Sewage treatment method based on photocatalyst sewage treatment process and intelligent illumination system adopted by sewage treatment method

Abstract

The invention discloses a sewage treatment method based on a photocatalyst sewage treatment process and an intelligent illumination system adopted thereof.The sewage treatment method comprises the following steps that S1, pretreatment is conducted, specifically, wastewater is injected into an adjusting tank and flows into a pre-reaction tank after being subjected to rough filtration and pH adjustment, and large-particle impurities are removed; s2, carrying out photocatalytic reaction; and S3, carrying out post-treatment such as filtration and disinfection on the treated sewage to ensure that the effluent quality reaches the standard. The core device intelligent illumination system for sewage treatment comprises a photocatalyst sewage treatment light control device, a pre-measurement control system and a man-machine multi-mode interaction system, and is suitable for treatment of industrial wastewater, domestic wastewater and sewage containing organic pollutants, heavy metal ions and microorganisms.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and in particular to a sewage treatment method based on a photocatalytic sewage treatment process and an intelligent lighting system used therein. Background Art

[0002] Traditional wastewater treatment technologies, such as physical adsorption, chemical precipitation, and biodegradation, suffer from low treatment efficiency, high costs, and secondary pollution. Photocatalytic water treatment technology is an advanced process that utilizes photocatalytic oxidation reactions to degrade pollutants in water. It relies primarily on photocatalysts (such as TiO2) to produce strong oxidizing free radicals (such as ·OH) under ultraviolet or visible light, effectively decomposing organic matter, killing microorganisms, and removing heavy metals. As an emerging advanced oxidation technology, photocatalytic technology offers significant advantages such as high efficiency, environmental friendliness, and sustainability, and exhibits enormous potential for application in wastewater treatment.

[0003] Existing technical research mainly focuses on the optimization of photocatalytic materials or reactor structures, but research on their intelligent operation and human-computer interaction optimization is still lagging behind. During the operation of photocatalytic sewage treatment plants, the time-varying nature of water quality, real-time monitoring of equipment efficiency degradation, management of the impact of human negative emotions, real-time status of human operational management capabilities, and real-time intelligence of control systems are all major factors affecting the high-efficiency, environmentally friendly, and energy-saving operation of water plants. However, with the rapid development of artificial intelligence disciplines such as soft measurement technology and human-computer interaction, the adaptive operation of water plants has reached a new level of intelligence. In order to avoid accidents caused by factors such as restrictive operation of equipment and operators, an operation method that combines intelligent prediction and interactive control is urgently needed to improve the adaptive capabilities of the system.

[0004] At present, the specific problems existing in the operation of photocatalytic wastewater treatment technology are as follows:

[0005] 1. Because photocatalyst materials have a narrow light response range, primarily limited to the ultraviolet region, their utilization of sunlight is low. Furthermore, they are affected by factors such as weather and structural obstruction, placing higher demands on real-time intelligent control of light sources. Traditional PLC control, however, suffers from poor processing performance and high energy consumption due to its time delay.

[0006] 2. Due to fluctuations in water quantity and quality, photocatalytic wastewater treatment processes are time-varying and nonlinear. Data collected by detection instruments and manual experiments are subject to errors and cannot meet the requirements of intelligent real-time control. Soft measurement technology is needed to replace traditional detection technology. However, the practical application of soft measurement technology in photocatalytic wastewater treatment processes is still weak.

[0007] 3. Human-machine multi-channel interaction is inadequate. Differences exist between human operator skills, equipment performance, and the multi-model control system, requiring a high degree of integration. For example, factors such as photocatalyst activity decay and light source efficiency decline affect treatment effectiveness. Human emotional management lacks intelligent predictive capabilities, making it impossible to anticipate system failures or efficiency reductions. All of these require human-machine integration.

[0008] In view of this, it is of great significance to develop a multimodal human-computer interaction method for an efficient, stable and intelligent photocatalytic wastewater treatment system. Summary of the Invention

[0009] The purpose of the present invention is to provide a sewage treatment method based on a photocatalytic sewage treatment process and the intelligent lighting system used therein, which, on the one hand, solves the problem of high energy consumption due to poor real-time performance of existing photocatalytic water treatment lighting control technology, and on the other hand, solves the problem of insufficient intelligent interactivity in the management of the existing status of water plant operators and machines, thereby realizing efficient, stable, energy-saving and sustainable sewage treatment.

[0010] To achieve the above objectives, the present invention provides a sewage treatment method based on a photocatalytic sewage treatment process. The sewage treatment system includes a regulating tank, a pre-reaction tank, a photocatalytic reactor, a clear water tank and a central control system. The sewage treatment method includes the following steps:

[0011] S1: Pretreatment: Wastewater is injected into the regulating tank, and after coarse filtration and pH adjustment, it flows into the pre-reaction tank to remove large particles of impurities;

[0012] S2: Photocatalytic reaction: The wastewater from the pre-reaction tank is introduced into the photocatalytic reactor of the reaction tank, and a photocatalytic reaction is carried out under light conditions to degrade organic pollutants, remove heavy metal ions, and inactivate microorganisms. The influent characteristics and effluent efficiency characteristics of the influent into the above-mentioned reaction tank are measured in a composite manner to determine the current dynamic values;

[0013] S3: Filter and disinfect the treated sewage to ensure that the effluent water quality meets the standards;

[0014] The central control system is connected to sensors and actuators provided in the regulating tank, pre-reaction tank, photocatalytic reactor, and clear water tank via telecommunications. The central control system collects measurement parameters of the sensors and operating parameters of the actuators, compares the measured parameters through a computer program, calculates a suitable target value, compares the calculated target value with the received corresponding parameters, and processes the calculated target value into analog and digital control output values. The obtained control output values ​​are used to regulate the processes of pretreatment, photocatalytic reaction, and post-treatment.

[0015] The program obtains the above-mentioned appropriate target value by adopting the control program of the ANFIS system.

[0016] According to the embodiment of the present application, based on the ANFIS system, according to the measurement parameters of the pretreatment stage, including sewage inflow flow, pH value, BOD and COD concentration of injected water, the execution parameters of the pretreatment stage are adjusted, including pump speed, valve opening, and chemical dosage.

[0017] According to an embodiment of the present application, based on the ANFIS system, the execution parameters of the photocatalytic reaction stage, including COD, TOC, UV transmittance, catalyst activity, and water flow rate, are adjusted according to the measurement parameters of the photocatalytic reaction stage, including UV-LED or mercury lamp intensity and reagent dosage.

[0018] According to an embodiment of the present application, the central control system includes a central computer and a PLC controller communicatively connected to the central computer, and the PLC controller is telecommunication-connected to sensors and actuators provided in the regulating tank, pre-reaction tank, photocatalyst reactor and clear water tank.

[0019] According to an embodiment of the present application, the ANFIS system includes an ANFIS neural network prediction model, a fuzzy controller and an optimization processor, wherein the ANFIS neural network prediction model takes the inlet flow, inlet COD and illuminance as input vectors, and the output vector is the outlet COD. The fuzzy controller takes the deviation between the predicted value and the expected value of the outlet COD of the ANFIS neural network prediction model and the deviation change rate as input, and the illuminance correction amount as the control output. The current light intensity is corrected through the control output, thereby completing the automatic adjustment of the light intensity.

[0020] Another aspect of the present application discloses an intelligent lighting system, which is applied to a sewage treatment method based on a photocatalytic sewage treatment process as described above, and is characterized in that it includes a photoreaction module, an intelligent prediction and control module and a human-computer interaction module. The intelligent prediction and control module receives the data signal of the photoreaction module for processing, and the processed data is integrated in the human-computer interaction module. The human-computer interaction module sends instructions to control the operation of the photoreaction module according to the data analysis results.

[0021] According to an embodiment of the present application, the photoreaction module includes a photoreactor, a light source and a multi-series sensor. The photoreactor adopts a fixed reactor, and the TiO2 / CdS composite photocatalyst material is loaded on a porous ceramic carrier and filled into the fixed reactor; the light source adopts an ultraviolet-visible composite light source, which is installed on the top of the fixed reactor to simulate the sunlight spectrum; the multi-series sensor includes a pH sensor, a temperature sensor and a flow meter equipped in the fixed reactor, and the multi-series sensor is telecommunication-connected to the intelligent predictive control module.

[0022] According to an embodiment of the present application, a photosensor is arranged in the photoreactor to ensure that there is no blind spot in the illumination.

[0023] According to the embodiments of the present application, the intelligent predictive control module is based on a fuzzy control algorithm and a neural network control program. According to the collected current water quality parameters, current light intensity and catalyst status, the ANFIS prediction model predicts the pollutant degradation efficiency or target concentration under different light intensities based on the input. Based on the prediction results, the light intensity is dynamically adjusted through the optimization algorithm, the power or switching time of the light source is adjusted, and the actual degradation effect is fed back through the sensor to correct the model error.

[0024] According to an embodiment of the present application, the human-computer interaction module includes a touch screen control terminal, a mobile APP remote monitoring system, an AR maintenance guidance system, a voice interaction interface and an emotion manager. The touch screen control terminal is connected to the central control system and serves as the main operation interface for real-time control and display of the status of the light reaction module; the mobile APP remote monitoring system is connected to the central control system via a network. The user can send instructions through the APP. These instructions are transmitted to the central control system via the network and then forwarded by the central control system to the intelligent prediction control module. The intelligent prediction control module then controls the status of the light reaction module. At the same time, the status information of the light reaction module is transmitted back to the APP via the same path; the AR maintenance guidance system is connected to the sensor and camera of the light reaction module to obtain real-time images and data of the module, and is connected to the central control system via the network to obtain maintenance guidance information and virtual models. These contents are displayed to maintenance personnel through the AR device; the voice interaction interface is connected to the voice processing unit of the intelligent prediction control module and interacts with the user through a microphone and a speaker; the emotion manager is connected to the sensor, camera and user behavior analysis unit of the light reaction module. The emotion manager judges the user's emotional state by analyzing the user's facial expressions and voice intonation, and adjusts the interaction mode according to the emotional state.

[0025] The beneficial effects of the technical solution of the present invention compared with the prior art are:

[0026] 1. The sewage treatment system based on the photocatalytic sewage treatment process of this application combines artificial intelligence prediction models and human-computer interaction methods to achieve efficient and intelligent water purification operation management.

[0027] 2. The sewage treatment based on the photocatalytic sewage treatment process of this application can achieve the following effects through an intelligent control system: energy efficiency optimization: dynamic lighting strategy can reduce energy consumption by 20%-40%; extend equipment life: avoid light source overload and reduce photocatalyst deactivation; operation-friendly: through a visual interface and AR assistance, reduce dependence on professionals.

[0028] 3. The sewage treatment system of this application deploys fast-response sensors for real-time prediction and response; simplifies the ANFIS structure (reduces the number of fuzzy rules) to meet real-time requirements, or uses edge computing equipment for deployment, lightweighting the system model; adopts the ANFIS system, emphasizes the influence of multiple variables, and realizes multi-parameter coupling. When the actual degradation effect deviates greatly from the prediction, the model parameters are triggered to be updated online, and feedback compensation can be performed to achieve anti-interference design; energy consumption is used as an additional output of ANFIS, and "treatment effect-energy consumption" multi-objective optimization is realized in the control to achieve energy-saving optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a schematic structural diagram of a photocatalytic water treatment system according to an example of the present invention;

[0030] Figure 2 This is a schematic diagram of human-computer interaction communication integration according to an example of the present invention;

[0031] Figure 3 This is a flow chart of a sewage treatment method based on a photocatalytic sewage treatment process according to an example of the present invention;

[0032] Figure 4 This is a schematic diagram of the structure of the intelligent lighting ANFIS system according to an example of the present invention;

[0033] Figure 5 Schematic diagram of the ANFIS neural network prediction model of the present invention.

[0034] The reference numerals are as follows:

[0035] 1. Equalization tank, 2. Pre-reaction tank, 3. Photocatalyst reactor, 4. Clear water tank, 5. Filter, 6. Pump, 7. Disinfector, 8. Light source, 9. Flushing pipe, 10. Multimodal information perception system, 20. Relay module, 30. Information fusion subsystem. DETAILED DESCRIPTION

[0036] The present invention will be further described below in conjunction with the accompanying drawings and specific examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0037] like Figure 3 As shown, one aspect of the present application shows a sewage treatment method based on a photocatalytic sewage treatment process, such as Figure 1 As shown, the sewage treatment system includes a regulating tank 1, a pre-reaction tank 2, a photocatalyst reactor 3, a clear water tank 4 and a central control system. The sewage treatment method includes the following steps:

[0038] S1: Pretreatment: Wastewater is injected into regulating tank 1, and after coarse filtration and pH adjustment, it flows into pre-reaction tank 2 to remove large particles of impurities;

[0039] Specifically, the following preprocessing methods can be used in the preprocessing stage:

[0040] Coarse filtration: Use filter 5 to remove suspended matter and particulate matter (such as sand filtration, activated carbon filtration);

[0041] Adjust pH: Optimize to the optimal activity range of the photocatalyst (TiO2 is usually pH 3 to 9);

[0042] Pre-oxidation (optional): For high concentration of organic matter or color water, ozone or H2O2 pretreatment can be combined.

[0043] S2: Photocatalytic reaction: The wastewater from the pre-reaction tank 2 is introduced into the photocatalytic reactor 3 of the reaction tank, and a photocatalytic reaction is carried out under light conditions to degrade organic pollutants, remove heavy metal ions, and inactivate microorganisms. The influent characteristics and effluent efficiency characteristics of the influent into the reaction tank are measured in a composite manner to determine the current dynamic values;

[0044] Specifically, the following reaction conditions can be employed for the core stage of the photocatalytic reaction: Photoreactor design: Fixed-bed reactor: The catalyst is supported on a carrier (e.g., ceramic or fiberglass), with water from flushing pipe 9 flowing through the illuminated area; Suspension reactor: Nanoscale catalysts are dispersed in water, requiring subsequent separation; Membrane reactor (integrated technology): Coupled ultrafiltration / microfiltration membranes simultaneously separate the catalyst and pollutants. Light source 8 options: Ultraviolet light (UVA 365nm) or visible light-modified catalysts (e.g., nitrogen-doped TiO2); Solar energy utilization: Parabolic concentrators or optical fibers are used to enhance light efficiency.

[0045] S3: Filter and disinfect the treated sewage to ensure that the effluent water quality meets the standards;

[0046] Specifically, the post-processing stage can adopt the following methods: catalyst recovery: centrifugation, membrane filtration (for suspension type) or direct fixed bed retention; deep purification: activated carbon adsorption of residual small molecular organic matter or ion exchange to remove heavy metals; disinfection (optional): photocatalysis itself has a bactericidal effect, using a disinfector 7, and supplementing chlorine or ultraviolet rays when necessary.

[0047] Specifically, the sewage treatment method also includes a recycling step, and the treated water can be reused for industrial cooling, irrigation, or further purified into drinking water.

[0048] In this embodiment, the central control system is connected to the sensors and actuators installed in the regulating tank 1, the pre-reaction tank 2, the photocatalytic reactor 3 and the clear water tank 4 by telecommunications. The central control system collects the measurement parameters of the sensors and the operating parameters of the actuators, compares the above-mentioned measured parameters through the program in the computer, and obtains a suitable target value. The obtained target value is then compared with the corresponding received parameters, and the calculation is processed into analog and digital control output values. The obtained control output values ​​are used to regulate the processes of pretreatment, photocatalytic reaction and post-treatment; the program obtains the above-mentioned suitable target value by adopting the control program of the ANFIS system.

[0049] Specifically, based on the ANFIS system, according to the measurement parameters of the pretreatment stage, including sewage inflow flow, pH value, BOD and COD concentration of injected water, the execution parameters of the pretreatment stage including the speed of pump 6, valve opening and dosage of chemical are adjusted.

[0050] Specifically, based on the ANFIS system, the execution parameters of the photocatalytic reaction stage, such as UV-LED or mercury lamp intensity and reagent dosage, are adjusted according to the measured parameters of the photocatalytic reaction stage, including COD, TOC, UV transmittance, catalyst activity, and water flow rate.

[0051] Specifically, the central control system includes a central computer and a PLC controller communicatively connected to the central computer, and the PLC controller is telecommunication-connected to sensors and actuators provided in the regulating tank 1 , the pre-reaction tank 2 , the photocatalytic reactor 3 and the clear water tank 4 .

[0052] In this embodiment, Figure 4 As shown in Figure 1, the ANFIS system includes an ANFIS neural network prediction model, a fuzzy controller, and an optimization processor. The ANFIS neural network prediction model takes the influent flow rate, influent COD, and illuminance as input vectors, and outputs the effluent COD. The fuzzy controller takes the deviation between the ANFIS neural network prediction model's effluent COD prediction value and the expected value, as well as the rate of change of the deviation, as input. The illuminance correction value is used as a control output, which is used to correct the current light intensity and thus automatically adjust the light intensity. The ANFIS neural network prediction model takes the influent flow rate q(t), influent COD x(t), and illuminance i(t) as input vectors, and outputs the effluent COD prediction value y(t+Δt). The fuzzy controller takes the deviation between the ANFIS neural network prediction model's effluent COD prediction value and the expected value, as well as the rate of change of the deviation ec, as input. The illuminance correction value Δi(t) is used as a control output, which is used to correct the current light intensity and thus automatically adjust the light intensity.

[0053] Specifically, if Figure 5As shown, the prediction model focuses on the photocatalytic illumination process of the reactor. Applying the theories of neural network control and fuzzy reasoning, a fuzzy neural network based on a clustering algorithm is used for illumination control design. By examining the relationships between the wastewater treatment system's effluent COD and influent COD, influent flow rate, dosage, illumination intensity, and historical values ​​of effluent COD, as well as their changing trends, the C-mean clustering method is used to summarize fuzzy rules from the samples. A hybrid algorithm is then used to identify the network structure and parameters, thereby establishing a prediction model. Finally, the configuration software and PLC programming are completed.

[0054] Specifically, fuzzy control is an intelligent control method based on fuzzy logic. It controls the system by using fuzzy language and fuzzy rules. The advantage of fuzzy control is that it has good adaptability to the uncertainty and nonlinearity of the system and can handle complex problems that are difficult to accurately model. The rules of fuzzy control are easy to understand and modify, making it convenient for operators to adjust according to actual conditions. However, fuzzy control also has some limitations. For example, the establishment of fuzzy rules requires a certain amount of experience and knowledge, and there may be subjectivity and uncertainty. In addition, the performance of fuzzy control depends largely on the quality and rationality of the fuzzy rules. Neural network control is an intelligent control method inspired by the biological nervous system. It controls complex systems by simulating the connection and information transmission mode of neurons. The advantage of neural network control is that it can automatically learn and adapt to the dynamic characteristics of the system and has strong nonlinear mapping capabilities and fault tolerance. However, the training of neural networks requires a large amount of data and computing resources, and its results may be affected by data quality and network structure. In summary, the prediction model of this application adopts the ANFIS system, which combines the advantages of fuzzy logic and neural networks to provide a powerful tool for processing complex nonlinear systems, data-driven modeling, and uncertainty processing.

[0055] Specifically, the ANFIS system can achieve the following functions for the sewage treatment system:

[0056] (1) Water quality control: Through fuzzy control algorithm, various parameters in the treatment process, such as the dosage of chemical agents, are automatically adjusted according to the water quality parameters of the sewage (such as COD, BOD, etc.) to achieve effective control of water quality.

[0057] (2) Equipment control: Fuzzy judgment is made on the operating status of the sewage treatment equipment, and its operating parameters, such as the speed of the pump 6 and the valve opening, are automatically adjusted according to the working conditions of the equipment, thereby improving the operating efficiency and stability of the equipment.

[0058] (3) Process optimization: Use fuzzy control technology to optimize the sewage treatment process and find the best operation strategy to reduce energy consumption, reduce chemical consumption and improve treatment effect.

[0059] (4) Abnormal situation handling: When abnormal situations occur during sewage treatment, the fuzzy control system can make judgments and handle them according to preset fuzzy rules and take corresponding measures, such as alarms and shutdowns, to ensure the safe operation of the system.

[0060] (5) Sewage treatment process modeling: By using neural networks to model the sewage treatment process, a soft measurement and prediction real-time control model can be established to better understand and predict the behavior of the system.

[0061] (6) Parameter optimization: Based on the predictive ability of neural networks, key parameters in the sewage treatment process, such as the amount of chemical agents added, are optimized to improve treatment effects and reduce costs.

[0062] (7) Fault diagnosis: Neural networks can learn the normal operating modes of sewage treatment equipment, thereby being able to detect and diagnose equipment faults.

[0063] (8) Intelligent Control: By combining neural networks with traditional control methods, a human-computer interaction intelligent model can be established to achieve intelligent control of the sewage treatment process and improve the system's adaptability and stability. Another aspect of this application discloses an intelligent lighting system, which is applied to a sewage treatment method based on a photocatalytic sewage treatment process as described above. The system includes a photoreaction module, an intelligent prediction and control module, and a human-computer interaction module. The intelligent prediction and control module receives data signals from the photoreaction module for processing, and the processed data is integrated into the human-computer interaction module. The human-computer interaction module sends instructions to control the operation of the photoreaction module based on the data analysis results.

[0064] In this embodiment, the photoreaction module includes a photoreactor, a light source 8 and a multi-series sensor. The photoreactor adopts a fixed reactor, and the TiO2 / CdS composite photocatalyst material is loaded on a porous ceramic carrier and filled into the fixed reactor; the light source 8 adopts an ultraviolet-visible composite light source, which is installed on the top of the fixed reactor to simulate the sunlight spectrum; the multi-series sensor includes a pH sensor, a temperature sensor and a flow meter equipped in the fixed reactor, and the multi-series sensor is connected to the intelligent prediction and control module by telecommunications.

[0065] Specifically, a photosensor is arranged in the photoreactor to ensure that there is no blind spot in the illumination.

[0066] In this embodiment, the intelligent prediction control module is based on a fuzzy control algorithm and a neural network control program. According to the collected current water quality parameters, current light intensity and catalyst status, the ANFIS prediction model predicts the pollutant degradation efficiency or target concentration under different light intensities based on the input. Based on the prediction results, the light intensity is dynamically adjusted through the optimization algorithm, the power or switching time of the light source 8 is adjusted, and the actual degradation effect is fed back through the sensor to correct the model error.

[0067] Specifically, the human-computer interaction module includes a touch screen control terminal, a mobile app remote monitoring system, an AR maintenance guidance system, a voice interaction interface, and an emotion manager. The touch screen control terminal is connected to the central control system and serves as the main operating interface for real-time control and display of the status of the light reaction module. The mobile app remote monitoring system is connected to the central control system via a network. Users can send commands through the app, which are transmitted to the central control system via the network and then forwarded to the intelligent prediction control module. The intelligent prediction control module then controls the status of the light reaction module. Simultaneously, the status information of the light reaction module is transmitted back to the app via the same path. The AR maintenance guidance system connects to the sensors and cameras of the light reaction module to obtain real-time images and data of the module. It also connects to the central control system via a network to obtain maintenance guidance information and virtual models, which are displayed to maintenance personnel through AR devices. The voice interaction interface connects to the voice processing unit of the intelligent prediction control module and interacts with the user through a microphone and speaker. The emotion manager is connected to the sensors, cameras, and user behavior analysis unit of the light reaction module. The emotion manager analyzes the user's facial expressions and voice intonation to determine the user's emotional state and adjusts the interaction mode based on the emotional state. Specifically, the human-machine interaction module consists of a touchscreen control terminal, a mobile app for remote monitoring, an AR maintenance guidance system, a voice interaction interface, and an emotion manager. This ensures efficient and stable human-machine operation. The application of human-machine interaction in photocatalytic wastewater treatment primarily improves the operational efficiency, monitoring accuracy, and management level of the wastewater treatment system through intelligent and visual technologies.

[0068] Example 1: An application case of an intelligent lighting system

[0069] In water treatment systems using photocatalysts (such as TiO2), light intensity, wavelength, and exposure time are key factors affecting photocatalytic efficiency. Intelligent light control dynamically adjusts light source parameters to achieve optimal degradation results while reducing energy consumption.

[0070] 1 Intelligent lighting control

[0071] 1.1 Intelligent Lighting Control Strategy

[0072] - Adaptive light intensity adjustment

[0073] - Automatically adjust UV-LED or mercury light intensity (10-100mW / cm2) based on real-time water quality data (such as COD, TOC, UV transmittance) 2 ).

[0074] - Example: When an elevated pollutant concentration is detected, the system automatically increases the light intensity by 20%.

[0075] -Pulse light mode

[0076] - Use intermittent irradiation (e.g., 10 seconds on / 5 seconds off) to reduce energy consumption and prevent overheating and passivation of the photocatalyst.

[0077] -Multi-wavelength collaboration

[0078] -Combining UV-A (365nm) and visible light (such as 450nm) light sources to broaden the photocatalyst excitation spectrum (suitable for modified TiO2).

[0079] 1.2 Hardware support for intelligent lighting

[0080] -Dimmable UV-LED array: light intensity is controlled by PWM (Pulse Width Modulation).

[0081] -Light uniformity detection: Place photosensors inside the reactor to ensure there are no blind spots.

[0082] - Light source 8 life prediction: Based on the cumulative working time and light decay curve, early warning replacement is required.

[0083] 2. ANFIS real-time prediction

[0084] 2.1 Integration of prediction model and control system

[0085] (1) Closed-loop control architecture

[0086] -Prediction → Decision → Execution Process:

[0087] - Real-time input: The sensor collects current water quality parameters (such as pollutant concentration, pH, temperature, etc.), current light intensity, and catalyst status.

[0088] -ANFIS prediction: The model predicts the pollutant degradation efficiency (or target concentration) at different light intensities based on the input.

[0089] -Control decision: Based on the prediction results, the light intensity is dynamically adjusted through optimization algorithms (such as PID, fuzzy logic or genetic algorithm).

[0090] -Execution feedback: Adjust the power or switching time of the LED / UV light source, and feedback the actual degradation effect through sensors to correct model errors.

[0091] (2) Control strategy

[0092] -Setpoint tracking:

[0093] If the target is a fixed pollutant removal rate (such as 95%), ANFIS predicts the treatment effect under different light intensities and selects the light intensity value closest to the target.

[0094] -Dynamic Optimization:

[0095] Under the energy consumption constraint, the standard processing is achieved with the minimum light intensity (such as predicting the relationship between light intensity and energy consumption through ANFIS to find the optimal solution).

[0096] 2.2 Implementation of key technologies

[0097] (1) Real-time prediction and response

[0098] - High-frequency data updates: Fast-response sensors (such as online COD detectors and UV intensity sensors) need to be deployed.

[0099] -Lightweight model: Simplify the ANFIS structure (reduce the number of fuzzy rules) to meet real-time requirements, or use edge computing devices for deployment.

[0100] (2) Anti-interference design

[0101] -Multi-parameter coupling: Light intensity control needs to consider the influence of other variables (such as catalyst activity and water flow rate), and ANFIS input needs to include these interference factors.

[0102] -Feedback compensation: When the actual degradation effect deviates significantly from the predicted one, the model parameters are triggered to be updated online (e.g., incremental learning).

[0103] (3) Energy-saving optimization

[0104] -Light intensity-energy consumption model: Energy consumption is used as an additional output of ANFIS to achieve multi-objective optimization of "treatment effect-energy consumption" in control.

[0105] -Time-of-day control: Based on changes in pollutant load (such as differences in water inflow during the day and night), ANFIS predicts the optimal light intensity at different times.

[0106] 2.3 Practical Application Cases

[0107] In a dyeing wastewater treatment project, ANFIS predicts the required UV light intensity based on the influent chromaticity and dynamically adjusts the number of LED arrays on, saving 30% energy while maintaining a decolorization rate of >90%.

[0108] - Combined with a photovoltaic power generation system, ANFIS automatically reduces the processing flow rate to match the available light intensity when sunlight is insufficient, maintaining processing stability.

[0109] 2.4 Implementation steps

[0110] (1) Offline stage:

[0111] -Train the ANFIS model using historical data to verify prediction accuracy.

[0112] -Design control logic (such as light intensity adjustment step size, constraints).

[0113] (2) Online stage:

[0114] -Deployment model and PLC / industrial computer linkage to achieve real-time control.

[0115] - Set safety thresholds (such as maximum allowed light intensity).

[0116] In short, ANFIS not only predicts the effectiveness of photocatalytic water treatment but also dynamically adjusts light intensity through closed-loop control, significantly improving energy efficiency. The key lies in: a high-precision real-time prediction model; seamless integration with actuators (such as adjustable light sources and inverters); and a design with anti-interference and adaptive capabilities.

[0117] Example 2: Working example of human-computer interaction module in sewage treatment system

[0118] (1) Real-time monitoring and data visualization

[0119] Sensor network: Integrates pH, turbidity, COD (chemical oxygen demand), light intensity and other sensors to collect water quality and photocatalyst reaction data in real time.

[0120] Interactive dashboard: Displays key parameters (such as pollutant degradation efficiency and reactor status) through a graphical interface (such as web or mobile terminal), allowing operators to quickly understand the system operation status.

[0121] Early warning system: Abnormal data (such as catalyst deactivation, insufficient light) automatically triggers an alarm and prompts the user through color, sound or message push.

[0122] (2) Intelligent control system

[0123] Automated adjustment: Based on machine learning or rule engines, parameters such as light intensity, sewage flow, and catalyst dosage are automatically adjusted to optimize reaction conditions.

[0124] Remote control: Remote operation (such as starting and stopping equipment, switching modes) can be achieved through touch screen, voice commands or gesture control, reducing manual intervention.

[0125] Adaptive learning: The system records historical data and learns operating habits to gradually improve the accuracy of control strategies.

[0126] (3) Virtual simulation and training

[0127] 3D modeling: Build a virtual model of the photocatalytic reactor 3 to simulate the sewage treatment effects under different working conditions and assist in process optimization.

[0128] AR / VR training: Operators are trained through augmented reality (AR) or virtual reality (VR) technology, visually demonstrating equipment disassembly and assembly, as well as troubleshooting procedures.

[0129] (4) Maintenance and fault diagnosis

[0130] Fault diagnosis assistant: Combined with expert systems or AI algorithms, it analyzes the causes of equipment abnormalities and provides maintenance suggestions (such as UV lamp replacement and catalyst regeneration).

[0131] Maintenance reminder: proactively push maintenance plans based on device usage time or performance degradation predictions.

[0132] (5) User-friendly design

[0133] Multimodal interaction: supports multiple operation modes such as voice, touch, gesture, etc. to adapt to the needs of different users (such as on-site workers or managers).

[0134] Multi-language support: Provide localized interfaces in multinational projects to reduce language barriers.

[0135] (6) Data sharing and collaboration

[0136] Cloud collaboration: Upload processed data to the cloud platform for shared analysis by cross-departmental or cross-regional teams.

[0137] Blockchain traceability: records key operations and data modification history to ensure transparency and traceability.

[0138] Through human-computer interaction technology, the photocatalytic sewage treatment system has been significantly improved in automation, intelligence and user experience, promoting the development of environmental protection technology towards a more efficient and easier-to-use direction.

[0139] Example 3: Working example of human-computer interaction of intelligent lighting system in sewage treatment system

[0140] The intelligent lighting system needs to interact efficiently with operators to achieve a hybrid control mode of "AI decision-making + manual correction".

[0141] 1Interactive interface design

[0142] Control panel function module

[0143]

[0144] 2. Mobile APP expansion

[0145] - Remotely monitor lighting status and receive alarm notifications (such as "The reactor is not evenly illuminated, please check the LED array").

[0146] -Provide historical data export function to facilitate energy efficiency analysis.

[0147] 3 Typical interaction scenarios

[0148] (1) Exception handling

[0149] -Problem: The sensor detects a sudden drop in light intensity in an area.

[0150] -Interaction process:

[0151] -System alarm and locate fault LED position;

[0152] - The interface displays AR guidance (such as highlighting damaged LED modules);

[0153] -After the operator confirms, the system starts the backup light source 8 and generates a maintenance work order.

[0154] (2) Parameter optimization

[0155] -Problem: The concentration of pollutants in the influent fluctuates greatly.

[0156] -Interaction process:

[0157] -The AI ​​model predicts that the light intensity needs to be increased to 90mW / cm 2 ;

[0158] -The operator reviews the prediction basis (e.g., "the current COD degradation rate is less than the target value") and chooses to accept or make fine adjustments;

[0159] -The system records the results of manual intervention for iterative model learning.

[0160] 4. Collaborative control of intelligent lighting and systems

[0161] -Linked with oxidant dosage: When light intensity increases, the amount of H2O2 added is automatically reduced to save costs.

[0162] -Match with hydraulic retention time (HRT): automatically extend HRT under low light intensity to ensure degradation effect.

[0163] 5. Operator performance control

[0164] - Detecting large mood swings and receiving intervention or adjustment

[0165] -Receive regular learning and training.

[0166] Example 4 Human-computer interaction process of photocatalytic sewage treatment method

[0167] Interaction Paradigms and Technologies

[0168] (1) Input method

[0169] - Traditional input: keyboard, mouse, touch screen.

[0170] -Emerging technologies: voice recognition (such as Siri), gesture control (such as VR), eye tracking, and brain-computer interfaces.

[0171] (2) Output method

[0172] -Visual output: graphical user interface (GUI), augmented reality (AR).

[0173] -Multimodal feedback: tactile (vibration), sound (prompt tone), and speech synthesis.

[0174] (3) Interaction Model

[0175] -WIMP paradigm: Window, Icon, Menu, Pointer.

[0176] -Natural User Interface (NUI): such as direct operation of touch screen and voice interaction.

[0177] Figure 2 The diagram of the integrated human-computer interaction communication is as follows: a multimodal information perception system 10 includes, but is not limited to, a Bluetooth perception module, a voice communication module, a camera module, an operating parameter sensor module, etc.; a relay module 20 is used for various operation controls of the automatic machine, such as speakers and switches; an information fusion subsystem 30 adaptively processes various data through machine learning and provides corresponding human-computer interaction.

[0178] In the photocatalytic wastewater treatment system, the integration of computer technology realizes the collaborative work of data collection, processing, control and interaction through a multi-level architecture.

[0179] The following is an overview of the integration of key technologies and their architecture:

[0180] (1) Hardware-layer fusion: perception and execution

[0181] -Embedded system: The photocatalytic reactor 3 has a built-in microcontroller (such as STM32, Arduino) or PLC (Programmable Logic Controller) that is directly connected to sensors (pH, turbidity, UV intensity, etc.) and actuators (water pumps, UV lamps, valves).

[0182] -Edge computing: Perform real-time data preprocessing (such as filtering and normalization) on the device side to reduce cloud transmission latency.

[0183] -Internet of Things (IoT) nodes:

[0184] - Through low-power wide area network protocols such as LoRa and NB-IoT, distributed device data is aggregated to the gateway to support remote monitoring.

[0185] (2) Data layer integration: unified management and analysis

[0186] -Data Lake / Cloud Platform:

[0187] -Use a time series database (such as InfluxDB) to store sensor data and a relational database (MySQL) to manage device information to form a unified data pool.

[0188] -ETL tools (such as Apache Kafka): clean and forward data streams to analysis modules in real time.

[0189] -AI model integration:

[0190] -Trained models (such as LSTM prediction of pollutant degradation efficiency) are deployed via Docker containers, receiving real-time data and outputting control recommendations.

[0191] (3) Control layer integration: intelligent decision-making

[0192] -Feedback control loop:

[0193] - Classic control: PID algorithm regulates UV lamp power or sewage flow.

[0194] -Intelligent control: Reinforcement learning (RL) dynamically optimizes reaction conditions (such as the combination of catalyst dosage and illumination time).

[0195] -Digital Twin:

[0196] -Simulate the control strategy effect in the virtual model and then send it to the physical equipment after verification.

[0197] (4) Interaction layer fusion: multimodal interface

[0198] -Front-end and back-end separation architecture:

[0199] -Front-end: Vue / React builds a visual dashboard, WebGL displays 3D device models, and supports drag-and-drop parameter settings.

[0200] -Backend: Spring Boot / Django provides RESTful API, processes user instructions and forwards them to the control layer.

[0201] -Multi-terminal adaptation:

[0202] - Responsive design is compatible with PCs, tablets, and mobile phones; voice interaction is connected to the control system through ASR (automatic speech recognition) technology.

[0203] (5) Collaboration layer integration: distributed collaboration

[0204] -Microservice architecture:

[0205] - Split monitoring, control, maintenance and other functions into independent services (such as Kubernetes cluster management) to improve system resilience.

[0206] -Blockchain:

[0207] -Hyperledger Fabric records operation logs to ensure that data cannot be tampered with during multi-party collaboration.

[0208] (6) Example process of technology integration

[0209] -Data flow:

[0210] Sensor → Edge node (filtering) → Cloud (AI analysis) → Control instructions → PLC execution.

[0211] -User operation:

[0212] Mobile phone APP voice command → cloud NLP analysis → control service → adjust reactor parameters.

[0213] (7) Communications and Security

[0214] - Compatible with heterogeneous protocols: Adopts OPC UA or MQTT unified communication protocols.

[0215] - Real-time guarantee: Edge computing reduces latency, and 5G transmits key instructions.

[0216] - Security: TLS encrypts data transmission, and RBAC (role-based access control) limits operation permissions.

[0217] Through the above integration, computer technology upgrades photocatalytic wastewater treatment from isolated equipment to an intelligent, adaptive, and user-friendly ecosystem.

[0218] Example 5 Photocatalytic Wastewater Treatment System for Operator Emotion Recognition

[0219] The recognition model consists of the perception layer and the application layer.

[0220] The perception layer uses advanced video or high-precision digital cameras to capture images of the human body (such as facial features or body movements) and transmits the data to a database. The ANFIS model processes and recognizes the data images to generate emotional results.

[0221] The application layer is responsible for judging whether the emotional changes are drastic, starting the automation and language system, and reminding users to make decisions.

[0222] In summary, the technical solution of this application has the following beneficial effects:

[0223] 1. The sewage treatment system based on the photocatalytic sewage treatment process of this application combines artificial intelligence prediction models and human-computer interaction methods to achieve efficient and intelligent water purification operation management.

[0224] 2. The sewage treatment based on the photocatalytic sewage treatment process of this application can achieve the following effects through an intelligent control system: energy efficiency optimization: dynamic lighting strategy can reduce energy consumption by 20%-40%; extend equipment life: avoid light source overload and reduce photocatalyst deactivation; operation-friendly: through a visual interface and AR assistance, reduce dependence on professionals.

[0225] 3. The sewage treatment system of this application deploys fast-response sensors for real-time prediction and response; simplifies the ANFIS structure (reduces the number of fuzzy rules) to meet real-time requirements, or uses edge computing equipment for deployment, lightweighting the system model; adopts the ANFIS system, emphasizes the influence of multiple variables, and realizes multi-parameter coupling. When the actual degradation effect deviates greatly from the prediction, the model parameters are triggered to be updated online, and feedback compensation can be performed to achieve anti-interference design; energy consumption is used as an additional output of ANFIS, and "treatment effect-energy consumption" multi-objective optimization is realized in the control to achieve energy-saving optimization.

[0226] The above are only preferred embodiments of the present invention and do not limit the implementation mode and protection scope of the present invention. For those skilled in the art, it should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the description and illustrations of the present invention should be included in the protection scope of the present invention.

Claims

1. A sewage treatment method based on a photocatalytic sewage treatment process, wherein the sewage treatment system comprises a regulating tank 1, a pre-reaction tank, a photocatalytic reactor, a clear water tank and a central control system, characterized in that: The sewage treatment method comprises the following steps: S1: Pretreatment: Wastewater is injected into the regulating tank, and after coarse filtration and pH adjustment, it flows into the pre-reaction tank to remove large particles of impurities; S2: Photocatalytic reaction: The wastewater from the pre-reaction tank is introduced into the photocatalytic reactor of the reaction tank, and a photocatalytic reaction is carried out under light conditions to degrade organic pollutants, remove heavy metal ions, and inactivate microorganisms. The influent characteristics and effluent efficiency characteristics of the influent into the above-mentioned reaction tank are measured in a composite manner to determine the current dynamic values; S3: Filter and disinfect the treated sewage to ensure that the effluent water quality meets the standards; The central control system is connected to the sensors and actuators provided in the regulating tank, pre-reaction tank, photocatalytic reactor, and clear water tank via telecommunications. The central control system collects measurement parameters of the sensors and operating parameters of the actuators, compares the measured parameters through a program in the computer, calculates a suitable target value, compares the calculated target value with the received corresponding parameters, and processes the calculated target value into analog and digital control output values. The obtained control output values ​​are used to regulate the processes of pretreatment, photocatalytic reaction, and post-treatment. The program obtains the above-mentioned appropriate target value by adopting the control program of the ANFIS system.

2. The sewage treatment method based on the photocatalytic sewage treatment process according to claim 1, characterized in that: Based on the ANFIS system, according to the measurement parameters of the pretreatment stage, including sewage inflow flow, pH value, BOD and COD concentration of injection water, the execution parameters of the pretreatment stage, including pump speed, valve opening, and reagent dosage, are adjusted.

3. The sewage treatment method based on the photocatalytic sewage treatment process according to claim 1, characterized in that: Based on the ANFIS system, the execution parameters of the photocatalytic reaction stage, such as UV-LED or mercury light intensity and reagent dosage, are adjusted according to the measured parameters of the photocatalytic reaction stage, including COD, TOC, UV transmittance, catalyst activity, and water flow rate.

4. The sewage treatment method based on the photocatalytic sewage treatment process according to claim 1, characterized in that: The central control system includes a central computer and a PLC controller communicatively connected to the central computer, and the PLC controller is electrically connected to sensors and actuators provided in the regulating tank, pre-reaction tank, photocatalyst reactor and clear water tank.

5. The sewage treatment method based on the photocatalytic sewage treatment process according to claim 1, characterized in that: The ANFIS system includes an ANFIS neural network prediction model, a fuzzy controller and an optimization processor, wherein the ANFIS neural network prediction model takes the inlet flow rate, inlet COD and illumination as input vectors, and the output vector is the outlet COD. The fuzzy controller takes the deviation between the predicted value and the expected value of the outlet COD of the ANFIS neural network prediction model and the deviation change rate as input, and the illumination correction amount as control output. The current light intensity is corrected through the control output, thereby completing the automatic adjustment of the light intensity.

6. An intelligent lighting system, applied to a sewage treatment method based on a photocatalytic sewage treatment process as claimed in any one of claims 1 to 5, characterized in that: It includes a light reaction module, an intelligent prediction and control module and a human-computer interaction module. The intelligent prediction and control module receives the data signal of the light reaction module for processing, and the processed data is integrated in the human-computer interaction module. The human-computer interaction module sends instructions to control the operation of the light reaction module according to the data analysis results.

7. The intelligent lighting system according to claim 6, characterized in that: The photoreaction module includes a photoreactor, a light source and a multi-series sensor. The photoreactor adopts a fixed reactor, and the TiO2 / CdS composite photocatalyst material is loaded on a porous ceramic carrier and filled into the fixed reactor; the light source adopts an ultraviolet-visible composite light source, which is installed on the top of the fixed reactor to simulate the sunlight spectrum; the multi-series sensor includes a pH sensor, a temperature sensor and a flow meter equipped with the fixed reactor, and the multi-series sensor is electrically connected to the intelligent prediction and control module.

8. The intelligent lighting system according to claim 7, characterized in that: A photosensitive sensor is arranged in the photoreactor to ensure that there is no blind spot in the illumination.

9. The intelligent lighting system according to claim 7, characterized in that: The intelligent prediction and control module is based on a fuzzy control algorithm and a neural network control program. Based on the collected current water quality parameters, current light intensity and catalyst status, the ANFIS prediction model predicts the pollutant degradation efficiency or target concentration under different light intensities based on the input. Based on the prediction results, the light intensity is dynamically adjusted through an optimization algorithm, and the power or switching time of the light source is adjusted. The actual degradation effect is fed back through sensors to correct model errors.

10. The intelligent lighting system according to claim 6, characterized in that: The human-computer interaction module includes a touch screen control terminal, a mobile APP remote monitoring system, an AR maintenance guidance system, a voice interaction interface and an emotion manager. The touch screen control terminal is connected to the central control system and serves as the main operation interface for real-time control and display of the status of the light reaction module; The mobile APP remote monitoring system is connected to the central control system via the network. Users can send instructions through the APP. These instructions are transmitted to the central control system via the network, and then forwarded by the central control system to the intelligent prediction control module. The intelligent prediction control module then controls the status of the light reaction module. At the same time, the status information of the light reaction module is transmitted back to the APP via the same path. The AR maintenance guidance system is connected to the sensor and camera of the light reaction module to obtain real-time images and data of the module, and is connected to the central control system via the network to obtain maintenance guidance information and virtual models, which are displayed to maintenance personnel through AR equipment; The voice interaction interface is connected to the voice processing unit of the intelligent prediction control module and performs voice interaction with the user through a microphone and a speaker; The emotion manager is connected to the sensor and camera of the light reaction module and the user behavior analysis unit of the intelligent prediction and control module. The emotion manager judges the user's emotional state by analyzing the user's facial expressions and voice intonation, and the device adjusts the interaction mode according to the emotional state.

Images