A medical multi-stage pure water integrated method and system based on modular design

Abstract

The present application relates to the technical field of medical pure water preparation, and specifically discloses a medical multi-stage pure water integrated method and system based on modular design, aiming at the problems that the existing reverse osmosis membrane is interfered by water temperature fluctuation, leading to pollution misdiagnosis, passive alarm and scheduled maintenance easily causing unplanned shutdown, by collecting parameters such as water inlet temperature, membrane section pressure difference and water production flow, completing standardized processing through temperature-viscosity compensation, identifying membrane fouling deposition and element damage failure through parameter change slope decoupling, predicting the remaining health running time of the membrane through linear regression fitting, automatically reducing the frequency of the high-pressure pump within the early warning threshold to implement preventive intervention, and realizing closed-loop self-correction of the diagnosis threshold and the data window based on the quantitative calculation of the intervention effect. The present application can strip water temperature interference to realize accurate pollution typing diagnosis, predict and actively delay pollution development in advance, eliminate unplanned shutdown from the root, control the adaptive evolution of the model, and ensure the continuous and stable operation of the medical pure water system.

Description

Technical Field

[0001] This invention relates to the field of medical pure water preparation technology, and in particular to a modular design-based integrated method and system for medical multi-stage pure water preparation. Background Technology

[0002] Medical pure water systems are critical water supply infrastructure for core departments in hospitals, such as disinfection supply centers, laboratories, and hemodialysis centers. Their continuous and stable operation is directly related to the safety of medical treatment. To adapt to the compact space and flexible water demand of hospitals, existing technologies highly integrate pretreatment, reverse osmosis, electro-deionization, and intelligent control modules into a single cabinet, forming a modular multi-stage medical pure water integrated device. Among them, the reverse osmosis module is the core unit for achieving water desalination and purification.

[0003] During long-term operation, reverse osmosis membranes inevitably experience fouling problems such as inorganic salt scaling and colloidal particle deposition. Meanwhile, the temperature of municipal tap water inlet fluctuates between 5°C and 30°C depending on the season. Water temperature changes alter the dynamic viscosity of the water, directly causing normal physical fluctuations in the pressure difference and permeate flow rate of the reverse osmosis membrane section. These normal fluctuations are coupled with the abnormal performance degradation caused by membrane fouling, making it impossible to distinguish between water temperature interference and actual membrane fouling based on the original operating parameters.

[0004] Current membrane fouling control in medical pure water systems relies solely on passive over-limit alarms or fixed-cycle periodic maintenance. Passive alarms are triggered only after membrane performance exceeds limits, by which time fouling has already reached a severe stage, easily leading to unplanned shutdowns and interruptions to the medical water supply. Periodic maintenance cannot match the actual fouling rate; in scenarios with good water quality, it results in over-maintenance and wasted chemicals, while in scenarios with poor water quality, maintenance is delayed and fails to curb fouling. Furthermore, the diagnostic thresholds and predictive parameters of existing control methods are factory-set and lack self-correction capabilities based on actual operating results, leading to a continuous decline in diagnostic and predictive accuracy over long-term operation. These shortcomings prevent current technologies from achieving early and accurate identification and proactive prevention of membrane fouling, resulting in unplanned shutdowns of medical pure water systems.

[0005] Therefore, there is an urgent need for a modular design-based integrated method and system for medical multi-stage pure water to solve the above problems. Summary of the Invention

[0006] The purpose of this invention is to provide a modular design-based integrated method for medical multi-stage pure water systems, comprising the following steps: The operating parameters of the first-stage reverse osmosis module are obtained, and the operating parameters are standardized by temperature compensation to obtain standardized operating parameters. Based on the operating parameters, the membrane performance index is calculated. Based on the changing trend characteristics of the standardized operating parameters within a preset time window, the membrane fouling state type is identified according to preset diagnostic rules. Only when the identification result is a preventable type of pollution, the performance degradation trend is generated using the historical data sequence of the standardized operating parameters to predict the remaining healthy operating time of the membrane; When the remaining healthy operating time of the membrane meets the preset warning conditions, a control command is generated to adjust the operating status of the first-stage reverse osmosis module and a warning message is pushed. After the control operation is performed or the membrane maintenance operation is completed, the operation data after the intervention is obtained, the intervention effect index is calculated, and the intervention effect index is used to perform closed-loop correction on the judgment parameters in the diagnostic rules and the data window parameters used to generate the performance degradation trend.

[0007] Furthermore, this invention also discloses a modularly designed medical multi-stage pure water integrated system, comprising: The acquisition submodule is used to acquire the operating parameters of the first-stage reverse osmosis module, perform temperature compensation standardization processing on the operating parameters to obtain standardized operating parameters, and calculate membrane performance indicators based on the operating parameters. The identification submodule is used to identify the membrane fouling state type based on the changing trend characteristics of the standardized operating parameters within a preset time window and according to preset diagnostic rules. The prediction submodule is used to generate a performance degradation trend and predict the remaining healthy operating time of the membrane only when the identification result is a preventable type of pollution. A generation submodule is used to generate control commands to adjust the operating status of the first-stage reverse osmosis module and push early warning information when the remaining healthy operating time of the membrane meets the preset early warning conditions. The correction submodule is used to acquire post-intervention operational data after the control operation is performed or the membrane maintenance operation is completed, calculate the intervention effect index, and use the intervention effect index to perform closed-loop correction on the judgment parameters in the diagnostic rules and the data window parameters used to generate the performance degradation trend.

[0008] Furthermore, the correction submodule includes: The first calculation unit is used to collect standardized operating parameters for the period after the intervention after the control operation is executed, and to calculate the rate of change of performance degradation before and after the intervention as the degradation mitigation efficiency. The second calculation unit is used to collect standardized operating parameters for the period after membrane maintenance is completed, compare them with the parameters recorded before maintenance, and calculate the membrane performance recovery rate. The adjustment unit is used to adjust the judgment threshold in the diagnostic rules and the data window parameters used to generate the performance degradation trend according to the comparison result between the degradation mitigation efficiency or the membrane performance recovery rate and the corresponding preset threshold, and to update the adjusted parameters.

[0009] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described modular design-based integrated method for medical multi-stage pure water.

[0010] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described modular design-based integrated method for medical multi-stage pure water.

[0011] The beneficial effects of this application are as follows: Firstly, this invention completely eliminates the physical interference of water temperature fluctuations on operating parameters through temperature-viscosity standardization compensation. Combined with multi-parameter slope trend decoupling, it achieves accurate classification and diagnosis of membrane fouling and membrane element damage. It can eliminate misjudgment and missed judgment caused by interference coupling from the data source, and significantly improve the accuracy and reliability of membrane fouling identification.

[0012] Secondly, this invention fits the membrane performance degradation trend through linear regression and extrapolates the remaining healthy operating time. It automatically triggers preventive frequency reduction intervention before the membrane performance reaches the minimum water consumption threshold, actively slows down the rate of membrane fouling, and reserves sufficient operating window for manual maintenance. This eliminates unplanned shutdowns of medical pure water systems from the source and ensures the continuity and safety of medical water supply.

[0013] Third, based on the actual effects of active intervention and manual cleaning, this invention can quantify and calculate the attenuation reduction efficiency and membrane performance recovery rate, realize dynamic closed-loop self-correction of diagnostic threshold and prediction data window, enable the control model to continuously adapt to the on-site raw water quality and membrane element aging conditions, maintain high accuracy of membrane fouling diagnosis and prediction throughout the entire life cycle, reduce system maintenance costs, and improve the operational economy and stability of modular medical pure water devices. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of a method flow proposed in an embodiment of this application.

[0015] Figure 2 This is a schematic diagram of the system structure proposed in an embodiment of the present invention.

[0016] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0018] like Figure 1 As shown, this application provides a modular design-based integrated method for medical multi-stage pure water systems, applied to a medical pure water device that integrates a pretreatment module, a first-stage reverse osmosis module, and an intelligent control module. The method includes the following steps: S1. Obtain the feed water temperature, membrane pressure difference, permeate flow rate, feed water conductivity, and permeate conductivity of the first-stage reverse osmosis module. Use the temperature-viscosity compensation relationship pre-existing in the intelligent control module to standardize the membrane pressure difference and permeate flow rate to obtain standardized membrane pressure difference and standardized permeate flow rate. Calculate the real-time desalination rate based on the feed water conductivity and permeate conductivity. S2. Based on the slopes of the changes in standardized membrane pressure difference, standardized permeate flow rate, and real-time desalination rate within a preset sliding time window, the membrane fouling status is identified according to the established diagnostic rules: When the standardized membrane pressure differential slope is greater than 0, the standardized permeate flow rate slope is less than 0, and the absolute value of the real-time desalination rate slope is less than the preset small threshold, it is determined that the membrane has scaled or deposited particles. When the standardized membrane pressure difference and the standardized permeate flow rate slope do not meet the above conditions, but the real-time desalination rate slope is less than 0 and its absolute value exceeds the preset desalination attenuation threshold, the membrane element is determined to be damaged or leaking. S3. Only when step S2 determines that scaling or particle deposition occurs, the historical sequence data of the standardized permeate flow rate in the annular buffer is used to fit the attenuation trend line using linear regression, and the intersection of the trend line and the preset lower limit of the standardized permeate flow rate is extrapolated. The difference between the intersection time and the current time is taken as the remaining healthy operating time of the membrane. S4. Compare the remaining healthy operating time of the membrane with the preset warning lead time. If the remaining healthy operating time of the membrane is less than or equal to the warning lead time, automatically send an instruction to the reverse osmosis high-pressure pump frequency converter to reduce the current operating frequency by a preset step size to slow down the fouling rate. At the same time, push a preventive maintenance warning containing the fouling type and the predicted remaining time through the human-machine interface. S5. After the frequency reduction action is executed or the membrane cleaning is completed and restarted manually according to the warning, standardized permeate flow data within a fixed period after the intervention is collected, the change ratio of the attenuation slope before and after the intervention is calculated to evaluate the attenuation mitigation efficiency, or the recovery rate of the standardized permeate flow before and after cleaning is calculated. The attenuation mitigation efficiency or recovery rate is used to automatically correct the slope threshold in the diagnostic rule and the length of the historical data window on which the linear regression in step S3 is based, so as to achieve closed-loop self-correction.

[0019] As described in steps S1-S5 above, the primary reverse osmosis module of the medical pure water system is the core desalination unit, and its operational stability directly determines the safety of the medical water supply. The inlet temperature of municipal tap water fluctuates between 5 and 30 degrees Celsius depending on the season. Water temperature changes alter the dynamic viscosity of the water; lower temperatures increase viscosity, leading to a passive increase in membrane pressure differential and a passive decrease in permeate flow rate under the same permeate demand—a normal physical change. This normal fluctuation is completely coupled with abnormal performance degradation caused by membrane scaling, particle deposition, and membrane element damage, making it impossible to distinguish between normal fluctuations and actual faults using original parameters. This results in misdiagnosis of membrane fouling and inaccurate prediction of remaining healthy operating time. Furthermore, medical settings have extremely high requirements for water supply continuity; any unplanned shutdown will affect the treatment process.

[0020] Existing medical pure water control methods only directly collect raw operating parameters without implementing standardized compensation for water temperature fluctuations, making it impossible to distinguish between physical interference and actual faults. Most methods use single-parameter instantaneous threshold judgment or timed maintenance strategies, which are passive responses after a fault occurs. They cannot predict membrane failure time in advance, nor can they actively control and delay the development of contamination. The diagnostic thresholds and prediction parameters are all fixed settings at the factory and do not have closed-loop self-correction capabilities. After long-term operation, the deviation between the model and the actual membrane performance continues to widen, which can easily lead to false alarms and missed alarms, and cannot meet the core requirement of zero unplanned downtime for medical pure water systems.

[0021] This invention employs a synergistic control approach, encompassing standardized operating parameter processing, decoupling and identifying membrane fouling trends, predicting remaining healthy operating time of the membrane, proactively reducing frequency for preventative intervention, and a closed-loop self-correction mechanism for intervention effects. This approach completely eliminates the physical interference of influent temperature fluctuations on membrane operating parameters, accurately identifies membrane fouling types and predicts performance failure points, proactively adjusts operating parameters to slow fouling development, and dynamically corrects diagnostic and predictive model parameters. Ultimately, it achieves advanced prevention and control of membrane fouling in medical pure water systems, eliminating unplanned downtime at its source. Through data standardization to remove interference, multi-parameter slope decoupling to identify faults, linear regression to extrapolate remaining operating time, proactive frequency reduction for preventative intervention, and a closed-loop control logic for quantitative self-correction of intervention effects, it systematically solves the entire chain of technical problems, including water temperature coupling interference, misdiagnosis of fouling, delayed early warning, lack of proactive intervention, and poor parameter adaptability. This provides medical pure water systems with adaptive membrane fouling control capabilities throughout their entire lifecycle.

[0022] This invention uses an intelligent control module as its core to construct a complete adaptive control closed loop encompassing data standardization, fault identification, predictive intervention, and closed-loop correction. First, temperature-viscosity compensation removes physical disturbances caused by influent temperature, yielding pure membrane performance indicators. Second, multi-parameter slope trend decoupling accurately distinguishes between membrane fouling and membrane element damage. For preventable fouling faults, linear regression is used to extrapolate the membrane performance failure critical time, quantifying the remaining healthy operating time. When the remaining time reaches the warning threshold, the high-pressure pump operating frequency is automatically reduced to slow the fouling rate, and precise maintenance guidance is simultaneously pushed. Finally, based on the actual effects of proactive frequency reduction or manual cleaning, the intervention efficiency is quantified and the diagnostic threshold, data window, and other core parameters are dynamically corrected, ensuring the control model continuously adapts to field conditions. The entire process requires no manual intervention, achieving proactive prevention of membrane fouling and long-term stable system operation.

[0023] In one embodiment, the standardization of membrane segment pressure difference and permeate flow rate, and the calculation of real-time desalination rate in step S1, specifically includes: The S1.1 intelligent control module collects the inlet water temperature from the inlet water temperature sensor at fixed time intervals, collects the pressure values ​​from the pressure sensors at the inlet and concentrate outlets of each section of the first-stage reverse osmosis module, calculates the difference between the inlet water pressure and the concentrate pressure as the membrane section pressure difference, collects the product water flow rate from the product water pipe flow meter, collects the inlet water conductivity from the inlet water conductivity sensor, and collects the product water conductivity from the product water conductivity sensor. The S1.2 module calls a pre-calibrated temperature-viscosity compensation lookup table in its internal non-volatile memory. This table records the membrane permeate flux correction factor and membrane pressure difference correction factor corresponding to the dynamic viscosity of water at different water temperatures. Using the current feed water temperature as an index, the module reads the membrane permeate flux correction factor and membrane pressure difference correction factor corresponding to the current temperature from the lookup table. The membrane pressure difference is multiplied by the membrane pressure difference correction factor to obtain the standardized membrane pressure difference, and the permeate flow rate is multiplied by the membrane permeate flux correction factor to obtain the standardized permeate flow rate. S1.3 Calculate the ratio of product water conductivity to influent conductivity, subtract this ratio from 1 and then multiply by 100% to obtain the real-time desalination rate; S1.4 stores the calculated standardized membrane pressure difference, standardized permeate flow rate, and real-time desalination rate, along with the timestamps, into a pre-defined circular data buffer for subsequent trend analysis.

[0024] As described in steps S1.1-S1.4 above, by collecting raw operating parameters at fixed intervals, performing temperature-viscosity standardization compensation, calculating the desalination rate in real time, and caching time-series data with timestamps, the physical interference of feed water temperature fluctuations on the operating parameters of the first-stage reverse osmosis module is completely eliminated. Standardized indicators that only reflect the performance of the membrane itself are generated, providing accurate input for subsequent membrane fouling identification and prediction of remaining healthy operating time. This ensures accurate membrane fouling diagnosis from the data source and supports the medical pure water system to achieve zero unplanned downtime.

[0025] The inlet temperature of municipal tap water fluctuates between 5°C and 30°C depending on the season. Temperature changes alter the dynamic viscosity of the water; lower temperatures increase viscosity, leading to a passive increase in membrane pressure differential and a passive decrease in permeate flow rate under the same permeate demand. These normal parameter fluctuations caused by water temperature can couple with abnormal increases in pressure differential and abnormal decreases in flow rate due to membrane fouling. It becomes impossible to distinguish between normal fluctuations and true membrane fouling using raw parameters, resulting in misdiagnosis of membrane fouling and inaccurate prediction of remaining healthy operating time. This fails to meet the core requirement of continuous and stable water supply for medical pure water systems. Therefore, it is essential to standardize the raw operating parameters to remove water temperature interference and accurately characterize the membrane's performance.

[0026] Current medical pure water control methods only directly collect parameters such as raw membrane pressure difference and permeate flow rate, without compensating for water temperature fluctuations. In low-temperature winter scenarios, normal parameter changes caused by water temperature may be misjudged as membrane fouling, leading to excessive maintenance and waste of chemicals. In scenarios with poor water quality, parameter interference may cause true membrane fouling to be missed, making early and accurate identification of membrane fouling impossible. A comprehensive solution, employing fixed-period full-scale parameter collection, standardized calculation using pre-stored compensation tables, real-time desalination rate calculation, and caching of time-series data, specifically addresses the technical challenges of water temperature interference and the inability to directly use raw parameters for membrane fouling diagnosis.

[0027] The intelligent control module performs data acquisition operations at fixed time intervals. It acquires the feed water temperature from the feed water temperature sensor, the pressure values ​​from the pressure sensors at the inlet and concentrate outlets of each section of the first-stage reverse osmosis module and calculates the membrane section pressure difference, the product water flow rate from the product water flow meter, and the feed water conductivity and product water conductivity from the feed water conductivity sensor. This completes the acquisition of all the basic raw parameters required for membrane fouling diagnosis, ensuring the integrity and reliability of the input data for subsequent calculations and providing data support for standardized processing.

[0028] The intelligent control module calls the pre-calibrated temperature-viscosity compensation lookup table in the internal non-volatile memory. This lookup table fully records the membrane permeate flux correction coefficient and membrane pressure difference correction coefficient corresponding to the dynamic viscosity of water at different water temperatures. Using the current feed water temperature as an index, the corresponding correction coefficient is read. The membrane pressure difference is multiplied by the membrane pressure difference correction coefficient to obtain the standardized membrane pressure difference, and the permeate flow rate is multiplied by the membrane permeate flux correction coefficient to obtain the standardized permeate flow rate.

[0029] For a specific example, when the influent temperature is 8℃, the system reads a membrane pressure differential correction factor of 0.75 and a membrane permeate flux correction factor of 1.2. The original membrane pressure differential of 1.1 bar is calculated to yield a standardized membrane pressure differential of 0.825 bar, and the original permeate flow rate of 480 L / h is calculated to yield a standardized permeate flow rate of 576 L / h. The above steps, through temperature-viscosity compensation, completely eliminate the physical interference of water temperature on membrane operating parameters, ensuring that the standardized parameters only reflect changes in the membrane's intrinsic performance, thus eliminating the core interfering factor in membrane fouling diagnosis.

[0030] The intelligent control module calculates the ratio of the product water conductivity to the influent conductivity, subtracts this ratio from 1, and then multiplies it by 100% to obtain the real-time desalination rate. The specific formula for calculating the real-time desalination rate is as follows: ; Among them, the The real-time desalination rate is used to characterize the real-time desalination performance of the reverse osmosis membrane. Indicates the conductivity of the produced water, the The influent conductivity and real-time desalination rate are core indicators for judging membrane element damage or sealing leakage. They can then accurately calculate the membrane's desalination performance parameters, providing key evidence for multi-dimensional identification of membrane fouling status and improving the indicator system for membrane fouling diagnosis.

[0031] The intelligent control module stores the calculated standardized membrane pressure difference, standardized permeate flow rate, and real-time desalination rate, along with timestamps, into a pre-defined circular data buffer. This circular data buffer continuously stores historical operating data, providing continuous time-series data support for subsequent parameter slope calculations and remaining healthy operating time predictions within a pre-defined sliding time window, ensuring the continuity and accuracy of trend analysis.

[0032] Through the coordinated operation of steps S1.1-S1.4, standardized preprocessing of membrane fouling monitoring indicators can be completed. The standardized operation throughout the entire process ensures accurate data collection, reliable compensation logic, and effective caching mechanism. This addresses the diagnostic misjudgment problem caused by the coupling of water temperature fluctuations and membrane fouling from the data source, providing a solid data foundation for subsequent membrane fouling status identification, predictive intervention, and closed-loop self-correction. Ultimately, this supports the core goal of achieving zero unplanned downtime in medical pure water systems.

[0033] In one embodiment, the step S2 of identifying the membrane fouling status based on the slope of the changes in standardized membrane pressure difference, standardized permeate flow rate, and real-time desalination rate within a preset sliding time window specifically includes: S2.1 Extract the standardized membrane pressure difference sequence, standardized permeate flow rate sequence, and real-time desalination rate sequence of the most recent N sampling points from the annular data buffer. Fit a straight line for each sequence using least squares linear regression. Use the slope of the fitted line as the slope of the change of each parameter within the preset sliding time window, and denoted as the standardized membrane pressure difference slope, the standardized permeate flow rate slope, and the real-time desalination rate slope, respectively. S2.2 compares the three calculated slopes with the preset thresholds in the internal curing diagnostic rule table one by one, and performs the judgment according to the following logic: If the standardized membrane pressure differential slope is greater than 0, the standardized permeate flow rate slope is less than 0, and the absolute value of the real-time desalination rate slope is less than the preset threshold for small changes in desalination rate, then a diagnostic indicator of membrane scaling or particle deposition will be output. If the standardized membrane pressure differential slope and the standardized permeate flow rate slope do not simultaneously meet the above conditions, but the real-time desalination rate slope is less than 0 and its absolute value is greater than the preset desalination rate attenuation threshold, then output a diagnostic sign of membrane element damage or seal leakage. If neither of the above two rules is met, the diagnostic marker remains normal. S2.3 When the diagnosis result is membrane scaling or particle deposition, the current standardized permeate flow rate slope, standardized membrane pressure differential slope, and diagnostic flag are output to the control logic corresponding to steps S3 and S4. When the diagnosis result is membrane element damage or seal leakage, step S3 is not executed. Instead, a high-priority alarm requiring immediate inspection of membrane integrity is issued directly through the human-machine interface and recorded in the operation log.

[0034] As described in steps S2.1-S2.3 above, by extracting the slope of parameter changes within a preset sliding time window and combining it with the solidified two-dimensional diagnostic rules, two types of fault states are distinguished: membrane fouling or particle deposition, and membrane element damage or sealing leakage. This provides an accurate basis for subsequent targeted predictive intervention or emergency alarm, and avoids misjudgment and omission of membrane fouling in the diagnostic process.

[0035] The standardized parameters have completely eliminated the physical interference of feed water temperature fluctuations and can only reflect the performance changes of the primary reverse osmosis membrane itself. Membrane fouling and membrane damage will exhibit completely different parameter change characteristics. Membrane scaling or particle deposition will block the membrane flow channels and increase the water flow resistance, which is manifested as a continuous increase in standardized membrane differential pressure and a continuous decrease in standardized permeate flow rate, with no significant change in real-time desalination rate. Membrane element damage or seal leakage will directly destroy the membrane desalination structure, which is manifested as a significant decrease in real-time desalination rate, with no corresponding trend in standardized membrane differential pressure and standardized permeate flow rate. The handling methods for the two types of faults are completely different, and they must be accurately distinguished by characteristic slopes.

[0036] Current membrane fault diagnosis in medical pure water systems relies solely on single-parameter instantaneous threshold judgments, failing to consider the combined assessment of multiple parameter trends. This makes it impossible to distinguish between minor fluctuations in standardized parameters and actual fault trends. Consequently, it easily misdiagnoses normal operating fluctuations as membrane fouling, cannot identify minor membrane fouling in its early stages, and cannot differentiate between membrane fouling and membrane rupture. This leads to incorrect maintenance measures and escalation of faults, failing to meet the core requirement of zero unplanned downtime for medical pure water systems. A comprehensive solution, employing time-series data slope fitting, dual-rule threshold comparison, and categorized fault handling, specifically addresses the technical problems of misdiagnosis based on single parameters and the inability to distinguish fault types.

[0037] The intelligent control module extracts the standardized membrane pressure differential sequence, standardized permeate flow rate sequence, and real-time desalination rate sequence from the N most recent sampling points within the circular data buffer. It then performs linear regression fitting on each sequence using the least squares method. The slope of the fitted line is used as the slope of the corresponding parameter's change within a preset sliding time window, denoted as the standardized membrane pressure differential slope, standardized permeate flow rate slope, and real-time desalination rate slope, respectively. This step eliminates the interference of instantaneous data fluctuations through linear regression fitting, quantifies the stable change trend of parameters over continuous periods, and provides stable characteristic evidence for fault diagnosis.

[0038] The intelligent control module compares the calculated slopes of the three parameters one by one with the preset thresholds in the internally solidified diagnostic rule table, and executes the judgment according to fixed logic. When the standardized membrane pressure differential slope is greater than 0, the standardized permeate flow rate slope is less than 0, and the absolute value of the real-time desalination rate slope is less than the preset threshold for minute changes in desalination rate, it is judged as membrane scaling or particle deposition. When the above conditions are not met, but the real-time desalination rate slope is less than 0 and its absolute value is greater than the preset threshold for desalination rate decay, it is judged as membrane element damage or seal leakage. If neither condition is met, it is judged as a normal state. For example, if the standardized membrane pressure differential slope increases by 0.02 bar per week, the standardized permeate flow rate slope decreases by 1.5% per week, and the absolute value of the real-time desalination rate slope is less than the preset threshold, the system judges that membrane scaling or particle deposition has occurred. Through dual-rule, multi-parameter joint judgment, the fault type can be accurately distinguished, avoiding the risk of misjudgment from single-parameter judgment.

[0039] When the diagnosis result is membrane fouling or particle deposition, the system outputs the current parameter slope and diagnostic flags to subsequent prediction and intervention steps. When the diagnosis result is membrane element damage or seal leakage, the system does not execute the remaining healthy operating time prediction step, but directly issues a high-priority alarm through the human-machine interface requiring immediate inspection of membrane integrity, and records the fault information to the operation log. By classifying and executing according to the severity of the fault and the handling logic, membrane fouling can be preventively intervened, and membrane damage requires immediate shutdown and repair, ensuring the safety of the medical pure water system's water supply. By collaboratively identifying and handling membrane fouling status, quantifying trends through slope fitting, accurately classifying faults using dual rules, and ensuring safety through classified handling, the system addresses the problems of misjudgment, missed judgment, and failure to classify faults in existing technologies from the diagnostic stage. This provides an accurate judgment basis for subsequent predictive intervention and closed-loop self-correction, continuously supporting the core goal of achieving zero unplanned downtime in medical pure water systems.

[0040] In one embodiment, step S3, which involves fitting a decline trend line using standardized historical permeate flow data and extrapolating to predict the remaining healthy operating time, specifically includes: S3.1 Extract the standardized permeate flow data sequence of the M nearest sampling points within the annular buffer, use the sampling number as the independent variable, and fit the sequence using the linear regression method to obtain the slope parameter and intercept parameter of the decay trend line. The slope parameter represents the decay rate of the standardized permeate flow and is negative. S3.2 Read the preset standardized water production flow rate lower limit threshold. This threshold is a fixed value determined based on the minimum water production flux that meets the minimum clinical water demand. Let the standardized water production flow rate in the expression of the decay trend line be equal to this lower limit threshold. Substitute the slope parameter and intercept parameter to obtain the corresponding sampling number. Convert the sampling number into the corresponding time and use it as the critical time when the water production flow rate decays to the lower limit threshold. S3.3 Subtract the current time from the critical time to obtain the remaining healthy operating time of the membrane. If the calculated remaining healthy operating time of the membrane is negative, set it to 0 and directly trigger the highest level warning.

[0041] As described in steps S3.1-S3.3 above, by fitting the trend of water flow rate decline through linear regression and extrapolating the critical moment when the flow rate declines to the lower limit threshold, the remaining healthy operating time of the membrane is quantitatively calculated. This provides an accurate time basis for subsequent preventive frequency reduction intervention, ensuring that sufficient maintenance window is reserved before the membrane performance can no longer meet the minimum clinical water demand, and preventing unplanned shutdowns of the medical pure water system from the prediction stage.

[0042] After scaling or particle deposition occurs on the primary reverse osmosis membrane, the standardized permeate flow rate will show a continuous and stable linear decline trend. Medical pure water devices must ensure that the permeate flow rate is not lower than the lower limit threshold required to meet the minimum clinical water demand. If an alarm is only triggered after the flow rate reaches the threshold, the membrane fouling has already developed to a severe stage, which will directly trigger an unplanned shutdown. Therefore, it is necessary to predict in advance when the flow rate will reach the threshold based on the historical decay rate to provide quantitative support for preventive intervention.

[0043] Existing medical pure water control technologies rely solely on instantaneous flow threshold alarms or fixed-cycle timed maintenance. Instantaneous threshold alarms are reactive responses to faults, failing to allow for advance maintenance time. Timed maintenance cannot match the actual membrane fouling rate, leading to over-maintenance in scenarios with good water quality and delayed maintenance in scenarios with poor water quality. Neither approach can accurately predict the remaining healthy operating time of the membrane, making it difficult to meet the core requirement of zero unplanned downtime in medical pure water systems. This paper proposes a comprehensive solution that addresses the technical problems of delayed membrane performance degradation prediction and lack of early warning by using linear regression to fit the degradation trend, extrapolating the critical flow time, and quantitatively calculating the remaining time.

[0044] The intelligent control module extracts the standardized permeate flow rate data sequence from the most recent M sampling points within the annular buffer. Using the sampling sequence number as the independent variable, a linear regression method is employed to fit this sequence, yielding the slope and intercept parameters of the decay trend line. The slope parameter, being negative, characterizes the decay rate of the standardized permeate flow rate. Linear regression fitting eliminates instantaneous data fluctuations, accurately quantifying the stable decay pattern of the membrane permeate flow rate and providing a mathematical model basis for predicting the remaining time. For example, the system extracts standardized permeate flow rate data from the most recent 90 days, and the fitted decay trend line shows a slope decreasing by 1.5% per week. This slope accurately reflects the permeate flow rate decay rate under membrane fouling.

[0045] The intelligent control module reads a preset standardized permeate flow rate lower limit threshold, which is determined based on the minimum permeate flux required to meet basic clinical water needs. The standardized permeate flow rate in the decay trend line expression is assigned to this lower limit threshold. Substituting the slope and intercept parameters obtained from the fitting, the corresponding sampling number is calculated and converted to a time. This time is the critical moment when the permeate flow rate decays to the lower limit threshold. If the calculated remaining healthy operating time of the membrane is negative, it is directly set to 0 and the highest level warning is triggered. The critical time point for membrane performance failure is determined through mathematical extrapolation, providing a benchmark for calculating the remaining operating time.

[0046] The intelligent control module subtracts the current time from the critical moment to obtain the remaining healthy operating time of the membrane. This step directly quantifies the remaining stable operating time of the membrane, providing core judgment parameters for subsequent comparison with the early warning period and triggering preventive frequency reduction intervention. The above steps can work together to accurately predict the remaining healthy operating time of the membrane. Linear regression fitting ensures accurate quantification of the decay trend, extrapolation calculation ensures reliable determination of the critical moment, and the remaining time calculation provides a quantitative basis for preventive intervention. This addresses the problem of lagging early warning in existing technologies from the prediction stage, lays the foundation for subsequent proactive regulation, and continuously supports the core goal of achieving zero unplanned downtime in medical pure water systems.

[0047] In one embodiment, the automatic reduction of the reverse osmosis high-pressure pump operating frequency and the push of preventive maintenance warnings in step S4 specifically includes: S4.1 Read the remaining healthy operating time of the membrane calculated in step S3 and compare it with the preset early warning period in the intelligent control module. If the remaining healthy operating time of the membrane is greater than the early warning period, only update the internal status record and do not trigger active control actions or early warning push. S4.2 If the remaining healthy operating time of the membrane is less than or equal to the early warning period, the current real-time operating frequency of the reverse osmosis high-pressure pump is obtained, and the current operating frequency is subtracted from the preset frequency reduction step size to obtain the frequency reduction target value. The value of the frequency reduction step size is in the range of 2 to 3 Hz. The frequency reduction target value is checked to see if it is not lower than the minimum allowable frequency to ensure the minimum water production. If it is satisfied, the frequency reduction target value is sent as a new frequency command to the high-pressure pump frequency converter for execution. S4.3 Simultaneously generate a text warning message. The warning message includes at least the following: the type of pollution determined in step S2, i.e., scaling or particle deposition, the currently predicted remaining healthy operating time of the membrane, and the recommended cleaning method within the remaining time, i.e., citric acid cleaning. The warning message is pushed to the human-machine interface for display and written to the operation log for archiving.

[0048] As described in steps S4.1-S4.3 above, by comparing the remaining healthy operating time of the membrane with the preset early warning period, the reverse osmosis high-pressure pump frequency reduction operation is automatically executed to slow down the membrane fouling rate. At the same time, a preventive maintenance warning containing the fouling type, remaining time and cleaning recommendations is pushed out. Under the premise of ensuring the minimum clinical water demand, the membrane maintenance window is extended, and membrane fouling is actively controlled from the intervention point, eliminating unplanned shutdowns of the medical pure water system.

[0049] The rate of membrane fouling or particle deposition is positively correlated with the permeate flux of the reverse osmosis membrane. Appropriately reducing the permeate flux can effectively slow down the deposition rate of pollutants on the membrane surface. When the remaining healthy operating time of the membrane is less than or equal to the preset warning lead time, the membrane performance degradation rate has reached the warning threshold. If no active intervention measures are taken, the membrane permeate flow rate will quickly drop to the minimum clinical water demand threshold, directly triggering unplanned shutdown and failing to meet the core requirement of continuous water supply in medical scenarios. Therefore, it is necessary to adjust the operating frequency of the high-pressure pump to achieve a slight reduction in permeate flux, leaving sufficient operating time for manual membrane cleaning.

[0050] Current medical pure water control technologies only trigger passive alarms when membrane performance exceeds limits, lacking proactive preventative control measures. They cannot slow the progression of membrane fouling by adjusting operating parameters. Furthermore, warning information only includes fault indications, without specific fouling types or maintenance plan guidance. Maintenance personnel cannot quickly develop targeted solutions, making unplanned downtime due to delayed maintenance response highly likely, thus failing to meet the requirements for safe and stable operation of medical pure water systems. A comprehensive solution, employing remaining time threshold comparison, precise high-pressure pump frequency reduction, and customized warning push notifications, specifically addresses the technical problems of delayed membrane fouling intervention and incomplete warning information.

[0051] The intelligent control module reads the remaining healthy operating time of the membrane calculated in step S3 and compares this value with the preset early warning period within the module. If the remaining healthy operating time of the membrane is greater than the early warning period, the system only updates the internal status record and does not trigger active control actions or early warning pushes, maintaining the current operating state. This step, through quantitative threshold comparison, only initiates intervention when membrane fouling develops to the early warning state, avoiding unnecessary adjustments to operating parameters and ensuring the system's water production efficiency.

[0052] After the intelligent control module determines that the remaining healthy operating time of the membrane is less than or equal to the warning lead time, it obtains the current real-time operating frequency of the reverse osmosis high-pressure pump. It subtracts a preset frequency reduction step size from the current operating frequency to obtain the target frequency reduction value, with the step size ranging from 2 to 3 Hz. After verifying that the target frequency reduction value is not lower than the minimum allowable frequency to ensure minimum permeate flow, the system sends this target value as a new frequency command to the high-pressure pump inverter for execution. This step achieves a slight reduction in permeate flux through a small frequency reduction, slowing down the membrane fouling rate while meeting minimum clinical water demand. Furthermore, the frequency reduction amplitude is controllable and will not affect the normal medical water supply. For example, if the current operating frequency of the reverse osmosis high-pressure pump is 43 Hz, the preset frequency reduction step size is 3 Hz, and the calculated target frequency reduction value is 40 Hz. This value is higher than the minimum allowable frequency, and the system sends a 40 Hz operating command to the inverter.

[0053] The intelligent control module automatically generates text-based early warning information. This information includes the type of fouling determined in step S2, the currently predicted remaining healthy operating time of the membrane, and the corresponding cleaning method. The warning information is displayed on the HMI and recorded in the operation log. This comprehensive warning information provides maintenance personnel with precise guidance, shortening maintenance response time and ensuring timely membrane cleaning operations. For example, if the system sends a warning indicating fouling or particle deposition on the membrane element, predicting a remaining healthy operating time of 14 days, and recommending citric acid cleaning, this information is simultaneously displayed on the HMI and recorded in the operation log.

[0054] The above steps work together to achieve preventive and proactive intervention against membrane fouling. By accurately triggering intervention actions through threshold comparison, reducing the frequency of operations scientifically slows down the fouling rate, and providing customized early warning guidance for rapid maintenance, the intervention addresses the problems of passive response and delayed maintenance in existing technologies. This allows sufficient time for membrane cleaning and maintenance, continuously supporting the medical pure water system to achieve the core goal of zero unplanned downtime.

[0055] In one embodiment, the closed-loop evaluation of the intervention effect and the self-calibration of the diagnostic threshold in step S5 specifically include: S5.1 After the active frequency reduction action is executed, the standardized permeate flow rate data for the first preset time period is continuously collected. The slope of the standardized permeate flow rate change within the first preset time period after frequency reduction is calculated again using linear regression and recorded as the attenuation slope after frequency reduction. The attenuation slope of the standardized permeate flow rate before frequency reduction recorded in step S2 is read, which is the attenuation slope before frequency reduction. The difference between the absolute value of the attenuation slope before frequency reduction and the absolute value of the attenuation slope after frequency reduction is calculated, divided by the absolute value of the attenuation slope before frequency reduction, and then multiplied by 100% to obtain the attenuation reduction efficiency, which is used to quantify the inhibitory effect of active frequency reduction on membrane fouling rate. S5.2 When artificial membrane cleaning occurs and the membrane is put back into operation, the standardized permeate flow rate after the second preset time of stable operation after cleaning is collected and recorded as the standardized permeate flow rate after cleaning. This is compared with the standardized permeate flow rate recorded last time before cleaning. The membrane performance recovery rate is obtained by dividing the standardized permeate flow rate after cleaning by the standardized permeate flow rate before cleaning and then multiplying by 100%, which is used to quantify the degree of recovery of the membrane permeate capacity by this cleaning. S5.3 Based on the calculated attenuation mitigation efficiency or membrane performance recovery rate, perform diagnostic threshold self-correction: If the attenuation mitigation efficiency exceeds the preset high efficiency threshold, or the membrane performance recovery rate exceeds the preset high quality recovery threshold, it is determined that the current diagnostic threshold setting is too conservative, and the length of the historical data window used in the linear regression in step S3 is automatically reduced to improve the sensitivity and response speed of subsequent predictions. If the attenuation mitigation efficiency is lower than the preset low efficiency threshold, or the membrane performance recovery rate is lower than the preset qualified recovery threshold, it is determined that the current diagnostic threshold may lead to false alarms, and the threshold for judging small changes in the desalination rate of scaling and the desalination rate attenuation threshold in step S2 are automatically increased to reduce the probability of false judgment or improve the cleaning evaluation criteria. S5.4 updates the corrected historical data window length, desalination rate small change threshold, and desalination rate decay threshold to the non-volatile memory of the intelligent control module, so that the corrected parameters take effect in the next round of day-night switching cycle, realizing the continuous adaptive evolution of the diagnostic threshold.

[0056] As described in steps S5.1-S5.4 above, by calculating the attenuation mitigation efficiency and membrane performance recovery rate, the length of the historical data window for linear regression, the threshold for small changes in desalination rate, and the threshold for attenuation of desalination rate are dynamically adjusted. This allows the diagnostic and prediction model to continuously adapt to the on-site raw water quality and membrane element aging characteristics, maintain the accuracy of membrane fouling identification and remaining healthy operating time prediction in the long term, achieve adaptive evolution of the control strategy, and solidify the zero unplanned downtime operation effect of the medical pure water system.

[0057] The rate of membrane fouling development and the sensitivity of diagnostic thresholds will dynamically change with the differences in hospital raw water quality and the aging of membrane elements over long-term operation. The fixed diagnostic thresholds at the factory and the length of historical data windows cannot match all field conditions. The effect of active frequency reduction on the rate of fouling and the degree of membrane performance recovery by manual cleaning need to be quantitatively verified. Only by dynamically correcting control parameters based on the actual intervention effect can we avoid diagnostic misjudgment and inaccurate prediction after long-term operation and continuously ensure the effectiveness of membrane fouling prevention and control.

[0058] Existing medical pure water control technologies use factory-set diagnostic thresholds and predictive parameters, lacking self-calibration capabilities based on actual intervention effects. Over long-term operation, the deviation between the model and the actual membrane performance continues to widen. This can lead to delayed warnings due to conservative thresholds or false alarms due to threshold sensitivity. Furthermore, these technologies cannot adapt to the characteristics of raw water in different hospitals and the membrane aging process, making it difficult to achieve long-term, stable, and precise control of membrane fouling. A comprehensive solution, employing quantitative calculation of intervention effects, adaptive parameter calibration, and iterative updates to the control strategy, specifically addresses the technical problems of poor adaptability of fixed parameters and long-term accuracy degradation.

[0059] After the active frequency reduction action is executed, the intelligent control module continuously collects standardized permeable flow rate data for a first preset duration. It then uses linear regression to calculate the slope of the standardized permeable flow rate change during this period, recording it as the attenuation slope after frequency reduction. The system reads the standardized permeable flow rate attenuation slope recorded before frequency reduction, divides the difference between the absolute value of the attenuation slope before and after frequency reduction by the absolute value of the attenuation slope before frequency reduction, and multiplies by 100% to obtain the attenuation mitigation efficiency. The formula for calculating the attenuation mitigation efficiency is as follows: ; Among them, the This indicates the attenuation reduction efficiency, used to characterize the inhibitory effect of frequency reduction intervention on the pollution rate. Indicates the attenuation slope before frequency reduction. This represents the attenuation slope after frequency reduction. For example, before frequency reduction, the attenuation slope decreased by 1.5% per week, and after frequency reduction, the attenuation slope decreased by 0.6% per week. The calculated attenuation mitigation efficiency is 60%. This value directly reflects the mitigation effect of frequency reduction intervention on membrane fouling attenuation.

[0060] After manual membrane cleaning and restarting, the intelligent control module collects the standardized permeate flow rate after a second preset period of stable operation following cleaning. This is recorded as the standardized permeate flow rate after cleaning. This value is compared with the last recorded standardized permeate flow rate before cleaning. The membrane performance recovery rate is obtained by dividing the post-cleaning value by the pre-cleaning value and multiplying by 100%. This value quantifies the degree to which the cleaning operation restores the membrane's permeate capacity. For example, if the standardized permeate flow rate before cleaning was 480 L / h and the stable value after cleaning is 576 L / h, the calculated membrane performance recovery rate is 95%. This value can directly evaluate the effectiveness of the cleaning operation.

[0061] The intelligent control module performs parameter correction based on the degradation mitigation efficiency or membrane performance recovery rate. When the degradation mitigation efficiency exceeds the preset high efficiency threshold or the membrane performance recovery rate exceeds the preset high quality recovery threshold, it is determined that the current diagnostic threshold is too conservative. The historical data window length used in the linear regression in step S3 is automatically reduced to improve the response sensitivity of subsequent predictions. When the degradation mitigation efficiency is lower than the preset low efficiency threshold or the membrane performance recovery rate is lower than the preset qualified recovery threshold, it is determined that the current threshold is prone to misjudgment. The threshold for small changes in the desalination rate and the desalination rate degradation threshold for determining scaling are automatically increased in step S2 to reduce the probability of misjudgment.

[0062] The intelligent control module updates the corrected historical data window length, desalination rate minute change threshold, and desalination rate decay threshold to the non-volatile memory. The corrected parameters officially take effect in the next day-night switching cycle, completing the closed-loop iteration of the diagnostic and predictive model. This step solidifies the self-calibration results into the system, enabling the control strategy to continuously adapt to field conditions and achieve long-term stability of model accuracy. Through the above steps, intervention effect evaluation and parameter closed-loop self-calibration can be completed collaboratively. The control and maintenance effects can be objectively verified through quantitative calculations. Core parameters can be dynamically optimized based on real operating data, solving the problem of poor long-term adaptability of fixed parameters. This ensures that membrane fouling diagnosis and remaining time prediction maintain high accuracy, providing a closed-loop guarantee for the long-term stable operation of the medical pure water system with zero unplanned downtime.

[0063] like Figure 2 As shown, this invention also discloses a modularly designed medical multi-stage pure water integrated system, comprising: The acquisition submodule is used to acquire the operating parameters of the first-stage reverse osmosis module, perform temperature compensation standardization processing on the operating parameters to obtain standardized operating parameters, and calculate membrane performance indicators based on the operating parameters. The identification submodule is used to identify the membrane fouling state type based on the changing trend characteristics of the standardized operating parameters within a preset time window and according to preset diagnostic rules. The prediction submodule is used to generate a performance degradation trend and predict the remaining healthy operating time of the membrane only when the identification result is a preventable type of pollution. A generation submodule is used to generate control commands to adjust the operating status of the first-stage reverse osmosis module and push early warning information when the remaining healthy operating time of the membrane meets the preset early warning conditions. The correction submodule is used to acquire post-intervention operational data after the control operation is performed or the membrane maintenance operation is completed, calculate the intervention effect index, and use the intervention effect index to perform closed-loop correction on the judgment parameters in the diagnostic rules and the data window parameters used to generate the performance degradation trend.

[0064] In one embodiment, the correction submodule includes: The first calculation unit is used to collect standardized operating parameters for the period after the intervention after the control operation is executed, and to calculate the rate of change of performance degradation before and after the intervention as the degradation mitigation efficiency. The second calculation unit is used to collect standardized operating parameters for the period after membrane maintenance is completed, compare them with the parameters recorded before maintenance, and calculate the membrane performance recovery rate. The adjustment unit is used to adjust the judgment threshold in the diagnostic rules and the data window parameters used to generate the performance degradation trend according to the comparison result between the degradation mitigation efficiency or the membrane performance recovery rate and the corresponding preset threshold, and to update the adjusted parameters.

[0065] This application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described modular design-based integrated method for medical multi-stage pure water.

[0066] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described modular design-based integrated method for medical multi-stage pure water.

[0067] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in this application and in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0068] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0069] The above description is merely a preferred embodiment of the present invention and does not limit the scope of this application. Any equivalent results or equivalent process transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of protection of this application.

Claims

1. A modular design-based integrated method for medical multi-stage pure water systems, characterized in that, A method for a medical pure water device integrating a pretreatment module, a primary reverse osmosis module, and an intelligent control module, wherein the method is executed by the intelligent control module and includes the following steps: The operating parameters of the first-stage reverse osmosis module are obtained, and the operating parameters are standardized by temperature compensation to obtain standardized operating parameters. Based on the operating parameters, the membrane performance index is calculated. Based on the changing trend characteristics of the standardized operating parameters within a preset time window, the membrane fouling state type is identified according to preset diagnostic rules. Only when the identification result is a preventable type of pollution, the performance degradation trend is generated using the historical data sequence of the standardized operating parameters to predict the remaining healthy operating time of the membrane; When the remaining healthy operating time of the membrane meets the preset warning conditions, a control command is generated to adjust the operating status of the first-stage reverse osmosis module and a warning message is pushed. After the control operation is performed or the membrane maintenance operation is completed, the operation data after the intervention is obtained, the intervention effect index is calculated, and the intervention effect index is used to perform closed-loop correction on the judgment parameters in the diagnostic rules and the data window parameters used to generate the performance degradation trend.

2. The integrated method for medical multi-stage pure water based on modular design according to claim 1, characterized in that, The steps of obtaining the operating parameters of the first-stage reverse osmosis module and performing temperature compensation standardization processing on the operating parameters to obtain standardized operating parameters include: The feed water temperature, membrane pressure difference, permeate flow rate, feed water conductivity, and permeate conductivity of the first-stage reverse osmosis module are acquired at a preset acquisition cycle. Call the pre-stored temperature compensation mapping data and obtain the corresponding correction coefficient using the inlet water temperature as the index; The membrane pressure difference and the permeate flow rate are standardized using the correction coefficient to obtain standardized membrane pressure difference and standardized permeate flow rate. The real-time desalination rate is calculated based on the influent conductivity and the product water conductivity. The standardized membrane pressure difference, the standardized product water flow rate, and the real-time desalination rate are stored in the data buffer as standardized operating parameters.

3. The integrated method for medical multi-stage pure water based on modular design according to claim 1, characterized in that, The step of identifying the membrane fouling state type based on the trend characteristics of the standardized operating parameters within a preset time window and according to preset diagnostic rules includes: Extract the standardized membrane pressure difference sequence, standardized permeate flow rate sequence, and real-time desalination rate sequence from the data buffer using a preset number of sampling points; Linear fitting was performed on each extracted sequence to obtain the slope of change as the trend feature; The slope of each change is compared with the corresponding judgment threshold in the preset diagnostic rule table, and the diagnostic result of the membrane fouling state is output. The diagnostic result includes at least membrane fouling or particle deposition, membrane element damage or seal leakage. When the diagnostic result is membrane fouling or particle deposition, the relevant parameters are output to the subsequent processing stage. When the diagnostic result is membrane element damage or seal leakage, an alarm message is generated.

4. The integrated method for medical multi-stage pure water based on modular design according to claim 1, characterized in that, The step of generating a performance degradation trend and predicting the remaining healthy operating time of the membrane using historical data sequences of the standardized operating parameters includes: Standardized permeable flow rate data sequences of a predetermined number of sampling points within the data buffer are extracted and linear regression fitting is performed to obtain parameters characterizing the performance degradation rate. Read the preset standardized lower limit threshold of water production flow rate, and calculate the critical moment based on the attenuation trend and the lower limit threshold. The difference between the critical moment and the current moment is taken as the remaining healthy operating time of the membrane. When the remaining healthy operating time of the membrane is negative, it is set to 0 and an early warning is triggered.

5. The integrated method for medical multi-stage pure water based on modular design according to claim 1, characterized in that, The step of generating control commands to adjust the operating status of the first-stage reverse osmosis module and pushing early warning information includes: The remaining healthy operating time of the membrane is compared with the preset early warning period; When the remaining healthy operating time of the membrane is less than or equal to the early warning period, a frequency reduction control command is generated based on the current operating frequency of the first-stage reverse osmosis module and the preset adjustment step size, and sent to the frequency converter for execution under the condition of meeting the minimum operating frequency constraint. Simultaneously generate early warning information including the type of pollution, the remaining healthy operating time of the membrane, and recommended maintenance methods, push it to the human-machine interface, and record it in the operation log.

6. The integrated method for medical multi-stage pure water based on modular design according to claim 1, characterized in that, The step of using the intervention effect index to perform closed-loop correction on the judgment parameters in the diagnostic rules and the data window parameters used to generate the performance degradation trend includes: After the control operation is executed, standardized operating parameters are collected for the period after the intervention, and the rate of change of performance degradation before and after the intervention is calculated as the degradation mitigation efficiency. After the membrane maintenance operation is completed, standardized operating parameters for the period after maintenance are collected and compared with the parameters recorded before maintenance to calculate the membrane performance recovery rate. Based on the comparison results between the degradation mitigation efficiency or the membrane performance recovery rate and the corresponding preset threshold, the judgment threshold in the diagnostic rules and the data window parameters used to generate the performance degradation trend are adjusted accordingly, and the adjusted parameters are updated.

7. A modularly designed integrated medical multi-stage pure water system, characterized in that, include: The acquisition submodule is used to acquire the operating parameters of the first-stage reverse osmosis module, perform temperature compensation standardization processing on the operating parameters to obtain standardized operating parameters, and calculate membrane performance indicators based on the operating parameters. The identification submodule is used to identify the membrane fouling state type based on the changing trend characteristics of the standardized operating parameters within a preset time window and according to preset diagnostic rules. The prediction submodule is used to generate a performance degradation trend and predict the remaining healthy operating time of the membrane only when the identification result is a preventable type of pollution. A generation submodule is used to generate control commands to adjust the operating status of the first-stage reverse osmosis module and push early warning information when the remaining healthy operating time of the membrane meets the preset early warning conditions. The correction submodule is used to acquire post-intervention operational data after the control operation is performed or the membrane maintenance operation is completed, calculate the intervention effect index, and use the intervention effect index to perform closed-loop correction on the judgment parameters in the diagnostic rules and the data window parameters used to generate the performance degradation trend.

8. The medical multi-stage pure water integrated system based on modular design according to claim 7, characterized in that, The correction submodule includes: The first calculation unit is used to collect standardized operating parameters for the period after the intervention after the control operation is executed, and to calculate the rate of change of performance degradation before and after the intervention as the degradation mitigation efficiency. The second calculation unit is used to collect standardized operating parameters for the period after membrane maintenance is completed, compare them with the parameters recorded before maintenance, and calculate the membrane performance recovery rate. The adjustment unit is used to adjust the judgment threshold in the diagnostic rules and the data window parameters used to generate the performance degradation trend according to the comparison result between the degradation mitigation efficiency or the membrane performance recovery rate and the corresponding preset threshold, and to update the adjusted parameters.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images