Modeling method of RO membrane risk diagnosis model

Abstract

The invention relates to the technical field of membrane operation, in particular to a modeling method of an RO membrane risk diagnosis model, which comprises the following steps of: establishing a mark event library of RO membrane operation, and marking according to a time sequence input by each mark event; constructing a basic model according to RO membrane operation data, water quality biochemical test data, expert analysis and diagnosis logic and general judgment rules; the preprocessed data set is input into the basic model, risk diagnosis results are obtained, and in the risk diagnosis results, one timestamp corresponds to one risk diagnosis event; according to the corresponding timestamps, mark events in a mark event library are called, and one risk diagnosis event corresponds to at least one mark event; and comparing the risk diagnosis event with the corresponding mark event to obtain a deviation result, and performing model iteration according to the deviation result until the deviation result meets requirements to obtain an optimized RO membrane risk diagnosis model. According to the invention, the risk diagnosis efficiency and the operation management efficiency of the RO membrane are improved.

Description

Technical Field

[0001] The invention relates to the technical field of membrane operation, and in particular to a modeling method of an RO membrane risk diagnosis model. Background Art

[0002] At present, the risk diagnosis of the operation process of the recycled water RO membrane mainly comes from the expert's judgment on the operation data such as pressure difference, water production, water conductivity, etc., or the judgment based on standardized data and test data. Such diagnostic methods have many limitations. The most obvious is the risk analysis and diagnosis based on the experience of professionals. When the relevant professionals of an enterprise leave, this part of the diagnostic ability will be lost; in addition, when the experience of professionals needs to be transferred from one site RO membrane to another site RO membrane with different working conditions, the learning of the operation characteristics of the RO membranes at different sites is slow and the migration time of the RO membrane operation diagnosis analysis knowledge between different sites is long. It also takes some time to understand the operation status of the RO membrane. In addition, the characterization of similar RO membranes may correspond to different risk events and different actions to be taken in different scenes at different sites. This also requires time accumulation, resulting in the inability of general diagnostic methods to be quickly applied to different scenes, greatly reducing the operational management efficiency of RO membranes.

[0003] Therefore, how to quickly diagnose the risks during the long-term operation of the RO membrane is crucial to extending the operating life of the RO membrane and improving the operational management efficiency of the RO membrane. Summary of the invention

[0004] The purpose of the present invention is to provide a modeling method for a RO membrane risk diagnosis model to solve the problem of rapid diagnosis of various risks that may occur during the long-term operation of the RO membrane.

[0005] In order to achieve the above object, the present invention provides a modeling method of an RO membrane risk diagnosis model, comprising the following steps:

[0006] Establish a landmark event library for RO membrane operation and mark each landmark event according to the time sequence of input;

[0007] Constructing a basic model based on the RO membrane operation data, water quality biochemical test data, expert analysis and diagnosis logic and general judgment rules;

[0008] Inputting the preprocessed data set into the basic model to obtain a risk diagnosis result, in which one timestamp corresponds to one risk diagnosis event;

[0009] According to the corresponding timestamp, retrieve the marker event in the marker event library, where one risk diagnosis event corresponds to at least one marker event;

[0010] The risk diagnosis event is compared with the corresponding marker event to obtain a deviation result, and the model is iterated according to the deviation result until the deviation result meets the requirements, thereby obtaining an optimized RO membrane risk diagnosis model.

[0011] Optionally, the content annotated according to the time sequence of each marker event input includes a timestamp, a risk severity label, an event type label, and a keyword label.

[0012] Optionally, according to the main risks occurring during the operation of the RO membrane, for each risk, a three-dimensional coordinate system is established based on the three dimensions of operation data, water quality biochemical test data and data operation characteristics, and the expert analysis and diagnosis logic is constructed based on the point area in the three-dimensional coordinate system;

[0013] The steps to obtain risk diagnosis results include:

[0014] According to the position of the preprocessed real-time data in the three-dimensional coordinate system, the expert analysis and diagnosis logic is called to determine the severity of the risk and output the risk diagnosis event.

[0015] Optionally, if the input data of the basic model comes from the same time point, the time point is used as the timestamp of the risk diagnosis event output by the model; if the input data of the basic model comes from different times, the last time point of all the input data is used as the timestamp of the risk diagnosis event output by the model.

[0016] Optionally, when comparing the risk diagnosis event with the corresponding marker event, if the risk diagnosis event fails to correspond to all the marker events, it indicates that the comparison has failed; otherwise, it indicates that the comparison has been successful.

[0017] Optionally, the number of risk diagnosis events for which the comparison fails is divided by the total number of risk diagnosis events to obtain a deviation value, and the basic model is iterated based on the deviation value.

[0018] Optionally, the iterative mechanism includes:

[0019] When the deviation value is greater than a first preset value, triggering model iteration;

[0020] When the deviation value is greater than a second preset value, sequentially traversing the tunable parameters of the basic model, wherein the second preset value is greater than the first preset value;

[0021] When the deviation value is greater than a first preset value and less than a third preset value, preferential tuning is performed according to tuning parameters related to the weakest type of diagnosis, wherein the second preset value is greater than the third preset value;

[0022] When the deviation value is between the third preset value and the third preset value, manual tuning is performed.

[0023] Optionally, the iterative mechanism further includes:

[0024] When the optimization of the deviation value is less than the fourth preset value after the basic model has been adjusted for a certain number of consecutive preset times, the basic model will fall back to the time before the parameter adjustment, and select the next priority tunable parameter for adjustment according to the current rule;

[0025] When the entire tuning process is completed according to the above iterative mechanism, if the deviation value still does not reach the set target, the iteration is terminated.

[0026] Optionally, when the deviation value is greater than a second preset value, the step of sequentially traversing the tunable parameters of the basic model specifically includes:

[0027] Three different types of diagnostic logic are set: f1(x), f2(y), and f3(z), where x, y, and z represent the RO membrane operation data, water quality biochemical test data, and data analysis characteristics, respectively;

[0028] The adjustment parameters for each indicator of the three diagnostic logics are set as: a1, a2, a3, a4...a n ; b1, b2, b3, b4...b n ; c1, c2, c3, c4...c n ;

[0029] The adjustment functions for the first two diagnostic logics f1(x) and f2(y) are set to be g1(z1,z2,z3,z4…z n ), g2(z1,z2,z3,z4…z n );

[0030] Then the basic model is recorded as:

[0031] {g1(z1,z2,z3,z4…z n )×F1(a1f1(x1),a2f1(x2),a3f1(x3)…a n f1(x n )),

[0032] g2(z1,z2,z3,z4…z n )×F2(b1f2(y1),b2f2(y2),b3f2(y3)…b n f2(y n )),

[0033] F3(c1f3(z1),c2f3(z2),c3f3(z3)…c n f3(z n ))}

[0034] According to the adjustment range [AL, AH] of each adjustable parameter, the adjustment interval is divided into N equal parts, where N is a positive integer greater than 1;

[0035] Starting from a1, including a1-a n ,b1-b n ,c1-c n All adjustable parameters, including, are traversed in all combinations within the parameter adjustment range;

[0036] After the traversal, the risk diagnosis event is compared with the corresponding marker event to obtain the deviation result. If the deviation result meets the requirements, the tuning is completed and the step is terminated; otherwise, the best several groups of tuning models are selected, and each adjustable parameter is set with the current optimal value of each adjustable parameter as the set point, and the interval is the parameter adjustment range. The parameter adjustment interval is divided into N equal parts again, and all combinations are traversed;

[0037] Repeat the above steps until the deviation result meets the requirements.

[0038] Optionally, for mature RO membranes, model iterations can be performed based on set time or membrane washing cycles;

[0039] For RO membranes that have been in operation for more than a certain number of years, when a new sign event defined as an important level is added to the sign event library, the model is iterated;

[0040] For newly commissioned RO membranes, model iterations are performed weekly or biweekly during a period defined as the trial operation period.

[0041] In the modeling method of the RO membrane risk diagnosis model provided by the present invention, at least one of the following beneficial effects is achieved:

[0042] 1) Through the methods of dynamic simulation and model parameter tuning, the problems of slow learning of RO operation characteristics at different sites, long migration time of RO operation diagnosis and analysis knowledge between different sites, and the inability of general judgment methods to be quickly applied to different scenarios are solved, which reduces the debugging time and manpower costs of the analysis and diagnosis logic algorithm and greatly improves the operation and management efficiency of RO membranes;

[0043] 2) By building a database of RO membrane operation event markers and expert analysis and diagnosis logic, the accumulated experience on site and the expert experience are matched, which solves the efficiency problems, potential risks (such as the inability of experts to be on duty in sudden or emergency situations) and cost problems (employment) that may be caused by the reliance on experts in risk judgment during the long-term operation of RO membranes. This improves the efficiency of RO membrane risk judgment, which can better extend the operating life of the membrane, reduce operating costs, and improve water production efficiency;

[0044] 3) For new problems arising from new working conditions, events can be quickly stored in the database and risk issues can be quickly compared to achieve more accurate diagnosis, ensure membrane life, reduce membrane risks, and improve operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Those skilled in the art should understand that the drawings are provided for a better understanding of the present invention and do not constitute any limitation on the scope of the present invention.

[0046] Figure 1 A flow chart of a modeling method for a RO membrane risk diagnosis model provided by one embodiment of the present invention;

[0047] Figure 2 A schematic diagram of extracting a landmark event provided by an embodiment of the present invention;

[0048] Figure 3 A schematic diagram of comparing risk diagnosis events and marker events provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, advantages and features of the present invention clearer, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the accompanying drawings are in a very simplified form and use non-precise proportions, which are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention. In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, please refer to the accompanying drawings. It should be noted that the structure, proportion, size, etc. illustrated in the drawings of this specification are only used to match the content disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention. Any modification of the structure, change in the proportional relationship or adjustment of the size, under the same or similar conditions as the effects that can be produced by the present invention and the purposes that can be achieved, should still fall within the scope of the technical content disclosed by the present invention.

[0050] As used in the present invention, the singular forms "one", "an", and "the" include plural objects, unless the content clearly indicates otherwise. As used in the present invention, the term "or" is generally used in a sense that includes "and / or", unless the content clearly indicates otherwise. As used in the present invention, the term "several" is generally used in a sense that includes "at least one", unless the content clearly indicates otherwise. As used in the present invention, the term "at least two" is generally used in a sense that includes "two or more", unless the content clearly indicates otherwise. In addition, the terms "first", "second", and "third" are used only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, the features defined as "first", "second", and "third" may explicitly or implicitly include one or at least two of the features.

[0051] In the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", and "fixed" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0052] Please refer to Figure 1 This embodiment provides a modeling method for a RO membrane risk diagnosis model, comprising the following steps:

[0053] S1. Establish a landmark event library for RO membrane operation and mark each landmark event according to the time sequence of input;

[0054] S2. Constructing a basic model based on the RO membrane operation data, water quality biochemical test data, expert analysis and diagnosis logic and general judgment rules;

[0055] S3, inputting the preprocessed data set into the basic model to obtain a risk diagnosis result, in which one timestamp corresponds to one risk diagnosis event;

[0056] S4. Retrieving a marker event in the marker event library according to the corresponding timestamp, wherein one risk diagnosis event corresponds to at least one marker event;

[0057] S5. Compare the risk diagnosis event with the corresponding marker event to obtain a deviation result, and iterate the model according to the deviation result until the deviation result meets the requirements, thereby obtaining an optimized RO membrane risk diagnosis model.

[0058] First, execute S1 to establish a marker event library for RO membrane operation, and mark each marker event according to the time sequence of input. In this embodiment, the marker event library comes from various events such as on-site service records, operation status records, inspection summary, reagent treatment summary, on-site trend analysis, etc., such as: a certain operation index is on the rise / downward / relatively stable, a certain test index is relatively stable, the system processing volume this week is relatively stable, a certain test index is low / high, the scaling trend of the system slows down after adding the scale inhibitor, the consumption of the scale inhibitor this week is low, the salt permeability of the system continues to rise abnormally, and the water production of the system continues to fall abnormally.

[0059] Preferably, the content marked according to the time sequence of each marker event input includes a timestamp, a risk severity label, an event type label, and a keyword label. In this embodiment, the events in the marker event library are marked with C1, C2, C3...Cn according to the input time sequence, and all events are provided with severity labels, type labels, and keyword labels. The severity labels are divided into low, medium, and high; the type labels are divided into operation events, water quality events, risk events, maintenance events, threshold events, etc., which can be added or deleted according to on-site needs; the threshold event refers to the cleaning of the membrane, the change of the type of agent addition, the change of the concentration of the agent addition, the change of the maintenance method, etc., which will affect the membrane operation status or the membrane health status; the keyword label includes the indicator name, the risk name such as scaling, membrane oxidation, microbial risk, etc.

[0060] The editing users of the landmark event library are divided into Class I users and Class II users. Class I users have the authority to mark existing events and add timestamps, as well as submit applications for new events, new types, and new labels. Class II users are responsible for approving new events, new types, and new labels.

[0061] The establishment of the marker event library can be done by directly copying the existing RO membrane operation marker event library, or by establishing a new marker event library based on a summary table of various on-site events, or by adding new on-site events based on the existing RO membrane operation marker event library.

[0062] The collection of the landmark events in the operation of the RO membrane comes from historical reports and inspection records, as well as from the information collected and recorded in a timely manner by Class I users; during the entry process after the collection is completed, Class I users will automatically retrieve the existing event entries displayed by keywords to determine whether the entered landmark event already exists. When Class I users determine that the entered event is not an existing event in the event library, they will submit the new event; Class I users submit the definition of severity, type and keywords at the same time; after submission, Class II users confirm the event description, severity, type and keywords. After confirmation, the incremental construction of the landmark event library is realized, and the new events are arranged in sequence numbers among the existing events. In addition, Class II users regularly clean up, reorganize or delete events in the event library.

[0063] In this embodiment, the marking process of the landmark event is as follows:

[0064] Online record of RO membrane operation events: The online record of RO membrane operation events refers to the entry of RO membrane operation events originally recorded in paper or Excel into the system after the user goes online with the RO membrane risk diagnosis model;

[0065] RO membrane operation event search and comparison: After the online record is completed, the system will automatically search for keywords and recommend the 3 event records with the highest matching degree

[0066] RO membrane operation event confirmation and annotation: After the system automatically compares and recommends the most matching event record, the Class I user manually confirms and enters the annotation record that matches the event.

[0067] Then, S2 is executed to construct a basic model based on the RO membrane operation data, water quality biochemical test data, expert analysis and diagnosis logic and general judgment rules. And S3 is executed to input the preprocessed data set into the basic model to obtain a risk diagnosis result, in which one timestamp corresponds to one risk diagnosis event.

[0068] In this embodiment, according to the main risks occurring during the operation of the RO membrane, for each risk, a three-dimensional coordinate system is established based on the three dimensions of operation data, water quality biochemical test data, and data operation characteristics, and the expert analysis and diagnosis logic is constructed based on the point area in the three-dimensional coordinate system;

[0069] The steps to obtain risk diagnosis results include:

[0070] According to the position of the preprocessed real-time data in the three-dimensional coordinate system, the expert analysis and diagnosis logic is called to determine the severity of the risk and output the risk diagnosis event.

[0071] In this embodiment, the main risks that may occur during the operation of the RO membrane are mainly evaluated from the perspective of membrane damage and membrane operation efficiency. The risks from the perspective of membrane damage include risks such as membrane oxidation and scaling; the risks from the perspective of membrane operation efficiency include the decrease in water production and salt permeability, and the accumulation of microorganisms.

[0072] The three dimensions are respectively the dimensions of operation data such as water production, conductivity, pressure, etc., the dimensions of water quality biochemical test data such as iron, aluminum, silica, SDI, etc., and the dimensions of data operation characteristics. The data operation characteristics refer to the mathematical meaning of operation data and water quality biochemical test data in the time dimension, such as rising, falling, continuous rising, continuous falling, fluctuations defined by users, correlation of multiple data, etc.

[0073] The preprocessing includes data filtering, effective data screening, data overbalancing processing, data normalization processing and other preprocessing work.

[0074] The expert analysis and diagnosis logic refers to the expert analysis and diagnosis logic for RO membrane risk assessment constructed by experts based on the point area in the three-dimensional coordinate system, including analysis and judgment of risk phenomena, speculation on the causes of risks, and suggestions for further screening or maintenance actions for existing risks. It is called according to the regional position of the influencing factors constructed in the expert analysis and diagnosis logic, and a textual reference opinion is output, which is the risk diagnosis.

[0075] In this embodiment, if the input data of the basic model comes from the same time point, the time point is used as the timestamp of the risk diagnosis event output by the model; if the input data of the basic model comes from different times, the last time point of all the input data is used as the timestamp of the risk diagnosis event output by the model; if the input data of the basic model comes from a time range, the last time point in the time range of all the input data is used as the timestamp of the risk diagnosis event output by the model.

[0076] Then, S4 is executed to retrieve a marker event in the marker event library according to the corresponding timestamp. One risk diagnosis event corresponds to at least one marker event.

[0077] Finally, S5 is executed to compare the risk diagnosis event with the corresponding marker event to obtain a deviation result, and the model is iterated according to the deviation result until the deviation result meets the requirement, thereby obtaining an optimized RO membrane risk diagnosis model.

[0078] For example, Figure 2-Figure 3 As shown, from the risk diagnosis event D nThe corresponding timestamp is YYYY-MM-DDHH:mm:ss. The first threshold event C is retrieved. k , then extract the threshold event C k The timestamp is YYYY1-MM1-DD1HH1:mm1:ss; extract all the marker events C between YYYY-MM-DD HH:mm:ss and YYYY1-MM1-DD1 HH1:mm1:ss1 (excluding C k ), the obtained flag event is C n , C n+1 , C n+2 , C n+3 ...;

[0079] like Figure 3 As shown, the risk diagnosis event D n and the corresponding landmark event C n , C n+1 , C n+2 , C n+3 ...for comparison, if D n and C n , C n+1 , C n+2 , C n+3 ...if any one of them is successful, then the D n The comparison is successful if D n and C n , C n+1 , C n+2 , C n+3 If none of the items in ... match successfully, the comparison fails.

[0080] The number of risk diagnosis events that failed the comparison is then divided by the total number of risk diagnosis events to obtain a deviation value, and the basic model is iterated based on the deviation value.

[0081] Furthermore, the iterative mechanism includes:

[0082] When the deviation value is greater than a first preset value, triggering model iteration;

[0083] When the deviation value is greater than a second preset value, sequentially traversing the tunable parameters of the basic model, wherein the second preset value is greater than the first preset value;

[0084] When the deviation value is greater than a first preset value and less than a third preset value, priority tuning is performed according to tuning parameters related to the weakest type of diagnosis, wherein the second preset value is greater than the third preset value;

[0085] When the deviation value is between the third preset value and the third preset value, manual tuning is performed.

[0086] Furthermore, the iterative mechanism also includes:

[0087] When the optimization of the deviation value is less than the fourth preset value after the basic model has been adjusted for a certain number of consecutive preset times, the basic model will fall back to the time before the parameter adjustment, and select the next priority tunable parameter for adjustment according to the current rule;

[0088] When the entire tuning process is completed according to the above iterative mechanism, if the deviation value still does not reach the set target, the iteration is terminated.

[0089] In this embodiment, the iterative mechanism includes:

[0090] Rule 1: Model iteration trigger rule: For example, when the deviation is greater than 5%, the model iteration is triggered;

[0091] Rule 2: Model sequential iteration rule: When the deviation meets a preset rule, such as greater than or equal to 50%, the tunable parameters of the basic model are sequentially traversed;

[0092] Rule 3: Prioritize the iterative rule: When the deviation meets a preset rule, such as the deviation is less than or equal to 30%, the tuning parameters related to the weakest class diagnosis are prioritized for tuning;

[0093] Rule 4: Manual intervention iteration rule: When the deviation meets a preset rule, such as the deviation is between 30% and 50%, the automatic iteration is stopped and the tuning rule is set manually;

[0094] Rule 5: Model storage rules: Model version preservation and rollback mechanism;

[0095] Rule 6: Model iteration fallback rule: When the model is tuned for a certain number of times (such as 100 times) in a row, and the optimization of the deviation is less than the optimization of the fourth preset value (such as 5%), it will fall back to the number before that, and select the next priority adjustable parameter for tuning according to the current rule.

[0096] Rule 7: Model iteration termination rule: In addition to when the tuning reaches the preset deviation requirement, when the entire tuning process is completed according to the sequence rule and fallback rule, but the deviation requirement is still not met, the model iteration is terminated.

[0097] Preferably, when the deviation value is greater than the second preset value, the step of sequentially traversing the tunable parameters of the basic model specifically includes:

[0098] Three different types of diagnostic logic are set: f1(x), f2(y), and f3(z), where x, y, and z represent the RO membrane operation data, water quality biochemical test data, and data analysis characteristics, respectively;

[0099] The adjustment parameters for each indicator of the three diagnostic logics are set as: a1, a2, a3, a4...a n ; b1, b2, b3, b4...b n ; c1, c2, c3, c4...c n ;

[0100] The adjustment functions for the first two diagnostic logics f1(x) and f2(y) are set to be g1(z1,z2,z3,z4…z n ), g2(z1,z2,z3,z4…z n );

[0101] Then the basic model is recorded as:

[0102] {g1(z1,z2,z3,z4…z n )×F1(a1f1(x1),a2f1(x2),a3f1(x3)…a n f1(x n )),

[0103] g2(z1,z2,z3,z4…z n )×F2(b1f2(y1),b2f2(y2),b3f2(y3)…b n f2(y n )),

[0104] F3(c1f3(z1),c2f3(z2),c3f3(z3)…c n f3(z n ))}

[0105] When the deviation value triggers the model sequential iteration rule, the following steps are performed:

[0106] According to the tuning range [AL, AH] of each adjustable parameter, the tuning interval is divided into N equal parts (e.g., 10 equal parts), where N is a positive integer greater than 1;

[0107] Starting from a1, including a1-a n ,b1-b n ,c1-c n All adjustable parameters, including, are traversed in all combinations within the parameter adjustment range;

[0108] After the traversal, the risk diagnosis event is compared with the corresponding marker event to obtain the deviation result. If the deviation result meets the requirements, the tuning is completed and the step is terminated; otherwise, the best several groups of tuning models are selected, and each adjustable parameter is set with the current optimal value of each adjustable parameter as the set point, and the interval is the parameter adjustment range. The parameter adjustment interval is divided into N equal parts again, and all combinations are traversed;

[0109] Repeat the above steps until the deviation result meets the requirements. If the deviation value requirement cannot be met after multiple consecutive times (such as 3 times), the model iteration is terminated and manual intervention is performed.

[0110] In this embodiment, when the deviation value triggers the model priority reinforcement iteration rule, the following steps are performed:

[0111] When the running model generates results D1, D2, D3... and C1, C2, C3... one by one, the deviation scores of the dynamic simulation deviation performance of f1(x), f2(y), and f3(z) are calculated respectively;

[0112] For the diagnostic logic with the lowest deviation score, the tunable parameters related to the diagnostic logic are tuned according to the tuning steps of the above model sequence iteration rules, and brought into the complete model for dynamic simulation;

[0113] If the diagnostic logic with the lowest score is optimized to meet the overall deviation requirement, the model iteration is terminated;

[0114] If the deviation value requirement is not met, the tunable parameters in the diagnostic logic with the second lowest score are tuned and dynamically simulated according to the tuning steps of the above-mentioned model sequential iteration rule.

[0115] If the deviation value requirement is not met, the tunable parameters related to the diagnostic logic are tuned and dynamically simulated in order from low to high according to the tuning steps of the above-mentioned model sequential iteration rules according to the scores.

[0116] If the deviation value requirement is not met, all adjustable parameters in the model are tuned according to the tuning steps of the above model sequential iteration rule until the iteration is completed or terminated.

[0117] After obtaining the optimized RO membrane risk diagnosis model, it can be used for rapid diagnosis of risks during the long-term operation of the RO membrane. For new problems arising from new working conditions, events can be quickly stored in the database and risk problems can be quickly compared to achieve more accurate diagnosis, ensure membrane life, reduce membrane risks, and improve operational efficiency.

[0118] Preferably, for a mature RO membrane, the model iteration is performed according to a set time or membrane washing cycle;

[0119] For RO membranes that have been in operation for more than a certain number of years, when a new sign event defined as an important level is added to the sign event library, the model is iterated;

[0120] For newly commissioned RO membranes, model iterations are performed weekly or biweekly during a period defined as the trial operation period.

[0121] In summary, the embodiments of the present invention provide a modeling method for an RO membrane risk diagnosis model. Through dynamic simulation and model parameter tuning methods, the problems of slow learning of RO operation characteristics at different sites, long migration time of RO operation diagnosis analysis knowledge between different sites, and inability of general judgment methods to be quickly applied to different scenarios are solved. This reduces the debugging time and labor costs of the analysis and diagnosis logic algorithm, and greatly improves the operational management efficiency of the RO membrane.

[0122] In addition, it should be recognized that although the present invention has been disclosed as a preferred embodiment, the above embodiment is not intended to limit the present invention. For any technician familiar with the art, without departing from the scope of the technical solution of the present invention, the technical content disclosed above can be used to make many possible changes and modifications to the technical solution of the present invention, or modified into equivalent embodiments of equivalent changes. Therefore, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still belongs to the scope of protection of the technical solution of the present invention.

Claims

1. A modeling method for a RO membrane risk diagnosis model, characterized in that: The following steps are involved: Establish a landmark event library for RO membrane operation and mark each landmark event according to the time sequence of input; Constructing a basic model based on the RO membrane operation data, water quality biochemical test data, expert analysis and diagnosis logic and general judgment rules; Inputting the preprocessed data set into the basic model to obtain a risk diagnosis result, in which one timestamp corresponds to one risk diagnosis event; According to the corresponding timestamp, retrieve the marker event in the marker event library, where one risk diagnosis event corresponds to at least one marker event; The risk diagnosis event is compared with the corresponding marker event to obtain a deviation result, and the model is iterated according to the deviation result until the deviation result meets the requirements, thereby obtaining an optimized RO membrane risk diagnosis model.

2. The modeling method of the RO membrane risk diagnosis model according to claim 1, characterized in that: The content annotated according to the time sequence of each marker event input includes a timestamp, a risk severity label, an event type label, and a keyword label.

3. The modeling method of the RO membrane risk diagnosis model according to claim 1, characterized in that: According to the main risks occurring during the operation of the RO membrane, for each risk, a three-dimensional coordinate system is established based on the three dimensions of operation data, water quality biochemical test data and data operation characteristics, and the expert analysis and diagnosis logic is constructed based on the point area in the three-dimensional coordinate system; The steps to obtain risk diagnosis results include: According to the position of the preprocessed real-time data in the three-dimensional coordinate system, the expert analysis and diagnosis logic is called to determine the severity of the risk and output the risk diagnosis event.

4. The modeling method of the RO membrane risk diagnosis model according to claim 1, characterized in that: If the input data of the basic model comes from the same time point, then this time point is used as the timestamp of the risk diagnosis event output by the model; if the input data of the basic model comes from different times, then the last time point of all the input data is used as the timestamp of the risk diagnosis event output by the model.

5. The modeling method of the RO membrane risk diagnosis model according to claim 1, characterized in that: When comparing the risk diagnosis event with the corresponding marker event, if the risk diagnosis event fails to correspond to all the marker events, it means the comparison has failed; otherwise, it means the comparison has been successful.

6. The modeling method of the RO membrane risk diagnosis model according to claim 5, characterized in that: The number of risk diagnosis events in which the comparison fails is divided by the total number of risk diagnosis events to obtain a deviation value, and the basic model is iterated based on the deviation value.

7. The modeling method of the RO membrane risk diagnosis model according to claim 1, characterized in that: The iterative mechanism includes: When the deviation value is greater than a first preset value, triggering model iteration; When the deviation value is greater than a second preset value, sequentially traversing the tunable parameters of the basic model, wherein the second preset value is greater than the first preset value; When the deviation value is greater than a first preset value and less than a third preset value, priority tuning is performed according to tuning parameters related to the weakest type of diagnosis, wherein the second preset value is greater than the third preset value; When the deviation value is between the third preset value and the third preset value, manual tuning is performed.

8. The modeling method of the RO membrane risk diagnosis model according to claim 7, characterized in that: The iterative mechanism also includes: When the optimization of the deviation value is less than the fourth preset value after the basic model has been adjusted for a certain number of consecutive preset times, the basic model will fall back to the time before the parameter adjustment, and select the next priority tunable parameter for adjustment according to the current rule; When the entire tuning process is completed according to the above iterative mechanism, if the deviation value still does not reach the set target, the iteration is terminated.

9. The modeling method of the RO membrane risk diagnosis model according to claim 7, characterized in that: When the deviation value is greater than the second preset value, the step of sequentially traversing the tunable parameters of the basic model specifically includes: Three different types of diagnostic logic are set: f1(x), f2(y), and f3(z), where x, y, and z represent the RO membrane operation data, water quality biochemical test data, and data analysis characteristics, respectively; The adjustment parameters for each indicator of the three diagnostic logics are set as: a1, a2, a3, a4...a n ; b1, b2, b3, b4...b n ; c1, c2, c3, c4...c n ; The adjustment functions for the first two diagnostic logics f1(x) and f2(y) are set to be g1(z1,z2,z3,z4…z n ), g2(z1,z2,z3,z4…z n ); Then the basic model is recorded as: <h2 style=";text-align:left;direction:ltr">{g1(z1,z2,z3,z4…z<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> )×F1(a1f1(x1),a2f1(x2),a3f1(x3)…a<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> f1(x<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> )), <h2 style=";text-align:left;direction:ltr">g2(z1,z2,z3,z4…z<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> )×F2(b1f2(y1),b2f2(y2),b3f2(y3)…b<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> f2(y<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> )), <h2 style=";text-align:left;direction:ltr">F3(c1f3(z1),c2f3(z2),c3f3(z3)…c<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> f3(z<h2 style=";text-align:left;direction:ltr"> n <h2 style=";text-align:left;direction:ltr"> ))} According to the adjustment range [AL, AH] of each adjustable parameter, the adjustment interval is divided into N equal parts, where N is a positive integer greater than 1; Starting from a1, including a1-a n ,b1-b n ,c1-c n All adjustable parameters, including, are traversed in all combinations within the parameter adjustment range; After the traversal, the risk diagnosis event is compared with the corresponding marker event to obtain the deviation result. If the deviation result meets the requirements, the tuning is completed and the step is terminated; otherwise, the best several groups of tuning models are selected, and each adjustable parameter is set with the current optimal value of each adjustable parameter as the set point, and the interval is the parameter adjustment range. The parameter adjustment interval is divided into N equal parts again, and all combinations are traversed; Repeat the above steps until the deviation result meets the requirements.

10. The modeling method of the RO membrane risk diagnosis model according to claim 1, characterized in that: For mature RO membranes, the model is iterated according to the set time or membrane washing cycle; For RO membranes that have been in operation for more than a certain number of years, when a new sign event defined as an important level is added to the sign event library, the model is iterated; For newly commissioned RO membranes, model iterations are performed weekly or biweekly during a period defined as the trial operation period.

Images