Reverse osmosis membrane cleaning control method and device

Abstract

The invention relates to the technical field of big data, in particular to a reverse osmosis membrane cleaning control method and device.The method comprises the steps that operation data and real-time operation data of a target reverse osmosis membrane system at historical target time are collected; based on the operation data of the historical target time and the real-time operation data, obtaining a change trend of a target index in the target reverse osmosis membrane system in the future target time; and determining a cleaning scheme of the target reverse osmosis membrane system according to the change trend, and controlling a cleaning actuator corresponding to the target reverse osmosis membrane system to execute the cleaning scheme according to the cleaning scheme. Therefore, the problems that in related technologies, a reverse osmosis membrane cleaning control method depending on artificial experience and a fixed period is prone to causing cleaning time lag, the real-time state of membrane pollution cannot be dynamically responded, the membrane performance is degraded, the economical efficiency and safety of a system are affected, and the service life of the system is influenced are solved. And increasingly strict water quality standards and energy-saving and consumption-reducing requirements are difficult to meet.

Description

Technical Field

[0001] This application relates to the field of big data technology, and in particular to a method and apparatus for cleaning and controlling reverse osmosis membranes. Background Technology

[0002] Reverse osmosis membrane technology has advantages such as simple operation and no secondary pollution. The pre-desalination of boiler feedwater in thermal power plants is highly dependent on reverse osmosis membrane technology.

[0003] In related technologies, the management and cleaning of reverse osmosis membranes in thermal power plants often rely on manual experience and fixed-cycle cleaning strategies such as quarterly cleaning.

[0004] However, in related technologies, the cleaning control method of reverse osmosis membranes, which relies on human experience and fixed cycles, is prone to delays in cleaning timing and cannot dynamically respond to the real-time state of membrane fouling. It is also prone to excessive (increasing costs) or insufficient (accelerating scaling) cleaning agents due to water quality fluctuations, leading to membrane performance degradation. The decrease in flux and increase in pressure drop caused by reverse osmosis membrane fouling (such as colloidal deposition, metal oxide scaling, microbial growth, etc.) will directly affect the economy and safety of the system, making it difficult to meet increasingly stringent water quality standards and energy conservation and consumption reduction requirements, which urgently need to be addressed. Summary of the Invention

[0005] This application provides a cleaning control method and apparatus for reverse osmosis membranes to solve the problems in related technologies, such as the reliance on manual experience and fixed cycles for cleaning control of reverse osmosis membranes, which easily leads to delayed cleaning timing, inability to dynamically respond to the real-time state of membrane fouling, resulting in membrane performance degradation, affecting the economy and safety of the system, and making it difficult to meet increasingly stringent water quality standards and energy conservation and consumption reduction requirements.

[0006] The first aspect of this application provides a cleaning control method for a reverse osmosis membrane, comprising the following steps: collecting historical and real-time operating data of a target reverse osmosis membrane system; based on the historical and real-time operating data, obtaining the changing trend of a target indicator in the target reverse osmosis membrane system over a future target time; determining a cleaning scheme for the target reverse osmosis membrane system according to the changing trend, and controlling the cleaning actuator corresponding to the target reverse osmosis membrane system to execute the cleaning scheme according to the cleaning scheme.

[0007] Optionally, in one embodiment of this application, the collection of the target reverse osmosis membrane system's operating data at a historical target time and real-time operating data includes: collecting the target reverse osmosis membrane system's water quality parameters, membrane module operating indicators, and module status information at the historical target time and the current time; and determining the operating data at the historical target time and the real-time operating data based on the water quality parameters, membrane module operating indicators, and module status information at the historical target time and the current time.

[0008] Optionally, in one embodiment of this application, obtaining the changing trend of the target indicator in the target reverse osmosis membrane system in the future target time based on the historical target time operation data and the real-time operation data includes: obtaining the time-series dependency, local features, and global features of the target indicator based on the historical target time operation data and the real-time operation data; and determining the changing trend by combining the time-series dependency, local features, and global features.

[0009] Optionally, in one embodiment of this application, the method further includes: acquiring the water quality parameter change trend of the target reverse osmosis membrane system, the inter-stage pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information based on the historical target time operation data and the real-time operation data; generating a pollution warning prompt for the target reverse osmosis membrane system based on the water quality parameter change trend, the inter-stage pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information, and feeding back the pollution warning prompt to the user of the target reverse osmosis membrane system.

[0010] Optionally, in one embodiment of this application, determining the cleaning scheme of the target reverse osmosis membrane system based on the changing trend includes: determining the cleaning duration, cleaning agent dosage, and valve opening degree of the target reverse osmosis membrane system based on the changing trend; and determining the cleaning scheme based on the cleaning duration, cleaning agent dosage, and valve opening degree.

[0011] A second aspect of this application provides a cleaning control device for a reverse osmosis membrane, comprising: a data acquisition module for acquiring historical and real-time operating data of a target reverse osmosis membrane system; a first acquisition module for acquiring, based on the historical and real-time operating data, the changing trend of a target indicator in the target reverse osmosis membrane system over a future target time; and a control module for determining a cleaning scheme for the target reverse osmosis membrane system based on the changing trend, and controlling a cleaning actuator corresponding to the target reverse osmosis membrane system to execute the cleaning scheme according to the cleaning scheme.

[0012] Optionally, in one embodiment of this application, the acquisition module includes: an acquisition unit, used to acquire water quality parameters, membrane module operating indicators, and module status information of the target reverse osmosis membrane system at a historical target time and at the current time; and a first determination unit, used to determine the operating data at the historical target time and the real-time operating data based on the water quality parameters, membrane module operating indicators, and module status information at the historical target time and at the current time.

[0013] Optionally, in one embodiment of this application, the acquisition module includes: an acquisition unit, configured to acquire the temporal dependency, local features, and global features of the target indicator based on the historical target time running data and the real-time running data; and a second determination unit, configured to determine the change trend by combining the temporal dependency, local features, and global features.

[0014] Optionally, in one embodiment of this application, it further includes: a second acquisition module, configured to acquire, based on the historical target time operation data and the real-time operation data, the water quality parameter change trend of the target reverse osmosis membrane system, the inter-segment pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information; and a generation module, configured to generate a pollution warning prompt for the target reverse osmosis membrane system based on the water quality parameter change trend, the inter-segment pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information, and to feed back the pollution warning prompt to the user of the target reverse osmosis membrane system.

[0015] Optionally, in one embodiment of this application, the root control module includes: a third determining unit, configured to determine the cleaning duration, cleaning agent dosage, and valve opening degree of the target reverse osmosis membrane system based on the changing trend; and a fourth determining unit, configured to determine the cleaning scheme based on the cleaning duration, cleaning agent dosage, and valve opening degree.

[0016] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the reverse osmosis membrane cleaning control method as described in the above embodiments.

[0017] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described reverse osmosis membrane cleaning control method.

[0018] A fifth aspect of this application provides a computer program product, including a computer program that, when executed, is used to implement the above-described reverse osmosis membrane cleaning control method.

[0019] This application embodiment can obtain the changing trends of key indicators in the future target time based on the historical and real-time operating data of the target reverse osmosis membrane system, thereby determining the cleaning plan for the target reverse osmosis membrane system and controlling the relevant actuators to execute it. This achieves closed-loop control of the entire process from data acquisition to decision execution by systematically collecting historical and real-time operating data, predicting the future changing trends of key indicators, and formulating cleaning plans accordingly. It transforms the traditional experience-driven or timed cleaning mode into a data-driven intelligent cleaning strategy, which can more accurately identify membrane fouling trends, predict performance degradation, and take targeted cleaning measures before or in the early stages of fouling. This effectively extends membrane life, improves system operating efficiency, reduces unnecessary cleaning frequency and reagent consumption, lowers manual intervention costs, and enhances the overall intelligence and precision of operation and maintenance. Therefore, it solves the problems in related technologies where the cleaning control method relying on manual experience and fixed cycles for reverse osmosis membranes easily leads to delayed cleaning timing, inability to dynamically respond to the real-time state of membrane fouling, resulting in membrane performance degradation, affecting the system's economy and safety, and making it difficult to meet increasingly stringent water quality standards and energy-saving and consumption-reducing requirements.

[0020] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0021] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the process logic of an intelligent operation and maintenance decision-making and automatic cleaning system for reverse osmosis membranes in thermal power plants based on big data analysis, according to an embodiment of this application. Figure 2 This is a flowchart of a reverse osmosis membrane cleaning control method according to an embodiment of this application; Figure 3 This is a schematic diagram of the feedback signal of a reverse osmosis automatic chemical dosing DCS system according to an embodiment of this application; Figure 4 This is a schematic diagram of the feedback signal of a reverse osmosis automatic dosing receiving DCS system according to an embodiment of this application; Figure 5 This is a schematic diagram of the structure of the reverse osmosis membrane cleaning control device provided according to an embodiment of this application; Figure 6 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application.

[0022] Figure label: 10-Reverse osmosis membrane cleaning control device: 100-Acquisition module, 200-First acquisition module and 300-Control module; 601-Memory, 602-Processor and 603-Communication interface. Detailed Implementation

[0023] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0024] The following description, with reference to the accompanying drawings, illustrates a reverse osmosis membrane cleaning control method and apparatus according to embodiments of this application. Addressing the issues raised in the background section regarding the aforementioned related technologies, the reliance on manual experience and fixed-cycle reverse osmosis membrane cleaning control methods often leads to delayed cleaning timing, an inability to dynamically respond to real-time membrane fouling, resulting in membrane performance degradation, impacting system economy and safety, and failing to meet increasingly stringent water quality standards and energy conservation requirements. This application provides a reverse osmosis membrane cleaning control method. In this method, the changing trends of key indicators over future target times can be obtained based on historical and real-time operating data of the target reverse osmosis membrane system, thereby determining the cleaning scheme for the target reverse osmosis membrane system and controlling the relevant actuators to execute it. This system achieves closed-loop control from data acquisition to decision-making by systematically collecting historical and real-time operational data, predicting future trends of key indicators, and formulating cleaning plans accordingly. It transforms traditional experience-driven or timed cleaning models into data-driven intelligent cleaning strategies, enabling more accurate identification of membrane fouling trends, prediction of performance degradation, and targeted cleaning measures before or in the early stages of fouling. This effectively extends membrane life, improves system operating efficiency, reduces unnecessary cleaning frequency and reagent consumption, lowers manual intervention costs, and enhances the overall intelligence and precision of operation and maintenance. This solves the problems in related technologies where reliance on manual experience and fixed-cycle reverse osmosis membrane cleaning control methods leads to delayed cleaning timing, inability to dynamically respond to real-time membrane fouling, resulting in membrane performance degradation, impacting system economy and safety, and failing to meet increasingly stringent water quality standards and energy conservation requirements.

[0025] Before explaining the cleaning and control method of the reverse osmosis membrane in the embodiments of this application, the architecture of the intelligent operation and maintenance decision-making and automatic cleaning system for reverse osmosis membranes in thermal power plants based on big data analysis in the embodiments of this application will be explained first.

[0026] Figure 1This is a schematic diagram of the process logic of a big data analysis-based intelligent operation and maintenance decision-making and automatic cleaning system for reverse osmosis membranes in thermal power plants, according to one embodiment of this application. Figure 1 As shown, the intelligent operation and maintenance decision-making and automatic cleaning system for reverse osmosis membranes in thermal power plants based on big data analysis (hereinafter referred to as the automatic cleaning system for reverse osmosis membranes) in this application embodiment mainly includes, but is not limited to, three parts: Real-time data layer: Integrates real-time water quality, operating parameters and historical operation and maintenance data, and uses big data technologies such as machine learning and statistical models to mine patterns, identify anomalies and predict trends from multi-source data.

[0027] Intelligent Analysis Layer: Based on the analysis results, it generates specific and executable operation and maintenance decisions. Through the feedback mechanism, it continuously optimizes the decision model, seeks to achieve dynamic optimization calculation of scale inhibitor dosage and accurate determination of the timing of reverse osmosis membrane cleaning, thereby providing core decision support for the intelligent operation and maintenance of reverse osmosis membranes in thermal power plants.

[0028] Decision output layer: Based on big data trend analysis, through multi-parameter linkage analysis and machine learning prediction models, it dynamically identifies membrane fouling risks and triggers graded early warnings, proactively prompting operation and maintenance interventions to prevent the deterioration of reverse osmosis membrane performance.

[0029] Specifically, Figure 2 This is a flowchart illustrating a reverse osmosis membrane cleaning control method provided in an embodiment of this application.

[0030] like Figure 2 As shown, the cleaning and control method for the reverse osmosis membrane includes the following steps: In step S201, the operating data and real-time operating data of the target reverse osmosis membrane system during the historical target time period are collected.

[0031] It is understandable that the target reverse osmosis membrane system here refers to the specific reverse osmosis membrane system object that is subject to cleaning control.

[0032] In some embodiments, this application can collect the operating data of the target reverse osmosis membrane system in real time during a historical target time period, as well as real-time operating data, such as membrane pressure difference, permeate flow rate, desalination rate, feed water temperature, and water ion concentration, so as to extract abnormal information of the target reverse osmosis membrane system based on these data.

[0033] Here, the historical target time refers to a continuous historical data period that traces back from the current moment, such as a historical two-hour period, a historical six-hour period, etc.

[0034] Furthermore, after collecting the historical and real-time operating data of the target reverse osmosis membrane system, this application can perform certain data processing on the collected data, such as cleaning, conversion, desensitization, and standardization, to build a unified data warehouse or data lake and ensure the quality and consistency of the data.

[0035] In the embodiments of this application, data cleaning and transformation can be performed on historical operating data and real-time operating data within a target time period, but not limited to; and, for data loss that may be caused by sensor failure or communication interruption (such as a flow meter suddenly losing signal), interpolation methods (such as linear interpolation) or default values ​​can be used to fill the gaps and avoid data gaps.

[0036] For example, this application can perform different data processing on abnormal data, such as missing values ​​(data with empty attribute values); outliers (noise) (data that deviates from the normal range); duplicate values ​​(records that are the same or highly similar (such as duplicate order information); inconsistent data (format, encoding or naming conflicts); and irrelevant data.

[0037] For example, for sensor drift data, embodiments of this application can identify and remove it using statistical methods, such as using IQR (interquartile range) to filter sensor drift data to prevent misleading analysis; if multiple devices collect the same indicator at the same time, duplicate data needs to be removed and only reliable data needs to be retained; and, a sliding window can be used to repair missing values, thereby reducing the amount of data and improving analysis efficiency while maintaining data integrity.

[0038] Furthermore, the cleaned and repaired data must be converted into a format that meets the requirements of the DCS system. For example, it is necessary to convert analog signals (4-20mA) and digital signals (such as RS485 / Modbus) from sensors (flow meters, temperature / conductivity meters). Analog signals can be converted into digital quantities (numerical values) through AD conversion (analog-to-digital conversion), but are not limited to; digital signals can be parsed into specific numerical values ​​(such as flow rate values, temperature values); digital signals from online monitoring instruments (such as residual chlorine, SDI) (such as Ethernet / IP, Profibus) need to have key indicator values ​​(such as residual chlorine concentration mg / L) extracted; manually recorded SDI values ​​may be in Excel / CSV format and must also be converted into digital quantities.

[0039] This application embodiment can also convert the transformed data into structured data (database tables) or binary files according to the requirements of the DCS / PLC system. The operating data stored in the DCS system needs to be exported in a structured format (such as SQL tables) for integration with data from other devices.

[0040] Additionally, the processed full data (such as cleaning time, dosage, and membrane state changes) can also be used for trend analysis of the DCS of the target reverse osmosis membrane system. For example, based on data within a historical target time period, the cleaning effect of a batch of cleaning agents can be evaluated through statistical models (such as regression analysis), such as the rate of decrease in membrane pressure difference and the rate of recovery of desalination rate after cleaning. This data can be fed back to the DCS of the target reverse osmosis membrane system to optimize the cleaning parameters of the target reverse osmosis membrane system for the next time, such as extending the soaking time.

[0041] By cleaning and transforming the data, the embodiments of this application can ensure the reliability of the data; by removing outliers and filling in missing values, the embodiments of this application can avoid erroneous decisions caused by "dirty data" (such as misjudging membrane fouling due to abnormal flow values).

[0042] Therefore, the embodiments of this application realize data integration, unifying data of different formats and sources into structured data and storing it in a database (such as MySQL or Oracle), supporting cross-device analysis (such as the correlation between flow rate and residual chlorine concentration), and effectively supporting subsequent applications. For example, the data after cleaning and conversion can be used for machine learning models (such as membrane fault early warning) and visualization dashboards (such as real-time monitoring of permeate flow and desalination rate), thereby providing strong data support for the efficient operation of thermal power plants (such as optimizing recovery rate) and cost control (such as reducing the number of membrane replacements).

[0043] Optionally, in one embodiment of this application, collecting the target reverse osmosis membrane system's operating data and real-time operating data within a historical target time period includes: collecting the target reverse osmosis membrane system's water quality parameters, membrane module operating indicators, and module status information at the historical target time and the current time; and determining the operating data within the historical target time period and the real-time operating data based on the water quality parameters, membrane module operating indicators, and module status information at the historical target time and the current time period.

[0044] In actual implementation, the operational data collected in this application during the historical target time period and the real-time operational data include, but are not limited to, the water quality parameters, membrane module operation indicators and module status information of the target reverse osmosis membrane system at the historical target time and the current time.

[0045] The water quality parameters, membrane module operating indicators, and module status information within the historical target time period constitute the operating data of the target reverse osmosis membrane system during that period. The water quality parameters, membrane module operating indicators, and module status information at the current moment constitute the real-time operating data of the target reverse osmosis membrane system.

[0046] In this context, "membrane module" refers to the core filtration unit and smallest operable unit of the reverse osmosis system, such as reverse osmosis membrane sheets, flow guide grids, and central permeate pipes. "Component" refers to the real-time physical parameters that directly reflect whether the key mechanical equipment of the reverse osmosis system is operating normally and stably, such as high-pressure pumps (which provide driving pressure for the reverse osmosis membrane) and various dosing pumps and cleaning pumps, numerous pneumatic / electric valves in the system, such as inlet valves, permeate valves, concentrate discharge valves, cleaning loop valves, pressure sensors installed at the inlet and concentrate outlets of each membrane section, etc.

[0047] For example, taking an automatic reverse osmosis membrane cleaning system as an example, the real-time data layer is based on an industrial-grade data integration solution for a thermal power plant SIS system, which can form the physical sensing and information integration foundation of the entire reverse osmosis membrane cleaning system.

[0048] The real-time data layer is characterized by the use of a multi-node high-precision sensor network to collect key water quality parameters (including but not limited to calcium and magnesium ion concentration, chloride ion content, pH value, turbidity, ORP (oxidation-reduction potential), temperature, etc.) during the operation of the reverse osmosis system, membrane module operating indicators (such as transmembrane pressure difference (the pressure difference required to drive water through the membrane, i.e., the difference between the average pressure on the feed water side and the pressure on the product water side), product water flow rate (the volume of qualified fresh water (permeate) produced by the reverse osmosis membrane system per unit time), desalination rate (the membrane's ability to remove dissolved salts from the water)), and core equipment status information (pump current, valve opening and closing status, inter-section pressure distribution (at a specific moment in the reverse osmosis system, along the water flow direction (from the first section to the last section), the spatial pressure pattern formed by the feed water pressure and concentrate pressure of each membrane module) at high frequency and synchronously.

[0049] Furthermore, the data acquisition devices deployed in this real-time data layer can, but are not limited to, upload data to the power plant's existing Supervisory Information System (SIS) via standardized communication protocols (such as Modbus TCP / IP or OPC UA) to achieve unified aggregation of raw monitoring and collected data within the plant-wide production control platform. Time alignment can be achieved using BeiDou clock technology to align with the SIS system's timestamps. Subsequently, a customized ETL (Extract-Transform-Load) middleware can extract structured and semi-structured data from the SIS system into a separately constructed water treatment-specific analysis database to complete data cleaning, outlier correction, and time-series reconstruction, forming a high-quality historical dataset suitable for advanced analysis.

[0050] In addition, in order to collect this data, embodiments of this application may, but are not limited to, first install certain data acquisition devices (monitoring equipment) in the target reverse osmosis membrane system.

[0051] For example, this application can install flow meters and differential pressure sensors at the reverse osmosis inlet, outlet, concentrate outlet, and high-pressure pump outlet to count the water volume and pressure at each location, calculate the recovery rate, and ensure that the membrane module operates stably under the design pressure.

[0052] In this embodiment, a residual chlorine detector can also be installed at the pretreatment outlet to monitor the residual chlorine concentration in the pretreated water, ensuring that it is below the threshold allowed by the reverse osmosis membrane and preventing membrane oxidation and degradation; and a thermometer can be installed at the reverse osmosis feed water end, because water temperature changes will affect membrane flux and desalination rate.

[0053] Furthermore, in this embodiment of the application, an SDI detector can also be installed at the pretreatment outlet to assess the risk of contamination by suspended solids and colloids in the water, ensuring that the water quality entering the membrane system meets the standards.

[0054] Monitoring instruments can, but are not limited to, use hard-wired mode to transmit the collected data to the power plant's SIS system. Then, ETL and other tools are used to transmit the point data of the SIS system to a real-time data platform (such as Flink) to perform ETL (extract-transform-load) on multi-source data (reverse osmosis membrane system operation data and real-time operation data over a certain period of time) from sensors, SCADA systems, and laboratory analysis equipment, thereby unifying the data format (such as timestamps and units) and building a "reverse osmosis membrane operation and maintenance data lake".

[0055] By installing monitoring equipment and collecting data, the embodiments of this application can connect the real-time data (such as flow rate, pressure difference, and residual chlorine) stored in the database to the DCS system (distributed control system) to achieve visual monitoring (such as displaying water production and desalination rate on the instrument panel) and automatic alarms (such as triggering an audible and visual alarm when residual chlorine exceeds the standard, prompting the operator to add reducing agent; and starting the cleaning program when the pressure difference is abnormal).

[0056] The data collected in this embodiment provides comprehensive and timely underlying data support for the target reverse osmosis membrane system. Intelligent analysis of this data enables quantitative assessment of cleaning risks, extracting cleaning determination patterns from massive historical and real-time data. Combining real-time data with intelligent analysis models and then connecting to the power plant's DCS system, data-driven closed-loop intelligent operation and maintenance is ultimately achieved, effectively improving the operating efficiency and lifespan of the reverse osmosis membrane system. By distinguishing between historical and real-time data, the system can consider both long-term performance evolution and short-term state fluctuations, enhancing its ability to perceive the health status of the membrane system and laying a data foundation for accurate prediction and timely intervention.

[0057] Step S202: Based on historical target time operation data and real-time operation data, obtain the changing trend of target indicators in the target reverse osmosis membrane system in the future target time.

[0058] In other embodiments, this application can obtain the changing trend of target indicators in the target reverse osmosis membrane system in the future target time based on the historical target time operation data and real-time operation data of the target reverse osmosis system.

[0059] Here, the target indicators can be understood as core monitoring parameters reflecting the operating status, treatment efficiency, and membrane element health of the reverse osmosis system, while also including human-controlled parameters that affect these states. In this embodiment, the target indicators can be, but are not limited to, divided into two categories: operating performance indicators and operational control indicators. I. Operational performance indicators (core monitoring parameters reflecting system status) These indicators directly reflect the operating performance of the reverse osmosis system and serve as the basis for judging whether the membrane elements are functioning properly. For example: The pressure difference between the inside and outside of the reverse osmosis membrane, which is the difference between the pressure on the feed water side and the pressure on the product water side of the reverse osmosis membrane, is a key indicator for measuring the degree of fouling and clogging of membrane elements. Suspended solids, colloids, and scale attached to the membrane surface will increase the resistance to water flow and cause the pressure difference to rise.

[0060] Desalination rate, which is the proportion of salt removed from raw water by a reverse osmosis system, is a core indicator for measuring the system's separation efficiency. A decrease in desalination rate usually indicates that the membrane element is damaged or aging, or that fouling on the membrane surface causes some salt to leak through the membrane layer.

[0061] Permeate production, which is the amount of qualified freshwater produced by a reverse osmosis system per unit time, is a core indicator for measuring the system's processing capacity. A decrease in permeate production may be caused by factors such as membrane fouling, increased pressure differential, and decreased feed water temperature.

[0062] Temperature, specifically the feed water temperature of a reverse osmosis system, is a key environmental parameter affecting membrane flux (the amount of water produced per unit area of ​​the membrane). The water permeability of a reverse osmosis membrane increases with increasing temperature (typically, the water production increases by about 2% to 3% for every 1°C increase in temperature), but excessively high temperatures will accelerate membrane aging.

[0063] II. Operational control indicators (key intervention parameters for human intervention) These types of indicators are control parameters set by operations and maintenance personnel to maintain stable system operation, and they directly affect the changing trends of operational performance indicators. For example: The dosage of chemicals refers to the amount of various chemicals added during the operation of the reverse osmosis system. These mainly include scale inhibitors (to prevent membrane scaling), bactericides (to inhibit microbial contamination), and reducing agents (to remove residual chlorine from the raw water and protect membrane elements).

[0064] The timing of agent addition refers to the frequency and duration of addition of various agents (such as continuous addition or intermittent addition).

[0065] The embodiments of this application can analyze historical operating data and real-time operating data over a certain period of time to obtain the correlation between these indicators and the target reverse osmosis system in time series or with some water quality parameters, and then obtain the changing trend of these indicators in the future target time based on these correlations.

[0066] Here, the future target time refers to a continuous future data period extending from the current moment, such as the next two hours or six hours.

[0067] For example, this application can, but is not limited to, present the current status and historical changes of key indicators through a data analysis layer, based on descriptive analysis (such as trend analysis, comparative analysis, and proportion analysis) and diagnostic analysis (such as anomaly detection and root cause analysis), revealing the potential correlations behind the data, such as the relationship between the pressure difference inside and outside the reverse osmosis membrane, desalination rate, temperature, permeate volume and the dosage and time of added chemicals, thereby predicting the changing trend of key indicators over a certain period of time in the future.

[0068] For example, insufficient dosage (such as insufficient scale inhibitor) or improper addition time will accelerate membrane scaling, thereby increasing the pressure differential; conversely, proper dosage can slow down the rate of pressure differential increase.

[0069] Insufficient dosage of bactericide or excessively long intervals between additions may lead to the growth of microorganisms on the membrane surface, damaging the membrane separation layer and reducing the desalination rate; excessively high temperatures may cause slight expansion of the membrane pore size, which may also slightly reduce the desalination rate.

[0070] Insufficient scale inhibitor dosage will cause membrane scaling, directly reducing water production; increased temperature will increase the permeation rate of water molecules, which can increase water production when other conditions remain unchanged.

[0071] Temperature changes directly affect water production and desalination rate, and also affect the effectiveness of chemicals (for example, the efficiency of some scale inhibitors will decrease at high temperatures, and the dosage needs to be adjusted).

[0072] Insufficient dosage can lead to an increase in pressure differential, a decrease in permeate flow, and a reduction in desalination rate. Excessive dosage may result in waste of reagents or even irreversible damage to membrane elements.

[0073] If the interval between intermittent addition of bactericides is too long, microorganisms on the membrane surface will continue to multiply, causing biofouling; if scale inhibitors are not added synchronously when the system is started, the high hardness of the raw water in the initial operation will directly form scale on the membrane surface, causing the pressure differential to rise rapidly.

[0074] By combining these fundamental relationships, and by using the historical and real-time operating data of the target reverse osmosis system at the target time, we can obtain the changing trends of some key indicators in the target reverse osmosis membrane system over a certain period of time in the future.

[0075] Optionally, in one embodiment of this application, the change trend of target indicators in the target reverse osmosis membrane system in the future target time is obtained based on historical target time operation data and real-time operation data, including: obtaining the time-series dependency, local features and global features of the target indicators based on historical target time operation data and real-time operation data; and determining the change trend by combining the time-series dependency, local features and global features.

[0076] Based on the descriptions of other embodiments, it is understood that this application can combine the potential correlations behind the data to obtain the changing trends of some key indicators in the target reverse osmosis membrane system over a certain period of time through the historical target time operation data and real-time operation data of the target reverse osmosis system.

[0077] In actual implementation, this application may, but is not limited to, use the historical operating data and real-time operating data of the target reverse osmosis system over a certain period of time to capture the long-term time-series dependence of key indicators in the target reverse osmosis system, as well as the global data characteristics presented by the historical operating data and the local characteristics presented by the real-time operating data, to jointly determine the changing trend of certain indicators in the target reverse osmosis membrane system over a certain period of time in the future.

[0078] For example, this application may, but is not limited to, utilize Long Short-Term Memory (LSTM) networks to capture the long-term temporal dependence of certain indicators in the target reverse osmosis membrane system (such as predicting whether the membrane pressure difference will reach the cleaning threshold within 7 days, adapting to the gradual nature of reverse osmosis membrane fouling); and use Temporal Convolutional Network (TCN) to process the time-series data in parallel, extracting global and local features (such as combining historical pressure difference data for 3 months with real-time water quality data to improve prediction accuracy), predicting the dosing time, permeate flow fluctuations, and equipment failure risks of the target reverse osmosis membrane system, thereby providing data support for forward-looking decision-making.

[0079] For example, in the intelligent operation and maintenance decision-making and automatic cleaning system for reverse osmosis membranes in thermal power plants based on big data analysis, the intelligent analysis layer serves as a bridge between the real-time data layer and the decision output layer. Its core objective is to transform real-time collected multi-source data (water quality, operating parameters) into "predictable trend signals," solving the key problem of "inability to detect performance degradation in advance" in the operation and maintenance of RO membranes (reverse osmosis membranes), and providing a basis for decision-making for subsequent dynamic dosing and precise cleaning.

[0080] In this embodiment, the intelligent analysis layer integrates four major categories of technologies: real-time stream processing, time series prediction, machine learning (ML), and anomaly detection.

[0081] Specifically, embodiments of this application may, but are not limited to, use data processing technologies such as Apache Flink and Spark Streaming for real-time stream processing, processing real-time data (such as flow rate, pH, and pressure difference) with low latency, and ensuring the "freshness" of the data (latency ≤ 1 second).

[0082] Then, in this embodiment of the application, time series prediction can be performed using LSTM (Long Short-Term Memory) and ARIMA: the LSTM model is used to predict the future trend of key RO membrane indicators (such as the pressure difference change in the next 24 hours): input "pressure difference, flow rate, and Ca²+ concentration in the past 7 days", output "predicted value of pressure difference in the next 24 hours" (error ≤ 5%); the ARIMA model is used to supplement short-term prediction (such as the desalination rate change in the next 1 hour), solving the problem that LSTM is not sensitive to "sudden small fluctuations", and combining the two to predict the future trend of key RO membrane indicators (such as pressure difference and desalination rate) in the future over a certain period of time (such as the pressure difference change in the next 24 hours).

[0083] Additionally, embodiments of this application may employ machine learning techniques such as XGBoost and Random Forest to uncover the correlation between "water quality parameters and operating status" (e.g., the correlation between increased Ca²⁺ concentration and increased pressure difference). Furthermore, anomaly detection can be performed based on the correlation between "water quality parameters and operating status" using Isolation Forest and LOF technologies.

[0084] For example, in this embodiment, an isolation forest can be used to identify anomalous data points (such as a sudden 20% increase in differential pressure or a 10% drop in desalination rate). Then, by calculating the isolation degree of the anomalous data points, it can be determined whether the anomalous data point belongs to normal fluctuations (such as a decrease in differential pressure after cleaning) or abnormal degradation (such as a sudden increase in differential pressure caused by membrane fouling). If it belongs to abnormal degradation, an early warning can be triggered in the decision output layer. The warning threshold for triggering the early warning can be set using, but is not limited to, the 3σ principle (normal distribution) (e.g., triggering a high-risk warning when the differential pressure exceeds the mean + 3σ).

[0085] By achieving efficient fusion of multi-source heterogeneous data through an integrated architecture, relying on advanced time series models, correlation mining, clustering and anomaly detection algorithms to deeply explore operation and pollution patterns, and using feature importance assessment and interpretable AI technology to accurately identify key features, the embodiments of this application can transform the cleaning of reverse osmosis membranes from a passive response to an active prevention, from experience-driven to data-driven, and intervene before significant performance degradation through prediction and early warning.

[0086] Furthermore, the precise operation and maintenance strategy formulated based on data patterns and key characteristics can effectively reduce human experience bias, improve system transparency and interpretability, help operation and maintenance personnel understand system status and model decisions, and enhance trust and operability. It can also effectively optimize cleaning strategies dynamically based on the real-time status, fouling type and historical effects of the target reverse osmosis system, improve cleaning efficiency, extend membrane life, reduce energy and material consumption (cleaning agent, water, electricity), reduce unplanned downtime, extend the life of key assets, and significantly reduce the total life cycle cost.

[0087] This application's embodiments can extract temporal dependencies, local features, and global features from data, and comprehensively utilize this information to predict future trends. This allows for a deeper understanding of the temporal dynamics and system behavior patterns within the data, avoiding misjudgments caused by single data points or short-term fluctuations. By combining long-term trends with real-time features, the system can more accurately predict the development path of membrane fouling and identify early signals of abnormal changes, thereby improving the reliability and timeliness of predictions and providing a scientific basis for proactive maintenance.

[0088] Step S203: Determine the cleaning plan for the target reverse osmosis membrane system based on the changing trend, and control the corresponding cleaning actuator of the target reverse osmosis membrane system to execute the cleaning plan according to the cleaning plan.

[0089] In some embodiments, after obtaining the changing trends of key indicators in the target reverse osmosis membrane system over a certain period of time, this application can determine the cleaning scheme for the target reverse osmosis system based on the changing trends of these key indicators, and then control the cleaning actuator corresponding to the target reverse osmosis membrane system to execute the cleaning scheme according to the cleaning scheme.

[0090] Here, the cleaning actuator can be understood as the specific actuator used to implement the cleaning scheme of the reverse osmosis membrane system, such as a dosing valve.

[0091] For example, if the predicted pressure difference will continue to rise in the next 24 hours, this application may appropriately add a certain amount of antiscalant to the target reverse osmosis system (because insufficient dosage (such as insufficient antiscalant addition) or unreasonable addition time will accelerate membrane scaling, thereby increasing the pressure difference; conversely, reasonable addition of the agent can slow down the rate of pressure difference growth).

[0092] This application's embodiments can systematically collect historical and real-time operational data, predict future trends of key indicators, and formulate cleaning plans accordingly, achieving closed-loop control throughout the entire process from data collection to decision execution. By transforming traditional experience-driven or timed cleaning models into data-driven intelligent cleaning strategies, it can more accurately identify membrane fouling trends, predict performance degradation, and take targeted cleaning measures before or in the early stages of fouling. This effectively extends membrane life, improves system operating efficiency, reduces unnecessary cleaning frequency and reagent consumption, lowers manual intervention costs, and enhances the overall intelligence and precision of operation and maintenance.

[0093] Optionally, in one embodiment of this application, determining the cleaning scheme of the target reverse osmosis membrane system based on the changing trend includes: determining the cleaning duration, cleaning agent dosage, and valve opening degree of the target reverse osmosis membrane system based on the changing trend; and determining the cleaning scheme based on the cleaning duration, cleaning agent dosage, and valve opening degree.

[0094] In some embodiments, the cleaning scheme for the target reverse osmosis membrane system determined in this application includes, but is not limited to, the cleaning duration, cleaning agent dosage, and valve opening degree of the target reverse osmosis membrane system being cleaned.

[0095] In the embodiments of this application, the cleaning scheme of the target reverse osmosis membrane system can be executed by the decision output layer, but is not limited to. If the membrane system triggers the cleaning condition, the DCS control unit is directly triggered to realize the opening and closing of the reverse osmosis membrane cleaning valve. If the data is within the healthy operating range specified by the reverse osmosis membrane manufacturer but there is a risk of fouling in the water quality, it is determined whether to open the valve to add scale inhibitor daily.

[0096] Furthermore, during the cleaning process, the embodiments of this application may, but are not limited to, use an electric regulating valve (adjustment accuracy ±1%) to receive control signals from the DCS system for adjustment (supporting 0-100% stepless adjustment); in the cleaning process, the variable frequency metering pump (range 0-50L / h) can adjust the output flow rate according to the frequency control signal (0-50Hz) of the DCS system to control the cleaning time of the reverse osmosis membrane, the amount of chemicals added, and the valve opening degree.

[0097] Specifically, the decision output layer in this application embodiment includes, but is not limited to, the following modules: a dedicated signal interaction module, a DCS system adaptation module, and a cleaning execution module.

[0098] Among them, the core chip of the dedicated signal interaction module can, but is not limited to, use a 480MHz main frequency and support multi-protocol signal processing to realize signal interaction between the DCS system adaptation module and the cleaning execution module.

[0099] The dedicated signal interaction module includes a signal interface module whose output interface can, but is not limited to, 8 analog outputs (4-20mA) for sending dosing control signals to the DCS; 16 digital outputs (DO) for sending valve opening / closing and dosing pump start / stop commands to the DCS; a communication verification module: integrating a CRC32 verification chip to verify the transmitted digital signals and avoid signal transmission errors; both analog and digital signals adopt opto-isolation design (isolation voltage ≥2500VAC) to prevent electromagnetic interference between the DCS system and the intelligent operation and maintenance system.

[0100] Figure 3 This is a schematic diagram of the feedback signal of a reverse osmosis automatic chemical dosing DCS system according to an embodiment of this application; Figure 4 This is a schematic diagram of the feedback signal received by a reverse osmosis automatic dosing DCS system according to an embodiment of this application. Figure 3 and Figure 4 The system generates cleaning commands (digital and analog signals) and sends them to the board. The core chip of the board parses the commands, adds a CRC32 checksum to the digital signal, and converts the analog signal into a 4-20mA standard industrial signal. After the isolation module performs electromagnetic isolation on the output signal, it is transmitted to the DCS system through hard-wired output interface. After receiving the signal, the DCS system sends a reception confirmation signal back to the board. After the board verifies that there are no errors, it sends a "command delivered" receipt to the intelligent operation and maintenance system.

[0101] Furthermore, the DCS system adaptation module and the cleaning execution module may, but are not limited to, adopt the following steps in the implementation of business logic: (1) Fouling monitoring stage: The reverse osmosis membrane automatic cleaning system receives reverse osmosis membrane operation data uploaded by the DCS system in real time through a dedicated board, and updates the core indicators such as pressure difference, permeate flow rate, and desalination rate every second; (2) Cleaning judgment stage: Based on the analysis of the operation data by the fusion algorithm, if the cleaning judgment conditions are met, a "cleaning start command" is generated, which includes the valve opening degree (100%), the dosage (calculated according to the level of fouling, such as 25L / h), and the cleaning time (30 minutes). (3) Command transmission stage: The reverse osmosis membrane automatic cleaning system sends the cleaning start command to the dedicated board. After the board completes signal conversion, verification and isolation, it transmits the command to the DCS system through hard wiring. After receiving the command, the DCS sends back a confirmation receipt. (4) Cleaning execution phase: The DCS system drives the dosing pipeline valves to be fully opened, controls the dosing pump to add scale inhibitor according to the set flow rate, and collects valve status and actual dosing data in real time, and feeds them back to the intelligent operation and maintenance system through the board; (5) Process monitoring stage: The reverse osmosis membrane automatic cleaning system compares the feedback data with the command parameters. If the actual dosage deviation exceeds ±5%, a correction command is sent to adjust the dosing pump frequency of the DCS through the board until the deviation meets the standard. (6) Cleaning stop stage: When the cleaning time reaches the set value, or the reverse osmosis membrane automatic cleaning system detects that the pressure difference and permeate flow rate have returned to the normal range (pressure difference ≤ 110% of the initial value, permeate flow rate ≥ 95% of the initial value), a "cleaning stop command" is generated and sent to the DCS system through the board; (7) Closed-loop closing stage: After receiving the stop command, the DCS system closes the dosing pipeline valve and dosing pump, and sends back the "execution completed" signal. The reverse osmosis membrane automatic cleaning system records the cleaning data (duration, dosing amount, cleaning effect) and updates the fouling judgment model parameters.

[0102] Furthermore, the DCS system adaptation in this embodiment adopts a hard-wired mode. The DCS system side is configured with corresponding analog input / output modules and digital input / output modules, which are hard-wired to a dedicated board (using shielded twisted-pair cable to reduce signal attenuation). A dedicated control program is written in the DCS system to achieve command reception and parsing functions. Receive cleaning commands sent by the board and parse parameters such as valve opening / closing status, dosage, and cleaning duration; Actuator drive: Controls the opening and closing degree (0-100%) of the dosing pipeline valves and the operating frequency of the dosing pump according to instructions, and adjusts the dosing amount; Status feedback: Real-time collection of actual valve opening and closing status, dosing pump operating current, and actual dosing amount detected by flow sensor, and feedback to intelligent operation and maintenance system via board; Emergency shutdown logic: When the DCS system detects an anomaly (such as overload of the dosing pump or valve jamming), it automatically triggers an emergency shutdown command, shuts down the valve and the dosing pump, and sends an alarm signal to the intelligent operation and maintenance system through the board.

[0103] In summary, the decision output layer in this application embodiment can, but is not limited to, be based on the logic of the DCS configuration (such as "membrane pressure difference > 15% and desalination rate < 97%), triggering an automatic cleaning command (replacing manual experience judgment), realizing bidirectional data transmission between the DCS and the reverse osmosis membrane automatic cleaning system (such as the DCS sending a "start cleaning" command, and the cleaning system responding with "current step (rinsing in progress)"); and then through the valve positioner, realize the opening and closing and opening degree control of key valves in the cleaning system (such as cleaning fluid inlet valve, concentrate discharge valve, permeate return valve) (such as adjusting the flushing flow rate), as well as the start and stop and frequency control of cleaning pumps (such as high-pressure flushing pumps) and dosing pumps (such as scale inhibitor pumps and cleaning agent pumps) (adjusting the flow rate through a frequency converter), thereby ensuring that the cleaning fluid circulation volume and dosing volume meet the requirements.

[0104] Additionally, the DCS system in this embodiment can also monitor the concentration of chemicals in the cleaning fluid (such as scale inhibitor concentration and acid / alkali concentration) in real time, and adjust the dosing pump frequency through DCS feedback to ensure that the concentration meets the membrane manufacturer's requirements (such as scale inhibitor concentration of 2-5 ppm).

[0105] Furthermore, to ensure data security during data interaction, embodiments of this application may, but are not limited to, incorporate the following functional mechanisms into the decision output layer: (1) Data encryption mechanism (dual encryption for transmission and storage): Encrypted transmission: The SSL / TLS 1.3 protocol is used to encrypt the transmission of real-time data (such as water quality sensors, flow meters, membrane pressure differentials, etc.) to prevent the data from being stolen or tampered with during network transmission. Encrypted storage: Sensitive data in the database (such as scale inhibitor formulations, membrane life prediction models, and historical cleaning records) are encrypted and stored using the AES-256 symmetric encryption algorithm.

[0106] (2) Fine-grained access control (role-based access control): Role-based hierarchy: System users are divided into four roles: system administrator (highest privileges), operations engineer (operational privileges), water quality analyst (data viewing privileges), and auditor (log viewing privileges). Detailed access control: Adopting the principle of least privilege, precise permissions are assigned to each role (e.g., maintenance engineers can only modify dosing parameters and cannot view the membrane manufacturer's core formula; water quality analysts can only access water quality data and cannot operate equipment). Dynamic permission adjustment: Supports real-time revocation or adjustment of permissions based on changes in user roles (such as resignation or job transfer) to prevent permission leaks.

[0107] (3) Data auditing and anomaly monitoring functions: Operation log traceability: Records all user operations (such as modification of dosing parameters, issuance of cleaning instructions, and data export), including information such as operator, operation time, operation content, and terminal IP. The log retention period is no less than 6 months, and it supports rapid traceability of violations. Abnormal behavior detection: Based on machine learning (ML) models (such as the Isolation Forest algorithm), analyze user access patterns, identify abnormal behaviors (such as exporting large amounts of data outside of working hours, or frequent access to the core database from IP addresses in different locations), and trigger real-time alarms (SMS / system pop-ups) while automatically blocking unauthorized operations.

[0108] This application embodiment can transform abstract cleaning strategies into specific, operable control commands through executable parameters such as cleaning duration, cleaning agent dosage, and valve opening degree, thereby quantifying and automating the cleaning process. By dynamically adjusting these parameters, the system can optimize the cleaning process based on the degree of membrane fouling, water quality conditions, and operating status, improving cleaning efficiency and effectiveness while avoiding over-cleaning or under-cleaning, reducing resource waste and system damage, and achieving precise and economical operation and maintenance management.

[0109] Optionally, in one embodiment of this application, the method further includes: acquiring the water quality parameter change trend of the target reverse osmosis membrane system, the inter-stage pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information based on historical target time operation data and real-time operation data; generating a pollution warning prompt for the target reverse osmosis membrane system based on the water quality parameter change trend, the inter-stage pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information, and feeding back the pollution warning prompt to the user of the target reverse osmosis membrane system.

[0110] As one possible approach, this application can first obtain the water quality parameter change trend of the target reverse osmosis membrane system, the inter-segment pressure drop curve of the reverse osmosis membrane, and the instantaneous concentration change information based on historical operating data and real-time operating data within the target time period.

[0111] Here, the trend of water quality parameters can be understood as the change information of water quality parameters in the reverse osmosis membrane system from the initial moment of the historical target time to the current moment.

[0112] Furthermore, in industrial reverse osmosis systems, to improve recovery rate (product water rate), membrane modules are typically used in series in multiple sections. The concentrate (wastewater that fails to permeate) of one section serves as the feed water for the next section, and so on. The inter-section pressure drop here refers to the pressure loss in each section as the water flows from the feed end to the outlet end (concentrate end). The inter-section pressure drop curve here refers to the trend line formed by the inter-section pressure drop value changing over time (e.g., from the initial moment of the historical target time to the current moment), and the fluctuation information is the fluctuation of this trend line over time.

[0113] Instantaneous concentration change information can be understood here as the change of various ion concentrations in the reverse osmosis system over time (e.g., from the initial moment of the historical target time to the current moment).

[0114] For example, this application may, but is not limited to, incorporate the ion concentration (Ca²) in the reverse osmosis system. + SO4² -Parameters such as pressure differential increase and desalination rate decline trend are used to identify and predict the probability and impact range of risks such as scaling, fouling, and microbial contamination in the reverse osmosis system within the next 24-72 hours. Then, based on machine learning models (such as LSTM and random forest) and statistical analysis, the membrane fouling risk level is dynamically assessed, and graded early warning signals (low risk, medium risk, high risk) are output and sent to users such as maintenance personnel to achieve effective management of the target reverse osmosis system.

[0115] In one embodiment of this application, early warning can be achieved through an early warning triggering and notification mechanism, that is, based on preset thresholds and model output results, a graded early warning can be automatically triggered and the operation and maintenance personnel can be notified through multiple channels (audio-visual alarm, SMS, email, work order).

[0116] The early warning triggering and notification mechanism also supports the following functions: dynamic threshold adjustment, allowing manual setting of thresholds (e.g., a pressure difference increase > 15% triggers medium risk) or adaptive adjustment by the model; multi-level notification strategies, with low risk only logging, medium risk sending SMS / emails, and high risk linking the DCS system and generating emergency work orders; mobile inspection work orders, allowing users to scan codes to obtain membrane module information, input on-site testing data (SDI, residual chlorine, etc.), and automatically link historical maintenance records for real-time alarm pushes; and support for uploading photos of abnormal phenomena (e.g., membrane shell leakage) and triggering emergency procedures. Through the mobile maintenance collaboration system, spatial limitations are broken, and emergency response efficiency is improved.

[0117] Additionally, embodiments of this application may also construct a dynamic knowledge base and case engine module, introduce a fault case map, structurally store membrane fouling characteristics (such as the differential pressure curve pattern of CaSO4 scaling), and recommend treatment solutions for similar historical cases; and add a chemical compatibility database to record experimental data on the compatibility of scale inhibitors / bactericides and provide early warnings of chemical conflict risks (such as the coexistence of sodium bisulfite and oxidizing bactericides).

[0118] This application embodiment can achieve early identification and proactive warning of membrane fouling risks by analyzing water quality parameter change trends, pressure drop curve fluctuations, and instantaneous concentration information. It generates warning prompts and provides feedback to users, helping maintenance personnel to promptly identify potential problems and take preventative measures. By providing warning information through visualization or notification, the system enhances the transparency and response speed of operations and maintenance, helps reduce sudden failures, extends equipment lifespan, and improves the safety and stability of system operation, realizing a shift from a "passive handling" to a "proactive prevention" operation and maintenance model.

[0119] The reverse osmosis membrane cleaning control method proposed in this application can obtain the changing trends of key indicators in the future target time based on the historical and real-time operating data of the target reverse osmosis membrane system, thereby determining the cleaning plan for the target reverse osmosis membrane system and controlling the relevant actuators to execute it. This achieves closed-loop control of the entire process from data acquisition to decision execution by systematically collecting historical and real-time operating data, predicting the future changing trends of key indicators, and formulating cleaning plans accordingly. It transforms the traditional experience-driven or timed cleaning mode into a data-driven intelligent cleaning strategy, which can more accurately identify membrane fouling trends, predict performance degradation, and take targeted cleaning measures before or in the early stages of fouling. This effectively extends membrane life, improves system operating efficiency, reduces unnecessary cleaning frequency and reagent consumption, lowers manual intervention costs, and enhances the overall intelligence and precision of operation and maintenance. Therefore, it solves the problems in related technologies where the cleaning control method for reverse osmosis membranes, which relies on manual experience and fixed cycles, easily leads to delayed cleaning timing, inability to dynamically respond to the real-time state of membrane fouling, resulting in membrane performance degradation, affecting the system's economy and safety, and making it difficult to meet increasingly stringent water quality standards and energy-saving requirements.

[0120] Next, referring to the accompanying drawings, a cleaning control device for a reverse osmosis membrane according to an embodiment of this application is described.

[0121] Figure 5 This is a schematic diagram of the structure of the reverse osmosis membrane cleaning control device according to an embodiment of this application.

[0122] like Figure 5 As shown, the reverse osmosis membrane cleaning control device 10 includes: a data acquisition module 100, a first acquisition module 200, and a control module 300.

[0123] The system includes a data acquisition module 100 for acquiring historical and real-time operating data of the target reverse osmosis membrane system; a first acquisition module 200 for acquiring the changing trend of target indicators in the target reverse osmosis membrane system in the future target time based on the historical and real-time operating data; and a control module 300 for determining a cleaning plan for the target reverse osmosis membrane system based on the changing trend, and controlling the corresponding cleaning actuator of the target reverse osmosis membrane system to execute the cleaning plan.

[0124] Optionally, in one embodiment of this application, the acquisition module 100 includes: an acquisition unit, used to acquire water quality parameters, membrane module operating indicators and module status information of the target reverse osmosis membrane system at a historical target time and at the current time; and a first determination unit, used to determine the historical target time operating data and real-time operating data based on the water quality parameters, membrane module operating indicators and module status information at the historical target time and at the current time.

[0125] Optionally, in one embodiment of this application, the first acquisition module 200 includes: an acquisition unit, used to acquire the temporal dependency, local features and global features of the target indicator based on historical target time running data and real-time running data; and a second determination unit, used to determine the change trend by combining the temporal dependency, local features and global features.

[0126] Optionally, in one embodiment of this application, it further includes: a second acquisition module, used to acquire, based on historical target time operating data and real-time operating data, the water quality parameter change trend of the target reverse osmosis membrane system, the inter-stage pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information; and a generation module, used to generate a pollution warning prompt for the target reverse osmosis membrane system based on the water quality parameter change trend, the inter-stage pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information, and to feed back the pollution warning prompt to the user of the target reverse osmosis membrane system.

[0127] Optionally, in one embodiment of this application, the control module 300 includes: a third determining unit, used to determine the cleaning time, cleaning agent dosage, and valve opening degree of the target reverse osmosis membrane system based on the changing trend; and a fourth determining unit, used to determine the cleaning plan based on the cleaning time, cleaning agent dosage, and valve opening degree.

[0128] It should be noted that the foregoing explanation of the reverse osmosis membrane cleaning control method embodiment also applies to the reverse osmosis membrane cleaning control device of this embodiment, and will not be repeated here.

[0129] The reverse osmosis membrane cleaning control device proposed in this application can obtain the changing trends of key indicators in the future target time based on the historical and real-time operating data of the target reverse osmosis membrane system, thereby determining the cleaning plan for the target reverse osmosis membrane system and controlling the relevant actuators to execute it. This achieves closed-loop control of the entire process from data acquisition to decision execution by systematically collecting historical and real-time operating data, predicting the future changing trends of key indicators, and formulating cleaning plans accordingly. It transforms the traditional experience-driven or timed cleaning mode into a data-driven intelligent cleaning strategy, which can more accurately identify membrane fouling trends, predict performance degradation, and take targeted cleaning measures before or in the early stages of fouling. This effectively extends membrane life, improves system operating efficiency, reduces unnecessary cleaning frequency and reagent consumption, lowers manual intervention costs, and enhances the overall intelligence and precision of operation and maintenance. Therefore, it solves the problems in related technologies where the cleaning control method for reverse osmosis membranes, which relies on manual experience and fixed cycles, easily leads to delayed cleaning timing, inability to dynamically respond to the real-time state of membrane fouling, resulting in membrane performance degradation, affecting the system's economy and safety, and making it difficult to meet increasingly stringent water quality standards and energy-saving requirements.

[0130] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 601, the processor 602, and the computer program stored on the memory 601 and capable of running on the processor 602.

[0131] When the processor 602 executes the program, it implements the reverse osmosis membrane cleaning control method provided in the above embodiments.

[0132] Furthermore, electronic devices also include: Communication interface 603 is used for communication between memory 601 and processor 602.

[0133] The memory 601 is used to store computer programs that can run on the processor 602.

[0134] The memory 601 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0135] If the memory 601, processor 602, and communication interface 603 are implemented independently, then the communication interface 603, memory 601, and processor 602 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 6 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0136] Optionally, in a specific implementation, if the memory 601, processor 602, and communication interface 603 are integrated on a single chip, then the memory 601, processor 602, and communication interface 603 can communicate with each other through an internal interface.

[0137] The processor 602 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0138] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described reverse osmosis membrane cleaning control method.

[0139] This application also provides a computer program product, including a computer program that can run computer instructions. When the computer instructions are executed by a processor, they implement the reverse osmosis membrane cleaning control method provided in this application.

[0140] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0141] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0142] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0143] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0144] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0145] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0146] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0147] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A method for cleaning and controlling a reverse osmosis membrane, characterized in that, Includes the following steps: Collect historical and real-time operating data of the target reverse osmosis membrane system; Based on the historical target time operation data and the real-time operation data, the changing trend of the target indicators in the target reverse osmosis membrane system in the future target time is obtained; Based on the changing trend, a cleaning scheme for the target reverse osmosis membrane system is determined, and the cleaning actuator corresponding to the target reverse osmosis membrane system is controlled to execute the cleaning scheme according to the cleaning scheme.

2. The method according to claim 1, characterized in that, The collected target reverse osmosis membrane system's operational data at historical target times and real-time operational data include: Collect water quality parameters, membrane module operating indicators, and module status information of the target reverse osmosis membrane system at the historical target time and the current time; Based on the historical target time and the water quality parameters, membrane module operation indicators and module status information at the current time, the operation data at the historical target time and the real-time operation data are determined.

3. The method according to claim 1, characterized in that, The process of obtaining the changing trend of target indicators in the target reverse osmosis membrane system over future target times based on the historical target time operating data and the real-time operating data includes: Based on the historical target time operation data and the real-time operation data, the temporal dependency, local features and global features of the target indicator are obtained; The change trend is determined by combining the temporal dependency, local features, and global features.

4. The method according to claim 1, characterized in that, Also includes: Based on the historical target time operation data and the real-time operation data, the water quality parameter change trend of the target reverse osmosis membrane system, the inter-segment pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information are obtained. Based on the water quality parameter change trend, the inter-segment pressure drop curve fluctuation information of the target reverse osmosis membrane system, and the instantaneous concentration change information, a pollution warning prompt for the target reverse osmosis membrane system is generated, and the pollution warning prompt is fed back to the user of the target reverse osmosis membrane system.

5. The method according to claim 1, characterized in that, The step of determining the cleaning scheme for the target reverse osmosis membrane system based on the changing trend includes: Based on the changing trend, determine the cleaning time, cleaning agent dosage, and valve opening degree of the target reverse osmosis membrane system; The cleaning plan is determined based on the cleaning duration, cleaning agent dosage, and valve opening degree.

6. A cleaning control device for a reverse osmosis membrane, characterized in that, include: The data acquisition module is used to collect historical and real-time operating data of the target reverse osmosis membrane system. The acquisition module is used to acquire the changing trend of the target indicators in the target reverse osmosis membrane system in the future target time based on the historical target time operation data and the real-time operation data; The control module is used to determine the cleaning scheme of the target reverse osmosis membrane system based on the changing trend, and to control the cleaning actuator corresponding to the target reverse osmosis membrane system to execute the cleaning scheme according to the cleaning scheme.

7. The apparatus according to claim 6, characterized in that, The acquisition module includes: The data acquisition unit is used to acquire water quality parameters, membrane module operating indicators, and module status information of the target reverse osmosis membrane system at the historical target time and the current time. The determining unit is used to determine the operating data at the historical target time and the real-time operating data based on the water quality parameters, membrane module operating indicators and module status information at the current time.

8. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the reverse osmosis membrane cleaning control method as described in any one of claims 1-5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the reverse osmosis membrane cleaning control method as described in any one of claims 1-5.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed, it is used to implement the reverse osmosis membrane cleaning control method as described in any one of claims 1-5.

Images