Water area pollution abatement method and system based on digital twinning

Abstract

The invention is suitable for the technical field of water environment intelligent monitoring, and provides a digital twinning-based water area pollution treatment method and system, and the method comprises the steps: obtaining an initial pollution response sensitivity parameter for evaluating a target fish species in a monitored water area based on a digital twinning technology, meanwhile, historical behavior monitoring data related to target fish species in a set time interval after the pollution event occurs in the monitored water area and conductivity time sequence change data are obtained; according to the method, the response lag correction coefficient and the conductivity sensitivity correction coefficient are introduced, the initial pollution response sensitivity parameter in the digital twin system is dynamically corrected, and the problem that in the prior art, fish behavior simulation lacks consideration on behavior response delay characteristics and water physical change factors is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent monitoring of water environment, and in particular relates to a water pollution control method and system based on digital twins. Background Art

[0002] At present, digital twin technology has been gradually applied to water pollution management scenarios. By integrating physical modeling, environmental perception and ecological response data, it provides auxiliary judgment support for pollution source tracing, path simulation and response strategies. In the existing technical framework, the system usually combines water flow information, water quality parameter change trends and ecological species behavior characteristics to simulate and restore the development process of pollution events. Among them, fish, as ecological indicator organisms with high sensitivity to pollution stimuli, have been widely used in the identification of water environment status, and their abnormal behavior is often used as an important reference for the occurrence of pollution.

[0003] In order to more accurately simulate the behavioral feedback of fish in pollution events, existing digital twin systems generally set the corresponding expected behavioral response time for specific fish species based on previous monitoring experience and experimental data, and use it as a time reference for the pollution behavior trigger mechanism. Although such settings can reflect the delayed perception of pollutants by fish to a certain extent, they are mostly fixed in the model in a static manner and lack a dynamic comparison mechanism with the actual observed behavior. In a complex water environment, the actual response time of fish may be affected by multiple factors such as the intensity of pollutant release, the degree of water disturbance, and the trend of conductivity changes, resulting in a behavioral process that deviates from the original setting, which in turn causes the model output results to be inconsistent with the actual ecological response. Summary of the invention

[0004] The purpose of the present invention is to provide a water pollution control method and system based on digital twins, aiming to solve the problems raised in the background technology.

[0005] The present invention is implemented as follows: a water pollution control method based on digital twins, the method comprising: Obtain the initial pollution response sensitivity parameters for the target fish species in the monitored waters based on digital twin technology, and obtain the historical behavior monitoring data related to the target fish species in the monitored waters within a set time interval after the pollution incident, as well as the conductivity time series change data; Analyze historical behavior monitoring data to screen out a preset number of group escape behavior events of target fish species that exceed a predefined behavior size threshold; Analyze each group escape behavior event in turn to determine whether the proportion of events with group response lag behavior characteristics exceeds the set proportion threshold. If so, calculate the corresponding average response delay time, and construct a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species; Based on the conductivity time series change data, the conductivity gradient change rate before and after the pollution event is extracted, and the conductivity sensitivity correction coefficient is constructed according to the conductivity gradient change rate; The response hysteresis correction factor is combined with the conductivity sensitivity correction factor to correct the initial pollution response sensitivity parameter.

[0006] As a further limitation of the technical solution of the embodiment of the present invention, the steps of analyzing each group escape behavior event in turn, determining whether the proportion coefficient of the events having group response lag behavior characteristics exceeds a set proportion threshold, and if so, calculating the corresponding average response delay time, and constructing a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species include: Analyze each group escape behavior event in turn to determine whether there is a behavior sequence in which the target fish species gather, stay in place for a certain period of time, and then escape collectively before escaping. If so, it is determined that the group escape behavior event has the behavior characteristics of group response lag; Determine whether the proportion coefficient of events with group response hysteresis behavior characteristics in a preset number of group escape behavior events exceeds a set proportion threshold. If so, obtain the hysteresis response time from the completion of the gathering behavior to the occurrence of the escape behavior in each group escape behavior event, and calculate the average of the hysteresis response time as the average response delay time; The response lag correction coefficient was constructed based on the average response delay duration and the standard response duration of fish corresponding to the target fish species.

[0007] As a further limitation of the technical solution of the embodiment of the present invention, the step of constructing a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species includes: Retrieving a fish response time benchmark model, and determining the pollution stimulus conditions of the pollution event, and matching the fish standard response time corresponding to the target fish species in the fish response time benchmark model based on the pollution stimulus conditions of the pollution event; The fish response time benchmark model is a response time mapping model constructed based on the standard behavioral response data of various fish species under known pollution stimulus conditions. The fish response time benchmark model selects environmental data formed without the influence of other interference factors as a reference basis, and is used to characterize the expected response time of different fish species to pollution events under standard environmental parameter conditions; The difference between the average response delay time and the matched standard response time of the fish was quantified to obtain the response lag correction coefficient.

[0008] As a further limitation of the technical solution of the embodiment of the present invention, the steps of extracting the conductivity gradient change rate before and after the pollution event based on the conductivity time series change data, and constructing the conductivity sensitivity correction coefficient according to the conductivity gradient change rate include: Based on the conductivity time series change data, the conductivity change values ​​at several consecutive time nodes before and after the pollution event are extracted, and the conductivity gradient change rate within the pollution event time interval is calculated accordingly; Determine the degree to which the rate of change of the conductivity gradient affects the behavioral response sensitivity of the target fish species and use the degree of influence as an adjustment factor; The conductivity sensitivity correction factor is calculated by combining the adjustment coefficient with the conductivity gradient change rate.

[0009] As a further limitation of the technical solution of the embodiment of the present invention, the step of combining the response hysteresis correction coefficient with the conductivity sensitivity correction coefficient to correct the initial pollution response sensitivity parameter includes: Retrieve the preset parameter correction formula, and substitute the response hysteresis correction coefficient and the conductivity sensitivity correction coefficient as input variables into the parameter correction formula, jointly correct the initial pollution response sensitivity parameter, and obtain the corrected pollution response sensitivity parameter; The corrected pollution response sensitivity parameters are applied to the pollution path inversion model in the digital twin system.

[0010] As a further limitation of the technical solution of the embodiment of the present invention, the parameter correction formula is: ,in Refers to the corrected pollution response sensitivity parameter, Refers to the initial contamination response sensitivity parameter, Refers to the average response delay time, Refers to the standard response time of fish. Refers to the response lag correction coefficient, which is the difference between the average response delay time and the standard response time of the matched fish. refers to the rate of change of conductivity gradient, Refers to the adjustment coefficient corresponding to the conductivity gradient change rate, Refers to the conductivity sensitivity correction factor, and They refer to the correction weighting factors for the response hysteresis correction factor and the conductivity sensitivity correction factor respectively.

[0011] A water pollution control system based on digital twins, the system comprising: a data acquisition module, a historical data analysis module, a first correction coefficient determination module, a second correction coefficient determination module and a sensitivity parameter correction module, wherein: The data acquisition module is used to obtain the initial pollution response sensitivity parameters for the target fish species in the monitored waters based on the digital twin technology, and to obtain the historical behavior monitoring data related to the target fish species in the monitored waters within a set time interval after the pollution incident occurs, as well as the conductivity time series change data; A historical data analysis module is used to analyze historical behavior monitoring data and screen out a preset number of group escape behavior events of target fish species that exceed a predefined behavior scale threshold; The first correction coefficient determination module is used to analyze each group escape behavior event in turn, determine whether the proportion coefficient of events with group response lag behavior characteristics exceeds a set proportion threshold, and if so, calculate the corresponding average response delay time, and construct a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species; A second correction coefficient determination module is used to extract the conductivity gradient change rate before and after the pollution event based on the conductivity time series change data, and construct a conductivity sensitivity correction coefficient according to the conductivity gradient change rate; The sensitivity parameter correction module is used to combine the response hysteresis correction coefficient with the conductivity sensitivity correction coefficient to correct the initial pollution response sensitivity parameter.

[0012] As a further limitation of the technical solution of the embodiment of the present invention, the first correction coefficient determination module specifically includes: The behavior analysis unit is used to analyze each group escape behavior event in turn to determine whether there is a behavior sequence in which the target fish species gathers, stays in place for a certain period of time, and then escapes collectively before escaping. If so, it is determined that the group escape behavior event has the behavior characteristics of group response lag; A delay time calculation unit is used to determine whether the proportion coefficient of events with group response hysteresis behavior characteristics in a preset number of group escape behavior events exceeds a set proportion threshold, and if so, obtain the delayed response time from the completion of the gathering behavior to the occurrence of the escape behavior in each group escape behavior event, and calculate the average of the delayed response time as the average response delay time; The correction coefficient construction unit is used to construct a response lag correction coefficient based on an average response delay time and a standard response time of fish corresponding to a target fish species.

[0013] As a further limitation of the technical solution of the embodiment of the present invention, the correction coefficient construction unit specifically includes: The model application subunit is used to retrieve the fish response time benchmark model and determine the pollution stimulus conditions of the pollution event, and match the fish standard response time corresponding to the target fish species in the fish response time benchmark model based on the pollution stimulus conditions of the pollution event; The fish response time benchmark model is a response time mapping model constructed based on the standard behavioral response data of various fish species under known pollution stimulus conditions. The fish response time benchmark model selects environmental data formed without the influence of other interference factors as a reference basis, and is used to characterize the expected response time of different fish species to pollution events under standard environmental parameter conditions; The duration difference quantification subunit is used to quantify the difference between the average response delay duration and the matched standard response duration of the fish to obtain a response lag correction coefficient.

[0014] As a further limitation of the technical solution of the embodiment of the present invention, the second correction coefficient determination module specifically includes: A change rate calculation unit is used to extract the conductivity change values ​​of a number of consecutive time nodes before and after the pollution event based on the conductivity time series change data, and calculate the conductivity gradient change rate within the pollution event time interval based on the conductivity change values; An adjustment coefficient determination unit is used to determine the degree of influence of the conductivity gradient change rate on the behavioral response sensitivity of the target fish species, and use the influence degree as the adjustment coefficient; The sensitivity correction coefficient calculation unit is used to calculate the conductivity sensitivity correction coefficient by combining the adjustment coefficient and the conductivity gradient change rate.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention dynamically corrects the initial pollution response sensitivity parameters in the digital twin system by introducing a response lag correction coefficient and a conductivity sensitivity correction coefficient, thereby solving the problem that the fish behavior simulation in the prior art lacks consideration of the behavior response delay characteristics and physical change factors of the water body.

[0016] The present invention is based on the recognition of the aggregation-stay-escape sequence in the escape behavior of fish groups, quantifies the hysteresis of group response, and combines the influence of the conductivity gradient change rate on the sensitivity of fish behavior in pollution events to construct a dual-factor correction mechanism, which effectively improves the adaptability and prediction accuracy of the pollution path inversion model to real ecological behavior. Through this correction method, the misjudgment rate caused by the idealized setting of the standard model can be significantly reduced, and more accurate tracking of complex pollution events can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a method provided by an embodiment of the present invention; Figure 2 A flowchart of calculating the average response delay duration in the method provided in an embodiment of the present invention; Figure 3 A flow chart of obtaining a response lag correction coefficient in the method provided in an embodiment of the present invention; Figure 4 A flow chart of obtaining a conductivity sensitivity correction coefficient in a method provided in an embodiment of the present invention; Figure 5 A flow chart of jointly correcting the initial pollution response sensitivity parameters in the method provided in an embodiment of the present invention; Figure 6 An application architecture diagram of a system provided by an embodiment of the present invention; Figure 7 A structural block diagram of a first correction coefficient determination module in a system provided by an embodiment of the present invention; Figure 8 A structural block diagram of a correction coefficient construction unit in a system provided by an embodiment of the present invention; Fig. 9 This is a structural block diagram of a second correction coefficient determination module in a system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] Figure 1 A flow chart of a method provided by an embodiment of the present invention is shown.

[0020] Specifically, a water pollution control method based on digital twins comprises the following steps: Step S100, obtain the initial pollution response sensitivity parameters for the target fish species in the monitored waters based on the digital twin technology, and simultaneously obtain the historical behavior monitoring data related to the target fish species in the monitored waters within a set time interval after the pollution incident occurs, as well as the conductivity time series change data.

[0021] In an embodiment of the present invention, the present invention is based on a typical application path of the pollution source tracing model in the existing digital twin technology, that is, by dynamically modeling the behavioral response of fish in the monitored waters when a pollution incident occurs, it is used to invert the pollutant diffusion path and the possible release source location. This path is usually manifested in the existing system as follows: the system records the collective avoidance behavior of fish schools in different spatial areas in the target waters through sensors or video behavior recognition methods, and then combines water flow rate, pollution diffusion simulation and other data to construct a pollutant source tracing model. In this type of model, whether the fish quickly escape from the polluted area is one of the important indicators for judging the intensity and scope of the pollution. The pollution response sensitivity parameter is set and referenced in this context.

[0022] The initial pollution response sensitivity parameter refers to the intensity of the possibility of behavioral response to pollutants within a given time frame based on the average behavioral response of the target fish species to specific pollution stimuli under standard environmental conditions. It is used to characterize the representativeness of the fish species as a "signal body" of ecological response to pollution events. This parameter is usually a standard value set by a priori rules within the system model or constructed through a historical behavior library. It is a common technical achievement in the prior art that can be calculated by the ecological response submodel or behavior simulation module in the digital twin model. Its generation depends on existing experimental data, ecological behavior laws or expert knowledge rules, and has a certain degree of maturity.

[0023] Target fish species do not refer to any fish, but should be fish populations that are representative in the monitored waters, have obvious response characteristics, or have been the subject of environmental monitoring for a long time. Such fish generally have characteristics such as high sensitivity to pollution events, obvious clustering behavior, and easy observation and identification, such as some freshwater fish, grass carp, carp, or indicator species selected based on the ecological characteristics of specific waters. The selection of target fish species is usually determined in combination with the ecological structure of the waters and the modeling objectives of the monitoring system, and has a certain degree of directionality and analytical pertinence.

[0024] Historical behavior monitoring data refers to structured data on the group behavior of target fish species collected by underwater video recognition systems, sonar imaging equipment or behavior sensors within a set time interval after a pollution incident. Such data should at least include information on the location distribution, speed changes, whether aggregation occurs, whether synchronous escape occurs, and the spatial location when behavioral changes occur of target fish species at different time points, so as to be used for subsequent screening of collective escape events and extraction of behavioral feature sequences.

[0025] Conductivity time series change data refers to the continuous record of water conductivity changes over time collected by conductivity sensors deployed at multiple locations in the monitored waters before and after the pollution event. This data usually includes the measured conductivity values ​​at multiple time nodes, the spatial distribution of conductivity changes, the rate of change, and the corresponding timestamps, which facilitates the system to analyze the conductivity change gradient and construct correction factors in the future. Conductivity data can be collected in real time through water quality multi-parameter sensors, or retrieved through the historical water quality monitoring platform, and should have a certain time resolution and spatial coverage.

[0026] Furthermore, the water pollution control method based on digital twins also includes the following steps: Step S200, parsing historical behavior monitoring data, and screening out a preset number of group escape behavior events of target fish species that exceed a predefined behavior scale threshold.

[0027] In the embodiment of the present invention, the above screening process is a structured processing of the original behavior observation data and the extraction of high-response behavior segments, which lays a foundation for subsequent behavior sequence analysis and response hysteresis judgment.

[0028] Among them, the preset number refers to the number of target event samples used by the system for subsequent statistical analysis, and its setting basis is usually the magnitude of empirical data, system computing power, and statistical stability requirements. Usually no less than 10, preferably 20 to 50 escape behavior events are selected to ensure that the results of subsequent judgments on the characteristics of the lagging behavior have sufficient statistical representativeness and regularity recognition capabilities. The preset number should not be too small to avoid misjudgment due to insufficient samples; nor should it be too large to avoid introducing too many interfering behavior fragments and increasing the burden on the system.

[0029] The predefined behavior scale threshold refers to one of the judgment conditions that the system refers to when determining whether a certain behavior segment constitutes a "collective escape behavior event". It is usually used to screen whether the intensity or range of the group response behavior reaches a recognizable level. The threshold can be set based on the average number of clusters of the target fish species, the minimum number of individuals involved in a single escape behavior, or the degree of synchronization of the escape behavior. For example, if a group escape behavior involves only a small number of individuals, or only manifests as individual changes in swimming speed, it may not be sufficient to constitute a collective response event and is not representative for model correction.

[0030] Furthermore, the water pollution control method based on digital twins also includes the following steps: Step S300, analyze each group escape behavior event in turn, determine whether the proportion coefficient of events with group response lag behavior characteristics exceeds the set proportion threshold, if so, calculate the corresponding average response delay time, and construct a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species.

[0031] Specifically, Figure 2 A flow chart for calculating the average response delay time is shown.

[0032] Among them, each group escape behavior event is analyzed in turn to determine whether the proportion coefficient of events with group response lag behavior characteristics exceeds the set proportion threshold. If so, the corresponding average response delay time is calculated, and the response lag correction coefficient is constructed based on the average response delay time and the fish standard response time corresponding to the target fish species. Specifically, the following steps are included: Step S301, analyzing each group escape behavior event in turn, determining whether there is a behavior sequence in which the target fish species gathers, stays in place for a certain period of time before escaping, and then escapes collectively. If so, determining that the group escape behavior event has the behavior characteristics of group response hysteresis; Step S302, determining whether the proportion coefficient of events with group response hysteresis behavior characteristics in a preset number of group escape behavior events exceeds a set proportion threshold, and if so, obtaining the hysteresis response time from the completion of the gathering behavior to the occurrence of the escape behavior in each group escape behavior event, and calculating the average of the hysteresis response time as the average response delay time; Step S303, constructing a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species.

[0033] In an embodiment of the present invention, in step S301, each group escape behavior event is analyzed in turn to determine whether there is a behavior sequence in which the target fish species gathers, stays in place, and then escapes collectively after a certain period of time before the escape behavior occurs. This process is the key recognition logic for the hysteresis of group response. The specific implementation method can be based on existing video behavior recognition technology, machine vision technology or behavior pattern recognition algorithm. By performing target detection, path tracking and spatial aggregation degree analysis on the time series images in the historical behavior monitoring data, data such as the relative distance change curve between individual fish, the group center stability index, and the local density change curve can be extracted to identify "aggregation behavior"; combined with the analysis of individual speed changes, it can be determined whether there is a stay-in-place state, and then the occurrence of collective escape behavior can be identified through the consistency of escape direction and speed transition. The recognition process of this type of behavior sequence can be constructed with the help of image recognition + rule judgment model, which is a mature application of behavior recognition algorithms in the field of ecological observation in the prior art.

[0034] In this step, it is determined that "if the above behavior sequence exists, it is determined that the group escape behavior event has the behavioral characteristics of group response lag", and its significance lies in clearly distinguishing the target behavior from ordinary escape behavior. Not all escape behaviors have obvious response delay characteristics. Some fish individuals may respond quickly after the pollution stimulus occurs, and there is no obvious observation period or aggregation process. Therefore, by introducing the judgment logic of this characteristic behavior sequence, it is helpful to screen out group events that truly have "slow response" characteristics, ensuring that the delay time extracted in subsequent calculations is a representative and ecologically significant sample, rather than an error caused by individual behavior fluctuations.

[0035] In step S302, the system determines whether the proportion of events with group response hysteresis behavior characteristics in a preset number of group escape behavior events exceeds the set proportion threshold. This step is used to confirm whether group response hysteresis is a common behavior characteristic of the target fish species in the monitored waters. If hysteresis behavior exists in only a small number of escape events, it may be an occasional or non-representative behavior, and the initial parameters should not be systematically corrected. Only when this behavior reaches a certain proportion in the sample (for example, exceeding 30%, 50% and other empirical thresholds), it indicates that the group generally has a delayed response phenomenon under the current water conditions, thereby having the necessity and rationality to adjust the pollution response sensitivity parameters. Therefore, this judgment step is used to enhance the robustness of hysteresis feature recognition and prevent improper corrections due to individual behavior samples.

[0036] The average response delay time refers to the average length of time from the completion of the aggregation behavior of the target fish species to the final collective escape behavior in several events with group response lag behavior characteristics. This time value objectively reflects the "behavioral hysteresis period" experienced by the pollution event in inducing the actual observable response of the target fish species, and is the core difference data between the actual response and the model standard response. This value not only reflects the reaction speed of fish to pollution stimuli, but is also affected by factors such as the complexity of the water environment, the stability of the fish population, and the strength of the perceived stimulus. Therefore, it is the direct basis for constructing the response lag correction coefficient. By comparing this time with the standard response time of the target fish species, the degree of response delay can be accurately quantified, thereby guiding the amplitude and direction of the adjustment of the sensitivity parameters.

[0037] Specifically, Figure 3 A flow chart for obtaining the response lag correction coefficient is shown.

[0038] Among them, constructing a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species specifically includes the following steps: Step S3031, calling the fish response time benchmark model, and determining the pollution stimulus conditions of the pollution event, and matching the fish standard response time corresponding to the target fish species in the fish response time benchmark model based on the pollution stimulus conditions of the pollution event; The fish response time benchmark model is a response time mapping model constructed based on the standard behavioral response data of various fish species under known pollution stimulus conditions. The fish response time benchmark model selects environmental data formed without the influence of other interference factors as a reference basis, and is used to characterize the expected response time of different fish species to pollution events under standard environmental parameter conditions; Step S3032, quantifying the difference between the average response delay time and the matched standard response time of the fish to obtain a response lag correction coefficient.

[0039] In an embodiment of the present invention, the fish response time benchmark model is a structured data model for storing and mapping the standard behavioral response time of various fish species under different pollution stimulus conditions, and its core source is the existing fish behavior experimental data, pollution response simulation experiments and historical ecological observation records in the field of ecological environment. The construction basis of this model is the group behavior response characteristics of the target fish species to known pollution factors (such as heavy metals, organic matter, temperature anomalies, etc.) under an ideal experimental environment (i.e., without interference from other non-pollution factors and stable environmental parameters), which usually include: the time point of occurrence of escape behavior, the time required for the maximum response amplitude, the starting time point of group synchronization, etc. This model can be constructed by existing technical means, and currently has similar structural forms in smart water affairs, water ecological simulation, and intelligent monitoring platforms, such as fish response rule libraries and water quality event response libraries in ecological simulation systems, so it has strong practical feasibility.

[0040] Pollution stimulus conditions refer to the specific manifestations of pollution factors that cause behavioral responses of target fish species in pollution events, mainly including pollutant types (such as ammonia nitrogen, heavy metals, pesticides, etc.), pollution concentration ranges, release methods (continuous, sudden), diffusion patterns, and affected spatial ranges. As an important influencing factor of fish response behavior, this condition is a key index field for matching in the benchmark model. In practical applications, the system can classify pollution stimulus conditions and form a standard description based on real-time detection data of pollution events, pollutant concentration logs, monitoring point traceability data, or historical pollution event label information.

[0041] During the matching process, the system uses the pollution stimulus conditions of the pollution event as input variables, retrieves the associated fields in the fish response time benchmark model, and performs dual-condition matching based on the identification information of the target fish species, that is, the correspondence between the fish species and the pollution factor is satisfied at the same time, and the corresponding standard response time is extracted. If the pollution stimulus conditions are completely matched, the corresponding value is directly extracted; if only partially matched, the standard response time under the closest conditions can be obtained through similarity calculation or the set error tolerance rules to ensure the scope of application and flexibility of the model.

[0042] In step S3032, the difference between the average response delay time and the matched standard response time of the fish is quantitatively calculated, and the difference reflects the degree of deviation between the behavioral response of the target fish species under the current pollution event and the standard behavior, which has important reference value for regulation. After relativization, the difference can be converted into a response lag correction coefficient, which is used to adjust the initial pollution response sensitivity parameters set in the digital twin system to make up for the local lack of adaptability of the standard model in specific waters or specific fish populations.

[0043] By introducing a response lag correction coefficient, the system can dynamically adapt to the actual situation of fish responses under different ecological conditions, avoid judgment bias caused by over-idealization of the standard model, and improve the fit of the pollution path tracking model to the actual ecological response, thereby enhancing the accuracy of pollution tracing and the overall environmental adaptability of the system.

[0044] Furthermore, the water pollution control method based on digital twins also includes the following steps: Step S400: extracting the conductivity gradient change rate before and after the pollution event based on the conductivity time series change data, and constructing a conductivity sensitivity correction coefficient according to the conductivity gradient change rate.

[0045] Specifically, Figure 4 A flow chart for obtaining the conductivity sensitivity correction factor is shown.

[0046] Among them, extracting the conductivity gradient change rate before and after the pollution event based on the conductivity time series change data, and constructing the conductivity sensitivity correction coefficient according to the conductivity gradient change rate specifically includes the following steps: Step S401, based on the conductivity time series change data, extract the conductivity change values ​​of several consecutive time nodes before and after the pollution event occurs, and calculate the conductivity gradient change rate within the pollution event time interval based on the values; Step S402, determining the influence of the conductivity gradient change rate on the behavioral response sensitivity of the target fish species, and using the influence as an adjustment coefficient; Step S403: combining the adjustment coefficient and the conductivity gradient change rate to calculate the conductivity sensitivity correction coefficient.

[0047] In an embodiment of the present invention, in step S401, the system first extracts data points of several consecutive time nodes before and after the pollution event from the conductivity time series change data based on the time tag of the pollution event, preferably including the conductivity values ​​within 5 to 10 minutes before the pollution occurs and 10 to 30 minutes after the pollution occurs, to form a conductivity sequence containing timestamps and measured conductivity values. This process can use the existing multi-sensor time series data aggregation technology, automatically extract the time window through the pollution event trigger mechanism, and smooth the sequence in combination with the sliding window or weighted average algorithm to eliminate extreme value interference. After extracting the valid data segment, the system calculates according to the slope of the conductivity changing with time, and uses a linear fitting method or a local interpolation fitting method to calculate the conductivity gradient change rate in the key time period of the pollution event as a physical indicator to measure the trend of pollution intensity changes. The technical means involved in this process include but are not limited to: sensor data time synchronization, time series fitting, differential calculation, data noise reduction, etc., which are mature methods widely used in existing environmental monitoring systems and data analysis platforms.

[0048] In step S402, the system further determines the degree of influence of the conductivity gradient change rate on the behavioral response sensitivity of the target fish species according to the conductivity gradient change rate, and uses the degree of influence as the adjustment coefficient. The specific implementation method is: the system calls the constructed sensitivity response mapping model, which is constructed based on historical pollution event data and reflects the behavioral response intensity relationship of different fish species under different conductivity change rate intervals. The system uses the conductivity change rate actually calculated in the pollution event as the input value, and matches the behavioral response sensitivity influence factor corresponding to the target fish species in the mapping model as the conductivity adjustment coefficient of this event. The mapping model can be implemented by a piecewise function, a regression model or a table lookup method. For example, when the conductivity change rate is in the range of [0,5] μS / cm / min, the reaction sensitivity of the target fish is maintained at a baseline value of 1.0, and when the rate is 10~15 μS / cm / min, the sensitivity is amplified to 1.4, and so on. After obtaining the adjustment coefficient, the system can further normalize it to ensure numerical stability.

[0049] In step S403, the system combines the adjustment coefficient obtained in step S402 with the conductivity gradient change rate to calculate the conductivity sensitivity correction coefficient. The correction coefficient is used to indicate whether the behavioral response of the target fish species in the current pollution event shows a sensitivity difference due to the conductivity changing too fast or too slow. The specific calculation method can adopt a linear weighted model, or select a product model or an exponential function according to the specific application.

[0050] The significance of this calculation is that by integrating the correlation between the rate of change of water conductivity and the sensitivity of fish behavior, the system can dynamically correct the pollution response model according to the physical characteristics of water disturbance under different pollution events, avoiding misjudgment or sensitivity imbalance caused by ignoring physical background changes, thereby improving the digital twin model's response capability and simulation accuracy to multidimensional pollution events.

[0051] Furthermore, the water pollution control method based on digital twins also includes the following steps: Step S500: combining the response hysteresis correction coefficient with the conductivity sensitivity correction coefficient to correct the initial pollution response sensitivity parameter.

[0052] Specifically, Figure 5 A flow chart for joint correction of initial pollution response sensitivity parameters is shown.

[0053] The correction of the initial pollution response sensitivity parameter by combining the response hysteresis correction coefficient with the conductivity sensitivity correction coefficient specifically includes the following steps: Step S501, calling a preset parameter correction formula, and substituting the response hysteresis correction coefficient and the conductivity sensitivity correction coefficient as input variables into the parameter correction formula, jointly correcting the initial pollution response sensitivity parameter, and obtaining the corrected pollution response sensitivity parameter; Step S502: Apply the corrected pollution response sensitivity parameters to the pollution path inversion model in the digital twin system.

[0054] The parameter correction formula is: ,in Refers to the corrected pollution response sensitivity parameter, Refers to the initial contamination response sensitivity parameter, Refers to the average response delay time, Refers to the standard response time of fish. Refers to the response lag correction coefficient, which is the difference between the average response delay time and the standard response time of the matched fish. refers to the rate of change of conductivity gradient, Refers to the adjustment coefficient corresponding to the conductivity gradient change rate, Refers to the conductivity sensitivity correction factor, and They refer to the correction weighting factors for the response hysteresis correction factor and the conductivity sensitivity correction factor respectively.

[0055] In the embodiment of the present invention, the steps S501 and S502 are used to jointly process the response hysteresis correction coefficient and the conductivity sensitivity correction coefficient, and dynamically correct the initial pollution response sensitivity parameters according to the set parameter correction formula. The corrected pollution response sensitivity parameters will be used in the pollution path inversion model in the digital twin system to improve the accuracy of the simulation of the pollution diffusion process and the pollution source location results.

[0056] The significance of introducing these two correction factors together is that we no longer rely solely on single-source data to judge pollution response, but instead consider the joint influence of the ecological behavioral characteristics of the target fish species and the physical properties of the water body on the behavioral response ability. The response lag correction coefficient quantifies the deviation between the actual response of fish and the standard response in pollution events by analyzing the aggregation, retention and delayed escape phenomena that occur in the behavior of fish groups; while the conductivity sensitivity correction coefficient reflects the intensity of stimulation caused by rapid or drastic changes in conductivity in the water body, which directly affects whether fish will be more sensitive or slower to make escape responses. Taking these two correction factors into consideration together will help form a comprehensive adjustment model that takes into account both behavioral delays and environmental interference, and improve the adaptability and prediction accuracy of pollution response sensitivity parameters in complex real-world scenarios.

[0057] Taking a set of specific data as an example, it is assumed that the pollution response sensitivity parameter initially set by the system is 0.75. Through the time series analysis and calculation of the group escape behavior, the system determines that the target fish species generally have obvious behavioral response delays, and the response lag correction coefficient is set to 0.25. At the same time, during the pollution incident, the monitored conductivity change rate is high, and the conductivity sensitivity correction coefficient is determined to be 0.35 after model matching. The correction weights set by the system are: the response lag part accounts for 60%, and the conductivity part accounts for 40%. The two correction coefficients are multiplied by their respective weights and added together to obtain a total correction coefficient of 0.29. The system adjusts the correction coefficient in conjunction with the initial pollution response sensitivity parameter, and the corrected pollution response sensitivity parameter is 0.84. This value will be used as the final result and input into the pollution path inversion module of the digital twin system, so that the simulation system can more realistically reflect the current water ecological response characteristics when judging the pollutant transmission trend, traceability path and risk diffusion range, and enhance the accuracy and reliability of judgment.

[0058] Furthermore, Figure 6 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0059] Among them, in another preferred embodiment provided by the present invention, a water pollution control system based on digital twins includes: The data acquisition module 100 is used to obtain the initial pollution response sensitivity parameters for the target fish species in the monitored waters based on the digital twin technology, and to obtain the historical behavior monitoring data related to the target fish species in the monitored waters within a set time interval after the pollution incident occurs, as well as the conductivity time series change data.

[0060] Furthermore, the water pollution control system based on digital twins also includes: The historical data analysis module 200 is used to analyze the historical behavior monitoring data and screen out a preset number of group escape behavior events of target fish species that exceed a predefined behavior scale threshold.

[0061] Furthermore, the water pollution control system based on digital twins also includes: The first correction coefficient determination module 300 is used to analyze each group escape behavior event in turn, and determine whether the proportion coefficient of events with group response lag behavior characteristics exceeds a set proportion threshold. If so, the corresponding average response delay time is calculated, and a response lag correction coefficient is constructed based on the average response delay time and the fish standard response time corresponding to the target fish species.

[0062] Specifically, Figure 7 It shows a structural block diagram of the first correction coefficient determination module 300 in the system provided by an embodiment of the present invention.

[0063] In a preferred embodiment of the present invention, the first correction coefficient determination module 300 specifically includes: The behavior analysis unit 301 is used to analyze each group escape behavior event in sequence to determine whether there is a behavior sequence in which the target fish species gathers, stays in place for a certain period of time before escaping, and then escapes collectively. If so, it is determined that the group escape behavior event has the behavior characteristics of group response hysteresis; The delay time calculation unit 302 is used to determine whether the proportion coefficient of events with group response hysteresis behavior characteristics in a preset number of group escape behavior events exceeds a set proportion threshold, and if so, obtain the delayed response time from the completion of the gathering behavior to the occurrence of the escape behavior in each group escape behavior event, and calculate the average of the delayed response time as the average response delay time; The correction coefficient construction unit 303 is used to construct a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species.

[0064] Specifically, Figure 8 It shows a structural block diagram of the correction coefficient construction unit 303 in the system provided by the embodiment of the present invention.

[0065] In the preferred embodiment of the present invention, the correction coefficient construction unit 303 specifically includes: The model application subunit 3031 is used to retrieve the fish response time benchmark model, determine the pollution stimulus conditions of the pollution event, and match the fish standard response time corresponding to the target fish species in the fish response time benchmark model based on the pollution stimulus conditions of the pollution event; The fish response time benchmark model is a response time mapping model constructed based on the standard behavioral response data of various fish species under known pollution stimulus conditions. The fish response time benchmark model selects environmental data formed without the influence of other interference factors as a reference basis, and is used to characterize the expected response time of different fish species to pollution events under standard environmental parameter conditions; The duration difference quantification subunit 3032 is used to quantify the difference between the average response delay duration and the matched standard response duration of the fish to obtain a response lag correction coefficient.

[0066] Furthermore, the water pollution control system based on digital twins also includes: The second correction coefficient determination module 400 is used to extract the conductivity gradient change rate before and after the pollution event based on the conductivity time series change data, and construct the conductivity sensitivity correction coefficient according to the conductivity gradient change rate.

[0067] Specifically, Fig. 9 It shows a structural block diagram of the second correction coefficient determination module 400 in the system provided by an embodiment of the present invention.

[0068] In a preferred embodiment of the present invention, the second correction coefficient determination module 400 specifically includes: The change rate calculation unit 401 is used to extract the conductivity change values ​​of a number of consecutive time nodes before and after the pollution event based on the conductivity time series change data, and calculate the conductivity gradient change rate within the pollution event time interval accordingly; An adjustment coefficient determination unit 402 is used to determine the influence of the conductivity gradient change rate on the behavioral response sensitivity of the target fish species, and use the influence degree as the adjustment coefficient; The sensitivity correction coefficient calculation unit 403 is used to calculate the conductivity sensitivity correction coefficient by combining the adjustment coefficient and the conductivity gradient change rate.

[0069] Furthermore, the water pollution control system based on digital twins also includes: The sensitivity parameter correction module 500 is used to correct the initial pollution response sensitivity parameter by combining the response hysteresis correction coefficient with the conductivity sensitivity correction coefficient.

[0070] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0071] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0072] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0073] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A water pollution control method based on digital twins, characterized in that: The method comprises: Obtain the initial pollution response sensitivity parameters for the target fish species in the monitored waters based on digital twin technology, and obtain the historical behavior monitoring data related to the target fish species in the monitored waters within a set time interval after the pollution incident, as well as the conductivity time series change data; Analyze historical behavior monitoring data to screen out a preset number of group escape behavior events of target fish species that exceed a predefined behavior size threshold; Analyze each group escape behavior event in turn to determine whether the proportion of events with group response lag behavior characteristics exceeds the set proportion threshold. If so, calculate the corresponding average response delay time, and construct a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species; Based on the conductivity time series change data, the conductivity gradient change rate before and after the pollution event is extracted, and the conductivity sensitivity correction coefficient is constructed according to the conductivity gradient change rate; The response hysteresis correction factor is combined with the conductivity sensitivity correction factor to correct the initial pollution response sensitivity parameter.

2. The water pollution control method based on digital twin according to claim 1 is characterized in that: The steps of analyzing each group escape behavior event in turn, determining whether the proportion coefficient of the events with group response lag behavior characteristics exceeds the set proportion threshold, and if so, calculating the corresponding average response delay time, and constructing a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species include: Analyze each group escape behavior event in turn to determine whether there is a behavior sequence in which the target fish species gather, stay in place for a certain period of time, and then escape collectively before escaping. If so, it is determined that the group escape behavior event has the behavior characteristics of group response lag; Determine whether the proportion coefficient of events with group response hysteresis behavior characteristics in a preset number of group escape behavior events exceeds a set proportion threshold. If so, obtain the hysteresis response time from the completion of the gathering behavior to the occurrence of the escape behavior in each group escape behavior event, and calculate the average of the hysteresis response time as the average response delay time; The response lag correction coefficient was constructed based on the average response delay duration and the standard response duration of fish corresponding to the target fish species.

3. The water pollution control method based on digital twin according to claim 2 is characterized in that: The steps of constructing a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species include: Retrieving a fish response time benchmark model, and determining the pollution stimulus conditions of the pollution event, and matching the fish standard response time corresponding to the target fish species in the fish response time benchmark model based on the pollution stimulus conditions of the pollution event; The fish response time benchmark model is a response time mapping model constructed based on the standard behavioral response data of various fish species under known pollution stimulus conditions. The fish response time benchmark model selects environmental data formed without the influence of other interference factors as a reference basis, and is used to characterize the expected response time of different fish species to pollution events under standard environmental parameter conditions; The difference between the average response delay time and the matched standard response time of the fish was quantified to obtain the response lag correction coefficient.

4. The water pollution control method based on digital twin according to claim 3 is characterized in that: The steps of extracting the conductivity gradient change rate before and after the pollution event based on the conductivity time series change data and constructing the conductivity sensitivity correction coefficient according to the conductivity gradient change rate include: Based on the conductivity time series change data, the conductivity change values ​​at several consecutive time nodes before and after the pollution event are extracted, and the conductivity gradient change rate within the pollution event time interval is calculated accordingly; Determine the degree to which the rate of change of the conductivity gradient affects the behavioral response sensitivity of the target fish species and use the degree of influence as an adjustment factor; The conductivity sensitivity correction factor is calculated by combining the adjustment coefficient with the conductivity gradient change rate.

5. The water pollution control method based on digital twin according to claim 4 is characterized in that: The steps of combining the response hysteresis correction factor with the conductivity sensitivity correction factor to correct the initial pollution response sensitivity parameter include: Retrieve the preset parameter correction formula, and substitute the response hysteresis correction coefficient and the conductivity sensitivity correction coefficient as input variables into the parameter correction formula, jointly correct the initial pollution response sensitivity parameter, and obtain the corrected pollution response sensitivity parameter; The corrected pollution response sensitivity parameters are applied to the pollution path inversion model in the digital twin system.

6. The water pollution control method based on digital twin according to claim 5 is characterized in that: The parameter correction formula is: ,in Refers to the corrected pollution response sensitivity parameter, Refers to the initial contamination response sensitivity parameter, Refers to the average response delay time, Refers to the standard response time of fish. Refers to the response lag correction coefficient, which is the difference between the average response delay time and the standard response time of the matched fish. refers to the rate of change of conductivity gradient, Refers to the adjustment coefficient corresponding to the conductivity gradient change rate, Refers to the conductivity sensitivity correction factor, and They refer to the correction weighting factors for the response hysteresis correction factor and the conductivity sensitivity correction factor respectively.

7. A water pollution control system based on digital twins, characterized in that: The system comprises: a data acquisition module, a historical data analysis module, a first correction coefficient determination module, a second correction coefficient determination module and a sensitivity parameter correction module, wherein: The data acquisition module is used to obtain the initial pollution response sensitivity parameters for the target fish species in the monitored waters based on the digital twin technology, and to obtain the historical behavior monitoring data related to the target fish species in the monitored waters within a set time interval after the pollution incident occurs, as well as the conductivity time series change data; A historical data analysis module is used to analyze historical behavior monitoring data and screen out a preset number of group escape behavior events of target fish species that exceed a predefined behavior scale threshold; The first correction coefficient determination module is used to analyze each group escape behavior event in turn, determine whether the proportion coefficient of events with group response lag behavior characteristics exceeds a set proportion threshold, and if so, calculate the corresponding average response delay time, and construct a response lag correction coefficient based on the average response delay time and the fish standard response time corresponding to the target fish species; A second correction coefficient determination module is used to extract the conductivity gradient change rate before and after the pollution event based on the conductivity time series change data, and construct a conductivity sensitivity correction coefficient according to the conductivity gradient change rate; The sensitivity parameter correction module is used to combine the response hysteresis correction coefficient with the conductivity sensitivity correction coefficient to correct the initial pollution response sensitivity parameter.

8. The water pollution control system based on digital twin according to claim 7 is characterized in that: The first correction coefficient determination module specifically includes: The behavior analysis unit is used to analyze each group escape behavior event in turn to determine whether there is a behavior sequence in which the target fish species gathers, stays in place for a certain period of time, and then escapes collectively before escaping. If so, it is determined that the group escape behavior event has the behavior characteristics of group response lag; A delay time calculation unit is used to determine whether the proportion coefficient of events with group response hysteresis behavior characteristics in a preset number of group escape behavior events exceeds a set proportion threshold, and if so, obtain the delayed response time from the completion of the gathering behavior to the occurrence of the escape behavior in each group escape behavior event, and calculate the average of the delayed response time as the average response delay time; The correction coefficient construction unit is used to construct a response lag correction coefficient based on an average response delay time and a standard response time of fish corresponding to a target fish species.

9. The water pollution control system based on digital twin according to claim 8 is characterized in that: The correction coefficient construction unit specifically includes: The model application subunit is used to retrieve the fish response time benchmark model and determine the pollution stimulus conditions of the pollution event, and match the fish standard response time corresponding to the target fish species in the fish response time benchmark model based on the pollution stimulus conditions of the pollution event; The fish response time benchmark model is a response time mapping model constructed based on the standard behavioral response data of various fish species under known pollution stimulus conditions. The fish response time benchmark model selects environmental data formed without the influence of other interference factors as a reference basis, and is used to characterize the expected response time of different fish species to pollution events under standard environmental parameter conditions; The duration difference quantification subunit is used to quantify the difference between the average response delay duration and the matched standard response duration of the fish to obtain a response lag correction coefficient.

10. The water pollution control system based on digital twin according to claim 9 is characterized in that: The second correction coefficient determination module specifically includes: A change rate calculation unit is used to extract the conductivity change values ​​of a number of consecutive time nodes before and after the pollution event based on the conductivity time series change data, and calculate the conductivity gradient change rate within the pollution event time interval based on the conductivity change values; An adjustment coefficient determination unit is used to determine the degree of influence of the conductivity gradient change rate on the behavioral response sensitivity of the target fish species, and use the influence degree as the adjustment coefficient; The sensitivity correction coefficient calculation unit is used to calculate the conductivity sensitivity correction coefficient by combining the adjustment coefficient and the conductivity gradient change rate.

Images