Modularized ultrapure water preparation and on-line monitoring system

Abstract

The invention discloses a modularized ultrapure water preparation and on-line monitoring system, which comprises a pretreatment module, a purification module, a monitoring module and a control module, and is characterized in that a dynamic monitoring end is arranged, and when ultrapure water preparation is monitored in real time, a multi-parameter fusion detection mechanism and self-adaptive threshold analysis are integrated; meanwhile, the abnormal state of the water quality is dynamically judged based on the conductivity, TOC and particulate matter data collected in real time, the sudden change risk of the water quality output by the purification module can be recognized in real time, error accumulation caused by traditional single-point monitoring is avoided, and therefore the purity detection reliability of the ultrapure water is improved; and the reverse osmosis unit and the electrode deionization unit of the purification module can immediately respond to the effluent parameter change of the pretreatment module, so that the hidden dangers of membrane blockage and recovery rate reduction caused by water quality fluctuation are reduced, and the energy consumption redundancy is reduced while the water quality stability is ensured.

Description

Technical Field

[0001] This invention relates to the field of water treatment technology, specifically to a modular ultrapure water preparation and online monitoring system. Background Technology

[0002] Water treatment refers to the physical and chemical measures taken to ensure that water quality meets certain usage standards.

[0003] Currently, because the ultrapure water preparation process involves multiple treatment units, the system relies on single-point sensor detection with fixed thresholds when conducting real-time water quality monitoring. When the raw water quality fluctuates or the pollutant concentration changes abruptly, the purification parameters cannot be dynamically adjusted to compensate for the error, which will lead to excessive conductivity of the produced water and an increase in total organic carbon (TOC).

[0004] Therefore, a modular ultrapure water preparation and online monitoring system is proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a modular ultrapure water preparation and online monitoring system, which solves the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a modular ultrapure water preparation and online monitoring system, the system comprising a pretreatment module, a purification module, a monitoring module, and a control module; The pretreatment module includes an inlet unit, a filtration unit, and a softening unit, which perform preliminary treatment on the raw water and generate pretreated water quality data. The purification module includes a reverse osmosis unit, an electrode deionization unit, and an ultraviolet sterilization unit, which are connected to the pretreatment module to perform deep purification treatment on the pretreated water and generate ultrapure water. The monitoring module includes a water quality sensing unit, a data acquisition unit, and an analysis unit. It is connected to the purification module and collects the conductivity, TOC, and particulate matter parameters of ultrapure water in real time. It also performs water quality anomaly detection based on artificial intelligence algorithms and generates real-time monitoring data. The control module includes an algorithm matching unit and an execution unit, which are connected to the monitoring module and the purification module. Based on the real-time monitoring data, the module calls an optimization algorithm to dynamically adjust the purification parameters and outputs control commands. The system uses modular interfaces to enable detachable connections between modules, supporting flexible expansion.

[0007] Preferably, the water inlet unit of the pretreatment module includes a flow sensor and a pressure regulator to monitor the inlet flow and pressure in real time. The filtration unit adopts a multi-stage filter structure, including a sedimentation filter, an activated carbon filter, and a microfiltration membrane to remove suspended solids and organic matter. The softening unit includes an ion exchange resin tank and a regeneration controller to reduce water hardness and generate softened water data.

[0008] Preferably, the reverse osmosis unit of the purification module includes a high-pressure pump, an RO membrane module, and a concentrate recovery device, which separates dissolved salts and generates reverse osmosis water. The electrode deionization unit includes an ion exchange membrane stack and a DC power supply, which are connected to the reverse osmosis unit to remove ionic impurities and generate deionized water data. The ultraviolet sterilization unit includes a UV lamp and a radiation sensor, which sterilize the water and output ultraviolet intensity data.

[0009] Preferably, the water quality sensing unit of the monitoring module includes a conductivity sensor, a TOC analyzer, and a laser particle counter, installed at the output end of the purification module, to measure the conductivity, total organic carbon, and particulate matter concentration of ultrapure water in real time. The data acquisition unit uses a multi-channel ADC converter to convert the sensing signals into digital data. The analysis unit embeds a machine learning model to compare historical data and detect water quality anomalies. The water quality anomaly detection uses a normalized comprehensive index formula: ; In the formula, QI is the water quality anomaly index, and an anomaly alarm is triggered when QI > 1.0; C represents real-time conductivity, in μS / cm, normalized reference. ; T represents real-time total organic carbon (TOC), in ppb, normalized basis. ; P represents the real-time particulate matter concentration, in units of particles / mL, normalized baseline. ; These are the weighting coefficients. Assignment based on parameter importance; Thresholds: The QI alarm threshold is 1.0, and the error tolerance threshold is less than 0.05.

[0010] Preferably, the algorithm matching unit of the control module includes a storage subunit and a matching processor. The storage subunit pre-stores various water quality optimization algorithms, including neural network algorithms, PID control algorithms and fuzzy logic algorithms. The matching processor performs algorithm matching based on the real-time monitoring data and preset thresholds to generate the optimal control strategy.

[0011] Preferably, the execution unit of the control module includes a motor driver and a valve controller, connected to the purification module, and adjusts the speed of the high-pressure pump and the voltage of the ion exchange membrane stack according to the optimal control strategy, wherein the PID control output uses a normalized formula: ; In the formula, For normalized control output, range [0,1], corresponding to the percentage of actual voltage and speed; To normalize the error, , To account for the error of the measured parameters, ; The normalized PID gain coefficient, Based on system optimization; Threshold: Error triggers adjustment threshold. Control the output saturation threshold .

[0012] Preferably, the system further includes an alarm module connected to the monitoring module, comprising an audible and visual alarm and a network communication unit. The audible and visual alarm triggers an alarm when water quality exceeds the standard, and the network communication unit supports 4G / 5G and IoT protocols to remotely transmit monitoring data to a cloud platform.

[0013] Preferably, the system further includes an energy management module integrated into the control module, comprising a solar converter and a battery pack. The solar converter collects renewable energy, and the battery pack stores electrical energy and supplies power to the ultraviolet sterilization unit and the monitoring module, achieving an energy self-sufficiency rate of more than 80%.

[0014] Preferably, the modular interface adopts standardized quick-connect couplings and O-rings, supports hot-swappable functionality, and ensures the sealing performance and rapid replacement of each module under pressure less than 1.0 MPa.

[0015] Preferably, the system has an external user interaction module, including a touch screen and a mobile APP. The touch screen displays real-time water quality parameters and system status, and the mobile APP connects to the control module through an API interface, supporting remote parameter setting and fault diagnosis.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. By setting up a dynamic monitoring terminal, when monitoring the preparation of ultrapure water in real time, a multi-parameter fusion detection mechanism and adaptive threshold analysis are integrated to ensure the monitoring accuracy under different water quality fluctuations. At the same time, based on the real-time collected conductivity, TOC and particulate matter data, the abnormal state of water quality is dynamically determined, which can identify the risk of sudden changes in the output water quality of the purification module in real time, avoid the error accumulation caused by traditional single-point monitoring, and thus improve the reliability of ultrapure water purity detection. 2. By setting up an algorithm coordination terminal, during system operation, the logical relationship between the algorithm library and real-time monitoring data is matched and optimized, and the optimal control strategy suitable for the current water quality is automatically selected. This enables the reverse osmosis unit and electrode deionization unit of the purification module to respond to changes in the effluent parameters of the pretreatment module in real time, reducing the risk of membrane blockage and decreased recovery rate caused by water quality fluctuations. At the same time, the operating intensity of the ultraviolet sterilization unit is optimized to ensure water quality stability while reducing energy consumption redundancy. 3. By setting up hot-swappable adapters, the system can be expanded and maintained by relying on standardized interfaces to achieve non-destructive and rapid disassembly and assembly of each module. This ensures that the sealing integrity is maintained under pressure fluctuations. Combined with the renewable energy conversion mechanism of the energy management module, the dependence on external power supply is reduced, forming a modular system with the dual advantages of deployment flexibility and operational sustainability, providing a highly adaptable solution for complex industrial scenarios. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall system architecture of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 The modular ultrapure water preparation and online monitoring system includes a pretreatment module, a purification module, a monitoring module, and a control module. The pretreatment module includes an inlet unit, a filtration unit, and a softening unit, which perform preliminary treatment on the raw water and generate pretreated water quality data. The purification module includes a reverse osmosis unit, an electrode deionization unit, and an ultraviolet sterilization unit, which are connected to the pretreatment module to perform deep purification treatment on the pretreated water and generate ultrapure water. The monitoring module includes a water quality sensing unit, a data acquisition unit, and an analysis unit. It is connected to the purification module to collect the conductivity, TOC, and particulate matter parameters of ultrapure water in real time, and to detect water quality anomalies based on artificial intelligence algorithms to generate real-time monitoring data. The control module includes an algorithm matching unit and an execution unit, which connects the monitoring module and the purification module. Based on real-time monitoring data, it calls the optimization algorithm to dynamically adjust the purification parameters and outputs control commands. The system uses modular interfaces to enable detachable connections between modules, supporting flexible expansion. The inlet unit of the pretreatment module includes a flow sensor and a pressure regulator to monitor the inlet flow and pressure in real time. The filtration unit adopts a multi-stage filter structure, including a sedimentation filter, an activated carbon filter, and a microfiltration membrane to remove suspended solids and organic matter. The softening unit includes an ion exchange resin tank and a regeneration controller to reduce water hardness and generate softened water data. The reverse osmosis unit of the purification module includes a high-pressure pump, an RO membrane module and a concentrate recovery device, which separates dissolved salts and generates reverse osmosis water. The electrode deionization unit includes an ion exchange membrane stack and a DC power supply, which is connected to the reverse osmosis unit to remove ionic impurities and generate deionized water data. The ultraviolet sterilization unit includes a UV lamp and a radiation sensor, which sterilizes the water and outputs ultraviolet intensity data. The water quality sensing unit of the monitoring module includes a conductivity sensor, a TOC analyzer, and a laser particle counter, installed at the output of the purification module. It measures the conductivity, total organic carbon, and particulate matter concentration of the ultrapure water in real time. The data acquisition unit uses a multi-channel ADC converter to convert the sensor signals into digital data. The analysis unit embeds a machine learning model to compare historical data and detect water quality anomalies. The water quality anomaly detection uses a normalized comprehensive index formula: ; In the formula, QI is the water quality anomaly index, and an anomaly alarm is triggered when QI > 1.0; C represents real-time conductivity, in μS / cm, normalized reference. ; T represents real-time total organic carbon (TOC), in ppb, normalized basis. ; P represents the real-time particulate matter concentration, in units of particles / mL, normalized baseline. ; These are the weighting coefficients. Assignment based on parameter importance; Thresholds: The QI alarm threshold is 1.0, and the error tolerance threshold is less than 0.05; The algorithm matching unit of the control module includes a storage subunit and a matching processor. The storage subunit pre-stores a variety of water quality optimization algorithms, including neural network algorithms, PID control algorithms and fuzzy logic algorithms. The matching processor performs algorithm matching based on real-time monitoring data and preset thresholds to generate the optimal control strategy. The control module's execution unit includes a motor driver and a valve controller, connected to the purification module. It adjusts the high-pressure pump's speed and the ion exchange membrane stack's voltage according to the optimal control strategy. The PID control output uses a normalized formula. ; In the formula, For normalized control output, range [0,1], corresponding to the percentage of actual voltage and speed; To normalize the error, , To account for the error of the measured parameters, ; The normalized PID gain coefficient, Based on system optimization; Threshold: Error triggers adjustment threshold. Control the output saturation threshold ; The system also includes an alarm module and a monitoring module, including an audible and visual alarm and a network communication unit. The audible and visual alarm triggers an alarm when water quality exceeds the standard. The network communication unit supports 4G / 5G and IoT protocols to remotely transmit monitoring data to the cloud platform. The system further includes an energy management module integrated into the control module, comprising a solar converter and a battery pack. The solar converter collects renewable energy, and the battery pack stores electrical energy and powers the ultraviolet sterilization unit and monitoring module, achieving an energy self-sufficiency rate of more than 80%. The modular interface uses standardized quick-connect couplings and O-rings, supports hot-swappable functionality, and ensures the sealing performance and quick replacement of each module under pressure less than 1.0MPa. The system has an external user interaction module, including a touch screen and a mobile APP. The touch screen displays real-time water quality parameters and system status, while the mobile APP connects to the control module via an API interface, supporting remote parameter setting and fault diagnosis.

[0020] Example 1: Ultrapure Water Supply Scenario in a Semiconductor Factory The system's raw water is municipal tap water with an initial conductivity of approximately 500 μS / cm and a TOC of less than 100 ppb. The target product water must meet a purity standard of 18.2 MΩ·cm.

[0021] Pretreatment module operation: Raw water first enters the inlet unit of the pretreatment module. A flow sensor detects the inlet flow rate in real time and transmits the data to the control module. A pressure regulator stabilizes the pressure at 0.4 MPa. The water then flows through a multi-stage filter structure. A sedimentation filter intercepts suspended particles larger than 50 μm, an activated carbon filter adsorbs organic matter and residual chlorine, and a microfiltration membrane further removes colloidal impurities. In the softening unit, an ion exchange resin tank replaces calcium and magnesium ions with sodium ions, reducing the water hardness to below 1 mg / L. The regeneration controller automatically triggers the resin regeneration program based on the softened water volume. The pretreated water quality data generated during this stage is uploaded to the monitoring module in real time.

[0022] The purification and monitoring process is coordinated and controlled as follows: After pretreatment, the water enters the reverse osmosis unit of the purification module. The high-pressure pump dynamically adjusts its speed according to the water quality data, pushing the water flow through the RO membrane module to generate reverse osmosis water with a conductivity of less than 5 μS / cm. The concentrate recovery device controls the wastewater rate to within 30%. The reverse osmosis water flows into the electrode deionization unit. The DC power supply applies a gradient voltage to drive the ion exchange membrane stack for deep desalination, producing primary ultrapure water with a conductivity of less than 0.1 μS / cm. The ultraviolet sterilization unit uses 254nm UV lamps to irradiate and sterilize the water flow, and the radiation sensor calibrates the ultraviolet intensity in real time.

[0023] Meanwhile, the monitoring module deploys a water quality sensing unit at the outlet of the purification module: the conductivity sensor collects data every 10 seconds, the TOC analyzer detects organic carbon content through ultraviolet oxidation, the laser particle counter counts particles larger than 0.1μm, the data acquisition unit converts multiple signals into digital quantities, and the analysis unit calls a machine learning model to compare with historical data. When the conductivity fluctuation exceeds the baseline value of 0.05μS / cm in real time, it determines that the water quality is abnormal and triggers an alarm.

[0024] Dynamic optimization and system expansion: After receiving the alarm signal, the algorithm matching unit of the control module calls the PID control algorithm from the storage subunit. The matching processor calculates the correlation between the conductivity deviation value and the preset threshold, and generates the optimal control strategy of increasing the high-pressure pump speed by 15% and increasing the ion exchange membrane stack voltage by 10V. The motor driver of the execution unit adjusts the speed of the high-pressure pump, and the valve controller synchronously increases the membrane stack voltage, so that the water quality recovers to stability within 120 seconds.

[0025] During system operation, the network communication unit of the alarm module uploads real-time data to the cloud platform. When the instantaneous TOC value exceeds 3ppb, the audible and visual alarm activates a level three alarm. The solar converter of the energy management module supplies power through the photovoltaic panels on the factory roof, and the battery pack prioritizes providing 48V DC power to the ultraviolet sterilization unit, reducing dependence on the external power grid. Maintenance personnel can view membrane blockage warnings through the touch screen of the user interaction module and replace the pretreatment filter module within 10 minutes using quick-connect couplings, ensuring uninterrupted water production throughout the process.

[0026] Implementation Results: After 72 hours of continuous operation testing, the system produced ultrapure water with a stable conductivity of 0.055 μS / cm, a TOC of less than 2 ppb, and a particulate matter concentration of less than 5 particles / mL, meeting the requirements for semiconductor wafer cleaning. The modular design improved the efficiency of reverse osmosis unit replacement, and the dynamic optimization of the algorithm reduced energy consumption.

[0027] Example 2: Nuclear Power Plant Auxiliary Cooling Water Supply Scenario The system's raw water is desalinated water after seawater pretreatment, with an initial chloride ion concentration of less than 200 ppm and a boron concentration of greater than 4.5 ppm. The target product water must meet the nuclear-grade ultrapure water standard, with a boron concentration of less than 0.01 ppb and a radioactivity of less than 0.1 Bq / L.

[0028] Pretreatment Enhancement Process: Desalinated seawater first enters the pressurization unit of the pretreatment module, where colloidal impurities are intercepted by a ceramic membrane filter, while organic pollutants are deeply removed by an activated carbon adsorption column. The softening unit uses a two-stage ion exchange process: a first-stage strong acid resin removes calcium and magnesium ions, and a second-stage special resin targets and adsorbs boron. The regeneration controller automatically switches the regeneration cycle according to the boron concentration of the influent. The real-time turbidity and boron concentration data are transmitted to the central control console via the Industrial Internet of Things. When the boron adsorption rate is detected to be less than 95%, a resin replacement warning is triggered.

[0029] Radiation-resistant purification process: The purification module is specially reinforced for the needs of the nuclear industry. The reverse osmosis unit uses a radiation-resistant composite RO membrane, achieving a chloride ion removal rate of more than 99.8% under an operating pressure of 1.8MPa. The electrode deionization unit is equipped with a radiation shield. The ion exchange membrane stack deeply removes trace boron in DC pulse mode. The ultraviolet sterilization unit is upgraded to a dual-wavelength high-power lamp, which simultaneously decomposes radioactive organic matter and sterilizes. All sealing components are made of fluororubber and can withstand a cumulative radiation dose of 100kGy.

[0030] Nuclear-grade monitoring and emergency control: The monitoring module is equipped with a neutron activation analyzer to detect trace radionuclides in the water in real time. When the data analysis unit detects abnormal fluctuations in radioactivity, it immediately activates a three-level response mechanism. The algorithm matching unit calls the emergency purification protocol, closes the conventional purification channel, and switches to the backup electrochemical deionization circuit. The execution unit switches the reverse osmosis concentrate discharge valve to the radioactive wastewater collection tank. The alarm module sends a real-time pollution spectrum to the nuclear safety supervision platform, triggering an audible and visual alarm in the plant area.

[0031] Extreme operating condition verification: Simulating highly contaminated influent after a nuclear accident, with boron concentration of 8 ppm and cesium-137 activity of 5 Bq / L, the pretreatment unit reduced the boron concentration to 0.5 ppb through a boron adsorption resin bed. The emergency purification channel suppressed the radioactivity activity to 0.08 Bq / L within 18 minutes. The energy management module switched to explosion-proof battery pack power supply to ensure 72 hours of uninterrupted operation. Maintenance personnel quickly replaced the resin module using a radiation-resistant robotic arm, and no secondary pollution was generated throughout the process.

[0032] Implementation Value: The system successfully intercepted three cooling water contamination incidents during the nuclear power plant's trial operation. The modular design improved the efficiency of radioactive resin replacement, reduced the risk of nuclide activation in the dual-wavelength ultraviolet unit, and ultimately ensured that the quality of the produced water consistently met the nuclear island makeup water standards, providing key technical support for nuclear power safety.

[0033] Example 3: Preparation scenario of water for injection in a biopharmaceutical plant The system's raw water is purified water with an initial endotoxin level greater than 0.25 EU / mL and a microbial load greater than 100 CFU / mL. The target product water must meet the pharmacopoeia's sterile and pyrogen-free standards, with an endotoxin level less than 0.001 EU / mL and a microbial load less than 1 CFU / 100mL.

[0034] Biological pollution control pretreatment: Raw water is pumped into the multi-stage barrier unit of the pretreatment module. First, it passes through an ultrafiltration membrane to remove endotoxins and large molecular organic matter. Then, it enters a dual-media filter to adsorb residual organic impurities. Finally, the water temperature is kept constant at 80℃ for 30 minutes through a heat exchanger to inactivate microorganisms. The softening unit uses high-temperature resistant chelating resin to simultaneously remove calcium and magnesium ions and capture metal ions. The regeneration controller automatically triggers resin regeneration based on the amount of endotoxin adsorbed. Pretreatment data is uploaded to the central database in real time.

[0035] Aseptic purification and real-time quality control: The purification module is specially designed for biopharmaceutical requirements. The reverse osmosis unit is equipped with a sanitary 316L stainless steel membrane housing, which operates at a high temperature of 85℃ and has a stable desalination rate of more than 99.5%. The electrode deionization unit integrates an ozone generator, which periodically injects 0.1ppm ozone water to inhibit biofilm formation. The ultraviolet sterilization unit uses dual-wavelength 185 / 254nm high-intensity lamps to ensure a pyrogen decomposition rate of more than 99.99%. All pipelines are made of electropolished seamless steel pipes. The in-situ cleaning system automatically performs pure steam sterilization every day.

[0036] Aseptic environment monitoring linkage: The monitoring module deploys special sensors at key nodes. The online endotoxin detector samples and analyzes every 15 minutes. The pyrogen content is dynamically determined using the Limulus amebocyte lysate (LAL) reagent method. The laser colony counter scans the microbial concentration in the water flow in real time. The temperature sensor controls the heat exchanger output in a closed loop. When the analysis unit detects an abnormal microbial count, the control module immediately initiates a three-level response. The algorithm matching unit calls the sterilization protocol, increases the power of the ultraviolet unit and activates ozone-assisted sterilization. The execution unit switches to the backup aseptic water storage tank, and the alarm module locks the contaminated batch and triggers an isolation alarm.

[0037] Process Validation and Maintenance: Simulated microbial contamination event test, Pseudomonas aeruginosa was artificially injected, the ultrafiltration membrane retained more than 99.9% of microorganisms within 2 minutes, the high temperature ultraviolet unit inactivated the remaining colonies within 10 seconds, the in-situ sterilization system was treated with pure steam at 121°C for 30 minutes, and maintenance personnel replaced consumables through a sterile transfer window. The entire process met the cleanliness requirements.

[0038] Implementation Value: The system ran continuously for 6 months on the monoclonal antibody production line, successfully preventing three microbial contamination events. The modular design improved the efficiency of ultrafiltration membrane replacement, reduced the energy consumption of the high-temperature reverse osmosis process, and ultimately ensured that the produced water stably met the standards for water for injection, thus providing a guarantee for the safe production of biological agents.

[0039] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0040] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A modular ultrapure water preparation and online monitoring system, characterized in that: It includes a pretreatment module, a purification module, a monitoring module, and a control module; The pretreatment module includes an inlet unit, a filtration unit, and a softening unit, which perform preliminary treatment on the raw water and generate pretreated water quality data. The purification module includes a reverse osmosis unit, an electrode deionization unit, and an ultraviolet sterilization unit, which are connected to the pretreatment module to perform deep purification treatment on the pretreated water and generate ultrapure water. The monitoring module includes a water quality sensing unit, a data acquisition unit, and an analysis unit. It is connected to the purification module and collects the conductivity, TOC, and particulate matter parameters of ultrapure water in real time. It also performs water quality anomaly detection based on artificial intelligence algorithms and generates real-time monitoring data. The control module includes an algorithm matching unit and an execution unit, which are connected to the monitoring module and the purification module. Based on the real-time monitoring data, the module calls an optimization algorithm to dynamically adjust the purification parameters and outputs control commands. The system uses modular interfaces to enable detachable connections between modules, supporting flexible expansion.

2. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The pretreatment module's inlet unit includes a flow sensor and a pressure regulator to monitor the inlet flow and pressure in real time. The filtration unit adopts a multi-stage filter structure, including a sedimentation filter, an activated carbon filter, and a microfiltration membrane, to remove suspended solids and organic matter. The softening unit includes an ion exchange resin tank and a regeneration controller to reduce water hardness and generate softened water data.

3. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The reverse osmosis unit of the purification module includes a high-pressure pump, an RO membrane module, and a concentrate recovery device, which separates dissolved salts and generates reverse osmosis water. The electrode deionization unit includes an ion exchange membrane stack and a DC power supply, which are connected to the reverse osmosis unit to remove ionic impurities and generate deionized water data. The ultraviolet sterilization unit includes a UV lamp and a radiation sensor, which sterilize the water and output ultraviolet intensity data.

4. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The water quality sensing unit of the monitoring module includes a conductivity sensor, a TOC analyzer, and a laser particle counter, installed at the output end of the purification module. It measures the conductivity, total organic carbon, and particulate matter concentration of the ultrapure water in real time. The data acquisition unit uses a multi-channel ADC converter to convert the sensor signals into digital data. The analysis unit embeds a machine learning model to compare historical data and detect water quality anomalies. The water quality anomaly detection uses a normalized comprehensive index formula: ； In the formula, QI is the water quality anomaly index, and an anomaly alarm is triggered when QI > 1.0; C represents real-time conductivity, in μS / cm, normalized reference. ; T represents real-time total organic carbon (TOC), in ppb, normalized basis. ; P represents the real-time particulate matter concentration, in units of particles / mL, normalized baseline. ; These are the weighting coefficients. Assignment based on parameter importance; Thresholds: The QI alarm threshold is 1.0, and the error tolerance threshold is less than 0.

05.

5. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The algorithm matching unit of the control module includes a storage subunit and a matching processor. The storage subunit pre-stores various water quality optimization algorithms, including neural network algorithms, PID control algorithms and fuzzy logic algorithms. The matching processor performs algorithm matching based on the real-time monitoring data and preset thresholds to generate the optimal control strategy.

6. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The execution unit of the control module includes a motor driver and a valve controller, which are connected to the purification module. It adjusts the speed of the high-pressure pump and the voltage of the ion exchange membrane stack according to the optimal control strategy. The PID control output uses a normalized formula. ； In the formula, For normalized control output, range [0,1], corresponding to the percentage of actual voltage and speed; To normalize the error, , To account for the error of the measured parameters, ; The normalized PID gain coefficient, Based on system optimization; Threshold: Error triggers adjustment threshold. Control the output saturation threshold .

7. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The system also includes an alarm module connected to the monitoring module, which includes an audible and visual alarm and a network communication unit. The audible and visual alarm triggers an alarm when water quality exceeds the standard. The network communication unit supports 4G / 5G and IoT protocols and remotely transmits monitoring data to the cloud platform.

8. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The system further includes an energy management module integrated into the control module, comprising a solar converter and a battery pack. The solar converter collects renewable energy, and the battery pack stores electrical energy and supplies power to the ultraviolet sterilization unit and the monitoring module, achieving an energy self-sufficiency rate of more than 80%.

9. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The modular interface uses standardized quick-connect couplings and O-rings, supports hot-swappable functionality, and ensures the sealing performance and rapid replacement of each module under pressures less than 1.0 MPa.

10. The modular ultrapure water preparation and online monitoring system according to claim 1, characterized in that: The system has an external user interaction module, including a touch screen and a mobile APP. The touch screen displays real-time water quality parameters and system status, and the mobile APP connects to the control module through an API interface, supporting remote parameter setting and fault diagnosis.

Images