that automatically records the following data: dispersion process time axis curve of dispersion quality parameters, process timing parameters, and is available for download through a web interface;

[0031] The dispersion quality parameters include: particle size distribution measured in real time by an online laser particle size analyzer; conductivity measured by a conductivity sensor; and pH value measured by a pH sensor.

[0032] Process timing parameters include: time axis curve generated by computer real-time database, valve action log derived from the 4-20mA valve position feedback signal of the electric valve;

[0033] Equipment operating parameters: The number of cycles was obtained from the OPC-UA interface of the microfluidizer;

[0034] The process monitoring system includes a preset basic anomaly detection module of the AI ​​prediction model, which triggers an alarm and switches to a preset safe mode when the collected parameters deviate from the threshold;

[0035] The system has a pre-set safety mode: when any dispersion quality parameter exceeds the threshold value three times in a row, the system automatically closes the feed valve, switches to internal circulation reflux, triggers an audible and visual alarm, and resets the equipment parameters to the baseline value.

[0036] The threshold range refers to the following: D90>300nm detected by the online laser particle size analyzer, conductivity fluctuation>15% detected by the conductivity sensor, and pH<6.5 or>9.0 detected by the pH sensor. The above parameter ranges can be adjusted according to actual conditions;

[0037] The reference value refers to the ultrasonic disperser outputting 20 kHz of ultrasonic waves and the microjet homogenizer setting the microjet at 500 bar;

[0038] The web interactive interface is used to support manual intervention.

[0039] Preferably, the ultrasonic disperser includes an ultrasonic generator, an ultrasonic probe and a sealed liquid storage tank; the ultrasonic generator is connected to the sealed liquid storage tank via the ultrasonic probe; the ultrasonic generator is connected to a computer, and the computer's intelligent predictive control and process monitoring system dynamically adjusts the ultrasonic frequency of the ultrasonic generator through an algorithm based on the particle size distribution monitored by the online laser particle size analyzer.

[0040] Preferably, the ultrasonic disperser and microfluidizer are equipped with a low-temperature control device to control the material temperature during the treatment process below 30° C. to suppress the attenuation of catalyst activity caused by heat generation by ultrasound and microjet.

[0041] Preferably, the inner walls of the pipeline system are coated with a super-hydrophobic coating to reduce catalyst adhesion loss.

[0042] Preferably, the two three-way valves each include a No. 1 interface, a No. 2 interface, and a No. 3 interface;

[0043] The No. 1 interface of the No. 2 electric valve is connected to the No. 2 interface of the No. 1 electric valve through the No. 4 delivery pipe, and the No. 1 interface of the No. 1 electric valve is connected to the No. 3 delivery pipe;

[0044] The No. 2 interface of the No. 2 electric valve is connected to the feed port of the sealed liquid storage tank of the ultrasonic disperser through the No. 1 delivery pipe;

[0045] The No. 3 interface of the No. 2 electric valve is connected to the mixed liquid of the catalyst and the water-based paint through the No. 6 delivery pipe for feeding, and the No. 3 interface of the No. 1 electric valve is connected to the No. 5 delivery pipe for discharging.

[0046] The present invention also provides an intelligent dispersion method for preparing high-performance nano-photocatalytic water-based coatings, comprising the following steps:

[0047] Step S1: Premixing: adding nano-photocatalyst to water-based coating and mixing by mechanical stirring to form a preliminary catalyst coating mixture;

[0048] Step S2: Dispersion: The catalyst-coating mixture enters the No. 2 port of the No. 2 electric valve through the No. 6 delivery pipe, and is introduced into the sealed liquid storage tank of the ultrasonic disperser through the No. 1 delivery pipe. The ultrasonic probe operates at a preset frequency to preliminarily break up the agglomerates to obtain the dispersed coating;

[0049] Step S3: Real-time monitoring: The particles are fed into the microfluidizer through the No. 2 delivery pipe to be refined, and the particle size distribution of the dispersed coating is monitored in real time in the online laser particle size analyzer, and the pH value and conductivity data of the conductivity sensor and pH sensor are monitored in real time; the data are collected synchronously and transmitted to the computer;

[0050] Step S4: Intelligent evaluation and adjustment: After data cleaning, it is input into a pre-trained AI prediction model on a computer. The AI ​​prediction model generates and executes dynamic adjustments to the ultrasonic frequency, micro-jet pressure, and valve opening based on historical data to optimize the dispersion effect;

[0051] Step S5: When the material parameters are qualified, the computer controls the No. 1 electric valve to open the No. 3 interface to start discharging, and automatically generates a process report;

[0052] The No. 1 interface of the No. 2 electric valve is controlled to close, and the No. 3 and No. 2 interfaces of the No. 2 electric valve are opened to receive the catalyst coating mixture through the No. 6 delivery pipe, and then introduced into the sealed liquid storage tank of the ultrasonic disperser through the No. 2 interface of the No. 2 electric valve through the No. 1 delivery pipe for subsequent processing;

[0053] When the material parameters are unqualified, the computer controls the opening of the No. 2 interface of the No. 1 electric valve to enter the No. 1 interface of the No. 2 electric valve through the No. 4 delivery pipe, closes the No. 6 delivery pipe of the No. 2 electric valve, stops receiving new catalyst coating mixture, opens the No. 2 interface of the No. 2 electric valve, receives the unqualified material through the No. 1 delivery pipe and introduces it into the sealed liquid storage tank of the ultrasonic disperser, and returns it to the system for reprocessing.

[0054] Preferably, step S2 adopts multi-stage dispersion: the preliminary catalyst coating mixture is subjected to ultrasonic dispersion and micro-jet homogenization treatment in sequence.

[0055] Preferably, in step S3, the real-time monitoring includes monitoring the size of the catalyst and coating particles in the coating and monitoring the conductivity of the coating to determine the agglomeration trend.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. The present invention significantly improves the catalyst dispersion efficiency through the synergistic effect of ultrasound and microfluidics, combined with AI dynamic regulation, avoids the agglomeration and performance degradation problems caused by traditional methods, and achieves efficient and stable dispersion.

[0058] 2. Low temperature control to protect catalyst activity: The system of the present invention adopts low temperature treatment below 30°C to effectively inhibit the influence of thermal effect on catalyst activity and extend the service life.

[0059] 3. The AI ​​prediction model of the present invention adjusts process parameters in real time to adapt to different catalyst types, improves production flexibility and stability, and realizes intelligent adaptive optimization.

[0060] 4. The online monitoring + automatic reflux mechanism of the present invention ensures that unqualified materials are reprocessed, improves the yield rate, reduces waste, and realizes closed-loop quality control.

[0061] 5. The system of the present invention supports Modbus communication, high-precision valve control and Web interaction, which is easy to integrate into existing production lines and is suitable for large-scale coating production.

[0062] In summary, this invention addresses the performance degradation of photocatalysts in coating production through ultrasonic-microfluidic synergistic dispersion, online monitoring, and AI-powered dynamic control technology, achieving efficient, low-cost, and stable nanodispersion. The system is adaptable to multiple catalysts and can be widely used in industrial-grade coating production. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic structural diagram of an intelligent dispersion system for preparing high-performance nano-photocatalytic water-based coatings provided by an embodiment of the present invention;

[0064] Figure 2 A schematic diagram of the structure of an electric valve in an intelligent dispersion system for preparing high-performance nano-photocatalytic water-based coatings provided by an embodiment of the present invention;

[0065] Figure 3 A schematic flow chart of an application method of an intelligent dispersion system for preparing high-performance nano-photocatalytic water-based coatings provided in an embodiment of the present invention.

[0066] The serial numbers in the figure are as follows:

[0067] Delivery pipe No. 1; 2. Ultrasonic disperser; 3. Delivery pipe No. 2; 4. Microfluidizer; 5. Delivery pipe No. 3; 6. Electric valve No. 1; 7. Delivery pipe No. 4; 8. Computer; 9. Delivery pipe No. 5; 10. Inlet tube for online laser particle size analyzer; 11. Delivery pipe No. 6; 12. pH sensor; 13. Electric valve No. 2; 14. Online laser particle size analyzer; 15. Conductivity sensor; 16. Sealed liquid storage tank; 17. Ultrasonic generator; 18. Ultrasonic probe; 19. Interface No. 1; 20. Interface No. 2; 21. Interface No. 3. DETAILED DESCRIPTION

[0068] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0069] like Figure 1 and Figure 2 As shown, an intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coatings provided in this embodiment includes a No. 1 delivery pipe 1, an ultrasonic disperser 2, a No. 2 delivery pipe 3, a microjet homogenizer 4, a No. 3 delivery pipe 5, a No. 1 electric valve 6, a No. 4 delivery pipe 7, a computer 8, a No. 5 delivery pipe 9, an online laser particle size analyzer sampling tube 10, a No. 6 delivery pipe 11, a pH sensor 12, a No. 2 electric valve 13, an online laser particle size analyzer 14, a conductivity sensor 15, a sealed liquid storage tank 16, an ultrasonic generator 17, and an ultrasonic probe 18.

[0070] Among them, the No. 1 delivery pipe 1, the ultrasonic disperser 2, the No. 2 delivery pipe 3, the microjet homogenizer 4, the No. 3 delivery pipe 5, the conductivity sensor 15, the pH sensor 12, the No. 1 electric valve 6, the No. 4 delivery pipe 7, and the No. 2 electric valve 13 are connected in sequence.

[0071] Computer 8 is connected to ultrasonic disperser 2, microfluidizer 4, electric valve No. 1 6, online laser particle size analyzer 14, conductivity sensor 15, pH sensor 12, and electric valve No. 2 13. Electric valve No. 1 6 is connected to delivery pipe No. 3 5, delivery pipe No. 4 7, and delivery pipe No. 5 9. Electric valve No. 2 13 is connected to delivery pipe No. 1 1, delivery pipe No. 4 7, and delivery pipe No. 6 11.

[0072] The microfluidizer 4 is connected to the computer 8 via a shielded cable with a digital signal transmitter. Its parameters such as pressure, flow, number of cycles, temperature, etc. are all regulated by the computer output signal and are monitored in real time by the above-mentioned intelligent predictive control and process supervision system.

[0073] The third delivery pipe 5 is connected to the feed port of the online laser particle size analyzer 14 and is used to collect a small amount of processed coating samples for the online laser particle size analyzer 14 to detect the catalyst particle size distribution in the slurry in real time.

[0074] The conductivity sensor 15 receives computer instructions via digital signals or bus protocols, is connected to the intelligent prediction control and process monitoring system, and transmits dynamic particle conductivity-related data for early warning of agglomeration trends.

[0075] The intelligent predictive control and process supervision system includes a real-time data acquisition module, a parameter identification module, a communication module and a safety protection mechanism. It connects the material handling unit, the online monitoring unit and the execution unit through the industrial communication protocol to achieve closed-loop control.

[0076] The specific implementation of the intelligent predictive control and process monitoring system is as follows:

[0077] The real-time data acquisition module uses an online laser particle size analyzer 14, a conductivity sensor 15, and a pH sensor 12 to synchronously collect slurry particle size distribution (including D10 / D50 / D90 values), conductivity, and pH data at a frequency of ≥1 Hz. This data is transmitted to a computer in real time via the Modbus-RTU / TCP industrial bus protocol, with a transmission delay of ≤20ms. The collected raw data undergoes sliding window mean filtering and z-score normalization to eliminate noise and dimensionality.

[0078] Furthermore, in this embodiment, the safety protection mechanism includes a multi-level interlocking protection module, a hardware redundancy module, an emergency power supply module, and a physical isolation module;

[0079] The multi-level interlocking protection module is started as follows:

[0080] When the online laser particle size analyzer 14 detects that D90>300nm exceeds the limit continuously, the first level protection is triggered and the micro jet pressure is automatically reduced to the reference value of 500bar;

[0081] When the conductivity sensor 15 detects a fluctuation of >15% and the pH exceeds the range of 6.5-9.0, the secondary protection is triggered, the feed valve is immediately closed and the internal circulation reflux is started;

[0082] When the temperature sensor detects that the material temperature is ≥32°C, the third level protection is triggered, the low temperature control device is forced to start cooling and the ultrasonic output is suspended;

[0083] The hardware redundancy module includes the conductivity sensor 15, the pH sensor 12, and the online laser particle size analyzer 14, all of which adopt a dual-probe redundancy design. When the difference between the master and slave probe data is greater than 5%, it automatically switches to the backup signal.

[0084] The electric valve is equipped with dual control signals, which will automatically switch to the 4-20mA analog backup channel when the main control signal fails;

[0085] The emergency power supply module includes a UPS uninterruptible power supply, which maintains the system's minimum power consumption for ≥30 minutes when the main power is interrupted, giving priority to sensor data storage and valve safety position switching;

[0086] The physical isolation module includes isolating the strong electric control circuit and the weak electric signal circuit through a photoelectric coupler to prevent malfunction caused by electromagnetic interference; the control instruction transmission adopts CRC-16 check, and automatically retransmits when the error rate is greater than 1‰.

[0087] Intelligent predictive control includes AI prediction model, anomaly detection module and Web interactive interface.

[0088] The AI ​​prediction model is a time-series prediction model based on a long short-term memory (LSTM) network. The model's input layer receives the time-series process parameters from the standardized dataset, including particle size distribution, conductivity, pH, and the current operating parameters of the ultrasonic disperser and microfluidizer 4. The hidden layer utilizes a two-layer LSTM structure to capture the dynamic relationships between process parameters. The output layer generates optimal control instructions for the ultrasonic frequency adjustment, microfluidic pressure adjustment, and electric valve opening.

[0089] The model is trained using supervised learning methods. The training dataset is derived from over 10,000 valid data sets accumulated during historical production processes. These data sets cover the dispersion process parameters and corresponding performance indicators of a variety of typical nanophotocatalysts, including TiO2, BiVO4, ZnO, and their composites. The training process uses mean squared error as the loss function, and the Adam optimizer is used for parameter optimization. After model training, it must undergo 5-fold cross-validation to ensure a prediction accuracy of at least 93% on an independent test set before deployment.

[0090] The model's anomaly detection module establishes a parameter fluctuation warning interval by setting the 3σ principle, triggering a reflux mechanism when three consecutive sampling values ​​exceed the interval. The web interactive interface supports engineers in manually setting parameter thresholds, viewing real-time production curves, and exporting historical process reports. It has a multi-level authority management function, enabling hierarchical control by administrators and operators.

[0091] Furthermore, in this embodiment, the model output layer of the AI ​​prediction model is mapped to control instructions through the fully connected layer: ultrasonic frequency adjustment amount (±5kHz step), microjet pressure adjustment amount (±50bar step), valve opening (0-100% linear adjustment), and the instructions are sent to the corresponding device in real time through the Modbus-RTU protocol.

[0092] The online laser particle size analyzer 14 includes a sampling system and an analysis system. The sampling system, pH sensor 12, and conductivity sensor 15 are all installed on the No. 3 conveying pipe 5. They are used to collect the particle size distribution, pH value, and conductivity parameters of the water-based paint in real time. After data cleaning, the data is input into a pre-trained AI prediction model. The AI ​​prediction model generates and executes dynamic adjustments to the ultrasonic frequency, micro-jet pressure, and valve opening based on historical data. When the parameters meet the requirements, the discharge is controlled and a process report is automatically generated.

[0093] In this embodiment, the process report is a PDF format document that automatically records the following data: a dispersion process timeline curve of dispersion quality parameters, process timing parameters, and is available for download via a web interface;

[0094] Dispersion quality parameters include: particle size distribution measured in real time by an online laser particle size analyzer; conductivity fluctuation measured by a conductivity sensor 15; and pH value measured by a pH sensor 12.

[0095] Process timing parameters include: time axis curve generated by computer real-time database, valve action log derived from the 4-20mA valve position feedback signal of the electric valve;

[0096] Equipment operating parameters: The cycle number is obtained from the OPC-UA interface of the microfluidizer 4.

[0097] The process supervision system includes a preset basic anomaly detection module of the AI ​​prediction model, which triggers an alarm and switches to a preset safety mode when the collected parameters deviate from the threshold.

[0098] The safety mode is preset in the system in advance: when any dispersion quality parameter exceeds the threshold range three times in a row, the system automatically closes the feed valve, switches to internal circulation reflux, triggers the sound and light alarm and resets the equipment parameters to the baseline value.

[0099] The above threshold range means that the online laser particle size analyzer 14 detects D90>300nm, the conductivity sensor 15 detects conductivity fluctuation>15%, and the pH sensor 12 detects pH<6.5 or>9.0. The above parameter ranges can be adjusted according to actual conditions.

[0100] The above-mentioned reference values ​​refer to the ultrasonic disperser 2 outputting an ultrasonic wave of 20 kHz and the microjet homogenizer 4 being set to a microjet of 500 bar. The above parameter ranges can be adjusted according to actual conditions.

[0101] The sampling system of the online laser particle size analyzer 14 can be configured with three sets of rotating sampling systems to achieve multi-point sampling. The sampling drive mechanism is simple and stable and reliable. The clean diluent in the dilution component can reverse clean the sampling line to ensure that no residual sample is collected during the next sampling.

[0102] Furthermore, the online laser particle size analyzer 14 is configured to support a dual-mode communication interface of Modbus-RTU serial protocol or Ethernet protocol, and retains a 4-20mA analog backup channel;

[0103] Furthermore, the data packets transmitted by the online laser particle size analyzer 14 to the computer 8 include: D10, D50, D90 cumulative particle size distribution data and interval particle size distribution data. The test cycle can be configured to 1-5 minutes according to production needs.

[0104] The trained AI prediction model uses a pre-trained artificial intelligence model to identify the parameter characteristics of the coating in production and obtain the coating monitoring and identification results, including:

[0105] First, a pre-trained artificial intelligence recognition model is used to identify the parameter characteristics of the coating in production according to the preset catalyst characteristics; the recognition results include the particle size distribution of the coating, the electrical conductivity of the coating, and the pH value of the coating.

[0106] When it is identified that the paint particle size distribution, conductivity and pH value are within the preset qualified range, the electric valve is controlled to open the discharge port, and the artificial intelligence model automatically outputs the process analysis report according to the preset analysis template.

[0107] The process supervision system includes a basic anomaly detection module preset in the AI ​​prediction model. When the collected parameters deviate from the threshold, an alarm is triggered and the system switches to the preset safety mode. The web interactive interface is used to support manual intervention.

[0108] The ultrasonic disperser 2 includes an ultrasonic generator 17, an ultrasonic probe 18 and a sealed liquid storage tank 16; the ultrasonic generator 17 is connected to the sealed liquid storage tank 16 through the ultrasonic probe 18; the bottom of the sealed liquid storage tank 16 is conical, which facilitates the output of the paint from the bottom infusion port.

[0109] Furthermore, a temperature sensor is provided on the inner wall of the sealed liquid storage tank 16, and a temperature display electrically connected to the temperature sensor is provided on the outer wall of the sealed liquid storage tank. The temperature sensor and temperature display cooperate to measure the temperature within the sonic disperser. The temperature sensor and temperature display can be of a type commonly used by those skilled in the art to achieve real-time temperature measurement and display.

[0110] The ultrasonic generator 17 is connected to the computer 8 via a shielded cable with a digital signal transmitter. The intelligent predictive control and process monitoring system of the computer 8 dynamically adjusts the ultrasonic frequency of the ultrasonic generator 17 through an algorithm based on the particle size distribution monitored by the online laser particle size analyzer 14.

[0111] Furthermore, the ultrasonic disperser 2 is configured to provide real-time feedback of power, frequency, and amplitude monitoring data via an RS-485 or Ethernet digital interface based on the Modbus RTU / TCP protocol, and supports adjustment of dispersion parameters via function code writing commands. Commercially available mature products can be used as the equipment.

[0112] Furthermore, the microfluidizer 4 is configured to provide real-time feedback of pressure, flow, and temperature monitoring data via an RS-485 or Ethernet digital interface based on the Modbus RTU / TCP protocol, and supports adjustment of dispersion parameters via function code writing instructions. The device can be a commercially available mature product.

[0113] Furthermore, the online laser particle size analyzer 14 is configured to transmit data via an RS-485 or Ethernet digital interface, and the sampling frequency is not less than 1 Hz. The equipment can be a commercially available mature product.

[0114] Furthermore, the electric valve supports PWM or analog input to adjust the opening and has a valve position feedback signal. The equipment can be a mature product available on the market.

[0115] Furthermore, the electric valve is configured to receive control instructions through a 4-20mA analog signal and feedback real-time opening, with an opening adjustment resolution of ≤0.5%, supporting positioning of fully open (100%), fully closed (0%) and any intermediate position; and automatically reset to a preset safety position when the signal is interrupted.

[0116] The catalytic performance of the coating after dispersion treatment is calculated and evaluated by the intelligent predictive control and process supervision system; the evaluation results include: catalyst dispersion degree and catalyst activity.

[0117] The ultrasonic disperser 2 and the micro jet homogenizer 4 are equipped with a low temperature control device to control the material temperature during the treatment process below 30° C. to suppress the catalyst activity attenuation caused by heat generation by ultrasonic waves and micro jets.

[0118] The inner wall of the delivery pipe is coated with a super-hydrophobic coating to reduce catalyst adhesion loss. Further, the delivery pipe model and type adopt the model and type commonly used by those skilled in the art, and can meet production needs. The type of super-hydrophobic coating adopts the type commonly used by those skilled in the art, and can reduce catalyst adhesion loss.

[0119] The two three-way valves each include a first interface 19, a second interface 20, and a third interface 21; both receive instructions from the computer 8 via digital signals or bus protocols;

[0120] The first interface 19 of the second electric valve 13 is connected to the second interface 20 of the first electric valve 6 through the fourth delivery pipe 7, and the first interface 19 of the first electric valve 6 is connected to the third delivery pipe 5;

[0121] The second interface 20 of the second electric valve 13 is connected to the feed port of the sealed liquid storage tank 16 of the ultrasonic disperser 2 through the first delivery pipe 1;

[0122] The No. 3 interface 21 of the No. 2 electric valve 13 is connected to the mixed liquid of the catalyst and the water-based paint through the No. 6 delivery pipe 11 for feeding, and the No. 3 interface 21 of the No. 1 electric valve 6 is connected to the No. 5 delivery pipe 9 for discharging.

[0123] The present invention also provides an intelligent dispersion method for preparing high-performance nano-photocatalytic water-based coatings, comprising the following steps:

[0124] Step S0: Preset Data: The intelligent predictive control and process monitoring system in computer 8 calls preset parameters based on the catalyst type, selecting appropriate parameters such as ultrasonic frequency, microfluidic pressure, flow rate, cooling temperature, and number of cycles. Input is sent to ultrasonic disperser 2 and microfluidizer 4. Furthermore, port 3 21 and port 2 20 of electric valve 2 are controlled to open.

[0125] Step S1: Premixing: adding nano-photocatalyst to water-based coating and mixing by mechanical stirring to form a preliminary catalyst coating mixture;

[0126] Step S2: Multi-stage dispersion: The catalyst-coating mixture enters the No. 2 port 20 of the No. 2 electric valve 13 through the No. 6 delivery pipe 11 and is introduced into the sealed liquid storage tank 16 of the ultrasonic disperser 2 through the No. 1 delivery pipe 1. The computer 8 sends a preset frequency parameter to the ultrasonic generator 17, and the ultrasonic probe 18 emits ultrasonic waves to break up the catalyst agglomerates. The preliminary catalyst-coating mixture is then subjected to ultrasonic dispersion and micro-jet homogenization by the micro-jet homogenizer 4 in sequence, preliminarily breaking up the agglomerates to obtain a dispersed coating. After the ultrasonic treatment of the coating is completed, the coating is discharged from the bottom of the sealed liquid storage tank 16. In addition, in this embodiment, dispersing the catalyst-coating mixture should include ultrasonic dispersion of the catalyst agglomerates and step-by-step mixing of the catalyst and coating particles. This embodiment presets a multi-stage dispersion parameter combination to achieve one-click call of the required parameters for each instrument in the system according to the usage conditions of different catalysts.

[0127] Step S3: Real-time monitoring: After discharge, the material enters the microfluidizer 4 through the second delivery pipe 3, and the shear force generated by the microjet refines the catalyst particles to the nanometer level;

[0128] The conductivity sensor 15, pH sensor 12 and online laser particle size analyzer 14 installed on the No. 3 conveying pipe 5 respectively collect samples from the sampling port installation location for analysis; the online laser particle size analyzer 14 sequentially monitors the dispersed paint particle size distribution, pH value and conductivity data of the conductivity sensor 15 and pH sensor 12 in real time; and synchronously collects the data and transmits it to the computer 8;

[0129] Among them, real-time monitoring should include monitoring the size of catalyst and coating particles in the coating and monitoring the conductivity of the coating to determine the agglomeration trend;

[0130] Step S4: Intelligent evaluation and adjustment: After data cleaning, the data is input into the AI ​​prediction model of the computer that has been trained in advance. The AI ​​prediction model generates and executes dynamic adjustment of ultrasonic frequency, microjet pressure and valve opening based on historical data to optimize the dispersion effect; by dynamically optimizing production parameters such as ultrasonic power, microfluidic flow rate, and dispersant addition amount.

[0131] The reference control parameter of the stirring speed is output to the electric valve, ultrasonic disperser 2 and microfluidizer 4 in the form of a 4-20 mA analog signal.

[0132] Step S5: When the material parameters are qualified, the computer 8 controls the No. 1 electric valve 6 to open the No. 3 interface 21 to start the material discharge, and automatically generates a process report;

[0133] The first interface 19 of the second electric valve 13 is controlled to be closed, and the third interface 21 and the second interface 20 of the second electric valve 13 are opened to receive the catalyst coating mixture through the sixth delivery pipe 11, and then introduced into the sealed liquid storage tank 16 of the ultrasonic disperser 2 through the second interface 20 of the second electric valve 13 through the first delivery pipe 1 for subsequent processing;

[0134] When the parameters of the material are unqualified, the computer 8 controls the opening of the second interface 20 of the No. 1 electric valve 6 to enter the No. 1 interface 19 of the No. 2 electric valve 13 through the No. 4 delivery pipe 7, closes the No. 6 delivery pipe 11 of the No. 2 electric valve 13, stops receiving new catalyst coating mixture, opens the No. 2 interface 20 of the No. 2 electric valve 13, receives the unqualified material through the No. 1 delivery pipe 1 and introduces it into the sealed liquid storage tank 16 of the ultrasonic disperser 2, and returns it to the system for reprocessing.

[0135] This example solves the problem of photocatalyst performance degradation in coating production by utilizing ultrasonic-microfluidic synergistic dispersion, online monitoring, and AI dynamic control technology, achieving efficient, low-cost, and stable nanodispersion. The system is adaptable to multiple catalysts and can be widely used in industrial-grade coating production.

[0136] The following is a detailed description of the specific parameters:

[0137] Example 1

[0138] This embodiment takes TiO2 photocatalyst as an example to specifically illustrate the implementation method of the intelligent dispersion system of the present invention. First, the preset parameters of the TiO2 catalyst are selected in the intelligent formula library of the computer 8, including ultrasonic frequency of 20kHz, microjet pressure of 1000bar, three cycles and cooling temperature of 25°C. After the TiO2 catalyst and water-based paint are mixed in a mass ratio of 1:100 in advance, it enters the sealed liquid storage tank 16 of the ultrasonic disperser 2 through the No. 6 delivery pipe 11. The ultrasonic probe 18 works at a preset frequency for 10 minutes to preliminarily break up the TiO2 agglomerates. Subsequently, the slurry enters the microjet homogenizer 4 through the No. 2 delivery pipe 3 to refine the particles to D50≤100nm. The online laser particle size analyzer 14 monitors the particle size distribution in real time, the conductivity sensor 15 and the pH sensor 12 synchronously collect data and transmit it to the computer 8, and the AI ​​prediction model dynamically adjusts the microjet pressure to 1100bar to optimize the dispersion effect. When D90≤200nm and the conductivity is stable, the No. 1 electric valve 6 opens the No. 3 interface to discharge the material, otherwise it returns to the system through the No. 2 interface for reprocessing.

[0139] Example 2

[0140] This example uses TiO2 photocatalyst as an example to specifically illustrate the implementation of the intelligent dispersion system of the present invention. First, the preset parameters of the TiO2 catalyst are selected from the intelligent formula library of computer 8, including an ultrasonic frequency of 20kHz, a microjet pressure of 1000bar, a number of cycles of 3, and a cooling temperature of 25°C. After the TiO2 catalyst and water-based paint are mixed in advance at a mass ratio of 1:100, it enters the sealed liquid storage tank 16 of the ultrasonic disperser 2 through the No. 6 delivery pipe 11. The ultrasonic probe 18 operates at a preset frequency for ten minutes to initially break up the TiO2 agglomerates. Subsequently, the slurry enters the microjet homogenizer 4 through the No. 2 delivery pipe 3 and refines the particles to D50 ≤ 100nm. The online laser particle size analyzer 14 monitors the particle size distribution in real time, and the conductivity sensor 15 and pH sensor 12 synchronously collect data and transmit it to the computer 8. The AI ​​prediction model dynamically adjusts the microjet pressure of the microjet homogenizer 4 to 1100bar to optimize the dispersion effect. When D90≤200nm and the conductivity is stable, the No. 1 electric valve 6 opens its No. 3 interface 21 to discharge the material, otherwise it returns to the system through the No. 2 interface 20 of the No. 1 electric valve 6 for reprocessing.

[0141] Example 3

[0142] This example uses the composite photocatalyst BiVO4 / ZnO as an example to demonstrate the system's adaptability to multi-component catalysts. The coordinated dispersion parameters for BiVO4 / ZnO are pre-loaded into the intelligent recipe library: alternating ultrasonic frequencies of 25kHz and 40kHz, a microjet pressure of 800 bar, and five cycles. BiVO4 and ZnO are pre-mixed in a 1:2 mass ratio and then added to a water-based coating. The mixture then enters ultrasonic disperser 2 via pipe 1, where alternating dual-frequency ultrasonic waves are applied for 15 minutes to overcome sedimentation caused by differences in component density. As the slurry passes through microfluidizer 4, an online laser particle size analyzer 14 detects a D50 of 120nm for BiVO4 particles and 80nm for ZnO. The AI ​​prediction model automatically increases the number of cycles to seven and adjusts the pH to 8.5 to balance dispersion efficiency. When the final slurry reaches a D90 of ≤180nm and a conductivity fluctuation of <5%, electric valve 6 opens the outlet for qualified products. The system also generates an optimization report including particle size distribution, energy consumption, and process parameters.

[0143] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like to indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0144] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0145] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. An intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coatings, characterized in that: It comprises a material processing unit, an online monitoring unit, an execution control unit, a computer (8) and a pipeline system; the computer (8) is in communication connection with the material processing unit, the online monitoring unit and the execution unit; The material processing unit includes an ultrasonic disperser (2) and a microfluidizer (4); The online monitoring unit includes an online laser particle size analyzer (14), a conductivity sensor (15) and a pH sensor (12); The execution control unit comprises two three-way valves of the same structure: a No. 1 electric valve (6) and a No. 2 electric valve (13), wherein the No. 1 electric valve (6) is used to switch between the internal circulation and discharge modes; and the No. 2 electric valve (13) is used to switch between the internal circulation and introduction modes. The computer (8) includes an intelligent predictive control and process monitoring system for real-time data collection, processing, and adjustment; The pipeline system comprises a No. 1 delivery pipe (1), a No. 2 delivery pipe (3), a No. 3 delivery pipe (5) and a No. 4 delivery pipe (7); The second electric valve (13) is connected to the ultrasonic disperser (2) through the first delivery pipe (1), the ultrasonic disperser (2) is connected to the micro jet homogenizer (4) through the second delivery pipe (3), the micro jet homogenizer (4) is connected in series to the first electric valve (6) through the third delivery pipe (5) with a conductivity sensor (15), a pH sensor (12) and an online laser particle size analyzer (14), and the first electric valve (6) is connected to the second electric valve (13) through the fourth delivery pipe (7); The intelligent prediction control includes an AI prediction model and an abnormality detection module; the conductivity sensor (15), pH sensor (12) and online laser particle size analyzer (14) collect the particle size distribution, pH value and conductivity parameters of the water-based paint in real time, and input the data into the pre-trained AI prediction model after data cleaning. The AI ​​prediction model generates and executes dynamic adjustment of ultrasonic frequency, micro-jet pressure and valve opening based on historical data, and controls the discharge opening when the parameters are qualified and automatically generates a process report; The process report automatically records the following data: dispersion process time axis curve of dispersion quality parameters, process timing parameters; The dispersion quality parameters include: particle size distribution measured in real time by an online laser particle size analyzer (14); conductivity fluctuation measured by a conductivity sensor (15); and pH value measured by a pH sensor (12); The process timing parameters include: the time axis curve is generated by the real-time database of the computer (8), and the valve action log is derived from the feedback signal of the electric valve; Equipment operating parameters: the number of cycles is obtained from the interface of the microfluidizer (4); The AI ​​prediction model is a time-series prediction model built based on a long short-term memory (LSTM) network. The model's input layer receives time-series process parameters from a standardized data set, including particle size distribution, conductivity, pH value, and the current operating parameters of the ultrasonic disperser and microfluidizer. The hidden layer uses a two-layer LSTM structure to capture the dynamic correlation between process parameters. The output layer generates optimal control instructions for the ultrasonic frequency adjustment amount, microfluidic pressure adjustment amount, and electric valve opening. The AI ​​prediction model is trained using a supervised learning method. The training data set is derived from more than 10,000 sets of data records accumulated during historical production processes. The records need to cover the dispersion process parameters of TiO2, BiVO4, ZnO and their composite materials, as well as the corresponding performance indicators of nano-photocatalysts. The training process uses mean square error as the loss function and uses the Adam optimizer for parameter optimization. After the AI ​​prediction model is trained, it must undergo a five-fold cross-validation to ensure that its prediction accuracy on an independent test set is not less than 93%; The anomaly detection module of the AI ​​prediction model establishes a parameter fluctuation warning interval by setting the 3σ principle, and triggers the reflux mechanism when three consecutive sampling values ​​exceed the interval; The process monitoring system includes a preset basic anomaly detection module of the AI ​​prediction model, which triggers an alarm and switches to a preset safe mode when the collected parameters deviate from the threshold; The system has a pre-set safety mode: when any dispersion quality parameter exceeds the threshold value three times in a row, the system automatically closes the feed valve, switches to internal circulation reflux, triggers an audible and visual alarm, and resets the equipment parameters to the baseline value. The threshold range refers to the following: the online laser particle size analyzer (14) detects D90>300nm, the conductivity sensor (15) detects conductivity fluctuation>15%, and the pH sensor (12) detects pH<6.5 or>9.0; The reference value refers to the ultrasonic disperser (2) outputting ultrasonic waves at 20 kHz and the microjet homogenizer (4) setting the microjet at 500 bar.

2. The intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coating according to claim 1, characterized in that: The intelligent predictive control and process monitoring system includes a real-time data acquisition module, a parameter identification module, a communication module and a safety protection mechanism, and connects the material processing unit, the online monitoring unit and the execution unit through an industrial communication protocol to achieve closed-loop control; The specific implementation of the intelligent predictive control and process monitoring system is as follows: The real-time data acquisition module synchronously collects the particle size distribution, conductivity and pH value data of the slurry at a frequency of ≥1 Hz through an online laser particle size analyzer (14), a conductivity sensor (15) and a pH sensor (12); the data is transmitted to the computer in real time through the Modbus-RTU / TCP industrial bus protocol, with a transmission delay of ≤20 ms; the collected raw data is pre-processed by sliding window mean filtering and z-score standardization to eliminate noise and dimensional differences; The safety protection mechanism includes a multi-level interlocking protection module, a hardware redundancy module, an emergency power supply module and a physical isolation module; The multi-level interlocking protection module is started as follows: When the online laser particle size analyzer (14) detects that D90>300nm exceeds the limit continuously, the first level protection is triggered and the micro jet pressure is automatically reduced to the reference value of 500bar; When the conductivity sensor (15) detects a fluctuation of >15% and the pH exceeds the range of 6.5-9.0, the secondary protection is triggered, the feed valve is immediately closed and the internal circulation reflux is started; When the temperature sensor detects that the material temperature is ≥32°C, the third level protection is triggered, the low temperature control device is forced to start cooling and the ultrasonic output is suspended; The hardware redundancy module includes a conductivity sensor (15), a pH sensor (12), and an online laser particle size analyzer (14), all of which adopt a dual-probe redundancy design and automatically switch to the backup signal when the difference between the master and slave probe data is greater than 5%; The electric valve is equipped with dual control signals, which will automatically switch to the 4-20mA analog backup channel when the main control signal fails; The emergency power supply module includes a UPS uninterruptible power supply, which maintains the system's minimum power consumption for ≥30 minutes when the main power is interrupted, giving priority to sensor data storage and valve safety position switching; The physical isolation module includes isolating the strong electric control circuit and the weak electric signal circuit through a photoelectric coupler to prevent malfunction caused by electromagnetic interference; the control instruction transmission adopts CRC-16 check, and automatically retransmits when the error rate is greater than 1‰.

3. The intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coating according to claim 2, characterized in that: The intelligent prediction control also includes a Web interactive interface; The process report is a PDF document and can be downloaded through the web interface; The valve action log is derived from the 4-20mA valve position feedback signal of the electric valve; The operating parameters of the device: the number of cycles is obtained from the OPC-UA interface of the microfluidizer (4); The web interactive interface allows engineers to manually set parameter thresholds, view real-time production curves, and export historical process reports. It also has multi-level authority management capabilities, enabling hierarchical control by administrators and operators. The web interactive interface is used to support manual intervention.

4. The intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coating according to claim 2, characterized in that: The ultrasonic disperser (2) includes an ultrasonic generator (17), an ultrasonic probe (18) and a sealed liquid storage tank (16); the ultrasonic generator (17) is connected to the sealed liquid storage tank (16) via the ultrasonic probe (18); the ultrasonic generator (17) is connected to a computer (8), and the intelligent predictive control and process monitoring system of the computer (8) dynamically adjusts the ultrasonic frequency of the ultrasonic generator (17) through an algorithm based on the particle size distribution monitored by the online laser particle size analyzer (14).

5. The intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coating according to claim 1, characterized in that: The ultrasonic disperser (2) and the micro jet homogenizer (4) are equipped with a low temperature control device for controlling the temperature of the material during the treatment process to below 30°C, so as to suppress the attenuation of the catalyst activity caused by the heat generated by the ultrasonic wave and the micro jet.

6. The intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coating according to claim 1, characterized in that: The inner walls of the pipeline system are coated with a super-hydrophobic coating to reduce catalyst adhesion loss.

7. The intelligent dispersing system for preparing high-performance nano-photocatalytic water-based coating according to claim 4, characterized in that: The two three-way valves each include a first interface (19), a second interface (20) and a third interface (21); The No. 1 interface (19) of the No. 2 electric valve (13) is connected to the No. 2 interface (20) of the No. 1 electric valve (6) via the No. 4 delivery pipe (7), and the No. 1 interface (19) of the No. 1 electric valve (6) is connected to the No. 3 delivery pipe (5); The second interface (20) of the second electric valve (13) is connected to the feed port of the sealed liquid storage tank (16) of the ultrasonic disperser (2) through the first delivery pipe (1); The No. 3 interface (21) of the No. 2 electric valve (13) is connected to the mixed liquid of the catalyst and the water-based paint through the No. 6 delivery pipe (11) for feeding, and the No. 3 interface (21) of the No. 1 electric valve (6) is connected to the No. 5 delivery pipe (9) for discharging.

8. A method for preparing an intelligent dispersion system for high-performance nano-photocatalytic water-based coatings according to claim 7, characterized in that: The steps include: Step S1: Premixing: adding nano-photocatalyst to water-based coating and mixing by mechanical stirring to form a preliminary catalyst coating mixture; Step S2: Dispersion: The catalyst coating mixture enters the No. 2 interface (20) of the No. 2 electric valve (13) through the No. 6 delivery pipe (11), and is introduced into the sealed liquid storage tank (16) of the ultrasonic disperser (2) through the No. 1 delivery pipe (1). The ultrasonic probe (18) operates at a preset frequency to preliminarily break up the agglomerates to obtain the dispersed coating; Step S3: Real-time monitoring: The particles are fed into the microfluidizer (4) through the second delivery pipe (3) to be refined, and the dispersed coating particle size distribution, the pH value and conductivity data of the conductivity sensor (15) and the pH sensor (12) are monitored in real time in the online laser particle size analyzer (14); the data are collected synchronously and transmitted to the computer (8); Step S4: Intelligent evaluation and adjustment: After data cleaning, the data is input into the AI ​​prediction model of the pre-trained computer (8). The AI ​​prediction model generates and executes dynamic adjustment of ultrasonic frequency, micro-jet pressure and valve opening based on historical data to optimize the dispersion effect; Step S5: When the material parameters are qualified, the computer (8) controls the No. 1 electric valve (6) to open the No. 3 interface (21) to start the material discharge, and automatically generates a process report; The first interface (19) of the second electric valve (13) is controlled to be closed, and the third interface (21) and the second interface (20) of the second electric valve (13) are opened to receive the catalyst coating mixture through the sixth delivery pipe (11), and the catalyst coating mixture is introduced into the sealed liquid storage tank (16) of the ultrasonic disperser (2) through the second interface (20) of the second electric valve (13) through the first delivery pipe (1) to continue the subsequent processing; When the material parameters are unqualified, the computer (8) controls the opening of the second interface (20) of the first electric valve (6) to enter the first interface (19) of the second electric valve (13) through the fourth delivery pipe (7), closes the sixth delivery pipe (11) of the second electric valve (13), stops receiving new catalyst coating mixture, opens the second interface (20) of the second electric valve (13), receives the unqualified material through the first delivery pipe (1) and introduces it into the sealed liquid storage tank (16) of the ultrasonic disperser (2), and returns it to the system for reprocessing.

9. The method for preparing an intelligent dispersion system for high-performance nano-photocatalytic water-based coating according to claim 8, characterized in that: Step S2 adopts multi-stage dispersion: the preliminary catalyst coating mixture is subjected to ultrasonic dispersion and microfluidizer treatment in sequence.

10. The method for preparing an intelligent dispersion system for high-performance nano-photocatalytic water-based coating according to claim 8, characterized in that: In step S3, the real-time monitoring includes monitoring the size of the catalyst and the coating particles in the coating and monitoring the conductivity of the coating to determine the agglomeration trend.

Images