Multi-parameter linkage water quality sampling method and system

Abstract

The invention discloses a multi-parameter linkage water quality sampling method and system, and relates to the field of water quality sampling, and the method comprises the steps: obtaining hydrological data, main transportation path data and pollutant discharge data; setting water area sampling points of the target monitoring water area according to the hydrological data, the pollutant emission data and the main transportation path data; water quality sampling is conducted on the water area surface sampling points and the water area vertical sampling points; verifying the validity of the sampling point data, and determining the validity of the water area surface sampling points and the water area vertical sampling points according to the validity verification result; acquiring water quality monitoring data of the water area surface sampling points and the water area vertical sampling points; calculating a pollution source linkage index according to the water quality monitoring data, and determining a pollution emission event type; determining a time sequence phase definition for the pollution event based on the pollution emission event type and the water quality monitoring data; and determining the corresponding water quality sampling frequency and water quality sampling volume by combining with the time sequence stage definition. The accuracy of water quality monitoring can be effectively improved.

Description

Technical Field

[0001] This application relates to the field of water quality sampling, and in particular to a multi-parameter linkage water quality sampling method and system. Background Technology

[0002] In the field of aquatic environment monitoring, intermittent, accidental, or excessive emissions from industrial enterprises can rapidly alter the physicochemical and biological characteristics of receiving water bodies, inducing a series of ecological and health risks such as decreased dissolved oxygen, accumulation of heavy metals and organic pollution, and abnormal algal blooms. Water quality testing is a fundamental means of pollution identification, concentration tracking, and spatial diffusion assessment. It can provide evidence support for regulatory enforcement, provide quantitative basis for enterprise pollution control and process optimization, and provide decision-making conditions for early warning and emergency dispatch of emergencies.

[0003] Current water quality monitoring and sampling systems employ a fixed-site, timed sampling model. However, this model struggles to capture the spatial heterogeneity and temporal fluctuations of water bodies. Fixed sites are typically located at the water's edge (shore), on bridges, or in historically polluted areas. Monitoring data from these locations cannot accurately reflect the water quality in the center of the water body, at different depths, or in hidden bays, resulting in inaccurate data reflecting the situation in the monitored area. Furthermore, timed sampling (e.g., weekly or monthly) misses rapid changes in water quality. For example, surface runoff from the initial stages of rainfall carries large amounts of pollutants into rivers, tidal forces cause periodic fluctuations in water quality, and algal blooms can erupt and subside within days. These critical dynamic processes cannot be captured in timed sampling data, leading to an inability to reflect the overall pollution distribution in the water body. This, in turn, affects the accuracy of water quality monitoring, resulting in inefficient allocation of monitoring resources, insufficient spatiotemporal data integrity, and delayed decision-making. Summary of the Invention

[0004] This application provides a multi-parameter linkage water quality sampling method and system to improve the accuracy of water quality monitoring.

[0005] To achieve the above objectives, the embodiments of this application adopt the following technical solutions: Firstly, a multi-parameter linkage water quality sampling method is provided, including the following steps: Acquire hydrological data and main transport route data of the target monitoring water area, as well as pollutant emission data of the target monitoring enterprises; Based on hydrological data, pollutant discharge data, and main transport path data, water sampling points are set for the target monitoring water area. The water sampling points include water surface sampling points and water vertical sampling points. Water quality samples were collected at both the surface sampling point and the vertical sampling point of the water area to obtain sampling point data; The validity of the sampling point data is verified to obtain the validity verification results, and the validity of the water surface sampling points and the water vertical sampling points is determined based on the validity verification results; When the surface sampling point and the vertical sampling point of the water body are valid sampling points, water quality monitoring data of the surface sampling point and the vertical sampling point of the water body are obtained; The pollution source linkage index is calculated based on water quality monitoring data, and the type of pollution discharge event is determined based on the pollution source linkage index. Pollution events are classified and their temporal stages are defined based on the type of pollution discharge event and water quality monitoring data. Based on the temporal stage definition, the water quality sampling frequency and water quality sampling volume corresponding to each surface sampling point and vertical sampling point of the water area are determined respectively.

[0006] In one possible implementation of the first aspect, the step of setting water sampling points for the target monitoring water area based on hydrological data, pollutant discharge data, and main transport path data includes: Wind field data and surface flow field data are obtained from the hydrological data of the target monitoring water area; By monitoring the pollutant emission data of target enterprises, we can obtain the location of pollution sources, emission time, emission flow rate, and emission concentration. Emission flow and emission concentration are converted into pollutant release flux, and wind field data are converted into wind stress according to a preset wind stress conversion rate; The simulation time window is set in conjunction with the emission time; Using the location of the pollution source as the starting point of diffusion, a migration and diffusion model is constructed within the simulation time window by combining data on pollutant release flux, wind stress, and surface flow field. Pollutant isochrones were extracted based on a migration and diffusion model, and target intersections were extracted by combining pollutant isochrones with main transport path data. The surface diffusion boundary and concentration gradient distribution of pollutants are calculated based on pollutant release flux, surface flow field data, and wind stress. Sampling points on the water surface were determined by combining the target intersection point, surface diffusion boundary, and concentration gradient distribution. Vertical sampling points for target monitoring water areas are set based on surface sampling points, hydrological data, and pollutant emission data.

[0007] In one possible implementation of the first aspect, the step of setting vertical sampling points for the target monitoring water area based on water surface sampling points, hydrological data, and pollutant discharge data includes: Water temperature profile data and water density profile data are obtained from the hydrological data of the target monitoring water area; Identify the water temperature gradient and water density gradient in the target monitoring area based on water temperature profile data and water density profile data; By combining the water temperature jump layer and the water density jump layer, the target monitoring water area is divided into an upper uniform layer, a gradient abrupt layer and a lower uniform layer. For any water surface sampling point, construct a surface vertical profile, and combine the surface vertical profile, upper uniform layer, gradient abrupt layer and lower uniform layer to determine the set of vertical candidate sampling points corresponding to each water surface sampling point; The priority vertical sampling layer is determined by filtering each set of vertical candidate sampling points based on the maximum information gain criterion. The types of pollutants emitted are determined by pollutant emission data, and the vertical distribution characteristics corresponding to the types of pollutants emitted are determined in a pre-set database; Vertical sampling points for the target monitoring water area are determined by combining vertical distribution characteristics and priority vertical sampling layers.

[0008] In one possible implementation of the first aspect, the sampling point data includes first time point data and second time point data. The step of validating the sampling point data to obtain a validity verification result, and determining the validity of the water surface sampling points and the water vertical sampling points based on the validity verification result, includes: For any water area sampling point, data at the first preset time point and data at the second preset time point are collected respectively. The relative deviation value is calculated based on the data from the first time point and the data from the second time point, and the first validity is determined by the relative deviation value. For any given water sampling point, obtain the original concentration of the water sample to be tested; The amount of the target pollutant spiked is obtained by adding a standard solution of the target pollutant of known concentration to the water sample to be tested. The concentration of the target pollutant after spiking is obtained by analyzing the water sample to be tested with a standard solution of the target pollutant at a known concentration. The second validity is calculated based on the spiked concentration of the target pollutant, the amount of the target pollutant spiked, and the original concentration. The overall validity is calculated based on the first and second validity levels; If the overall validity is greater than or equal to the preset validity threshold, then the corresponding water surface sampling point and water vertical sampling point are determined to be valid.

[0009] In one possible implementation of the first aspect, the step of classifying pollution events based on pollution discharge event type and water quality monitoring data to determine the temporal stages of pollution events includes: The concentration of pollutants, the rate of change of concentration, and the abnormal values ​​of pollutant concentration are obtained through water quality monitoring data. Based on the type of pollution emission event, select the corresponding type parameter group from the preset event type library. The type parameter group includes rising hysteresis threshold, falling hysteresis threshold, forgetting factor and minimum residence time. The forgetting factor is used to characterize the memory length of event energy for historical data and the weight decay rate. Based on the type parameter group, the type linkage index is calculated using a preset multidimensional fusion model. The type linkage index is used to characterize the degree of matching between the pollution emission event type and the target pollution type. A comprehensive anomaly score is obtained by weighted fusion calculation combining pollutant concentration, concentration change rate, and pollutant concentration anomalies. The event energy is calculated using a preset event energy formula by combining the forgetting factor and the comprehensive abnormality score. The time trigger moment and the time end moment are determined by combining the rising hysteresis threshold and the falling hysteresis threshold. Within the time trigger moment and time end moment, the event energy and type linkage index are combined with the minimum residence time to divide and determine the transient segment, transition segment and steady-state segment of the pollution emission event type. The transient segment, transition segment and steady-state segment are used to characterize the time sequence stage definition.

[0010] In one possible implementation of the first aspect, the pollution emission event types include volatile events and heavy particulate events, and the method further includes: When the pollution emission event type is a volatile event, obtain surface flow velocity data from sampling points on the water surface; The mean surface velocity is calculated based on the surface velocity data, and the sum of squares of velocity deviations is calculated based on the mean surface velocity. The surface disturbance index is calculated by combining the sum of squared velocity deviations and the number of sampling points on the water surface. The adjusted rise hysteresis threshold is obtained by combining the surface disturbance index with the threshold adjustment. When the pollution emission event type is a heavy particulate event, the heavy particulate concentrations of the upper homogeneous layer, gradient abrupt layer and lower homogeneous layer are obtained respectively. The absolute value of the heavy particle concentration between any two adjacent layers is calculated by combining the heavy particle concentrations of the upper homogeneous layer, the gradient abrupt layer, and the lower homogeneous layer, and the absolute values ​​are added together to obtain the stratification intensity index. The adjusted descent hysteresis threshold is obtained by combining the stratification intensity index with the threshold adjustment.

[0011] In one possible implementation of the first aspect, the step of defining the water quality sampling frequency and water quality sampling volume corresponding to each water surface sampling point and water vertical sampling point by combining the time-series stage definition includes: Determine the corresponding surface disturbance index and stratification intensity index for each time series stage; The corresponding surface disturbance index and stratification intensity index are defined for each time series stage to determine the corresponding surface disturbance level and vertical stratification level, where the surface disturbance level is high surface disturbance and low surface disturbance, and the vertical stratification level is high vertical stratification and low vertical stratification. Based on the definition of the corresponding surface disturbance level and vertical stratification level for each time series stage, the water quality sampling frequency and water quality sampling volume for surface sampling points and vertical sampling points in the water area are determined respectively.

[0012] In one possible implementation of the first aspect, the step of defining the corresponding surface disturbance level and vertical stratification level according to each time stage, and determining the water quality sampling frequency and water quality sampling volume of the water surface sampling point and the water vertical sampling point respectively, includes: When the timing phase is defined as the transient phase and the surface disturbance index is high, the water quality sampling frequency and water quality sampling volume corresponding to the water surface sampling point are determined by sampling using the first preset frequency and the third preset volume. When the timing phase is defined as the transient phase and the surface disturbance index is low, the water quality sampling frequency and water quality sampling volume corresponding to the water surface sampling point are determined by sampling using the second preset frequency and the second preset volume. When the timing phase is defined as the steady state phase, the water quality sampling frequency and water quality sampling volume corresponding to the sampling point on the water surface are determined by sampling using the fourth preset frequency and the first preset volume. When the timing phase is defined as a transient phase and the vertical stratification level is high vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the first preset frequency and the third preset volume. When the time sequence stage is defined as a transient segment and the vertical stratification level is low vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the second preset frequency and the second preset volume. When the time sequence stage is defined as a steady state and the vertical stratification level is high vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the third preset frequency and the first preset volume. When the time sequence stage is defined as a steady state and the vertical stratification level is low vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the fourth preset frequency and the first preset volume. Among them, the first preset frequency is greater than the second preset frequency, the second preset frequency is greater than the third preset frequency, and the third preset frequency is greater than the fourth preset frequency; the first preset volume is greater than the second preset volume, and the second preset volume is greater than the third preset volume.

[0013] Secondly, this application provides an electronic device, comprising: The memory is configured to store instructions; and The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the aforementioned multi-parameter linkage water quality sampling method.

[0014] Thirdly, this application provides a machine-readable storage medium storing instructions that cause a machine to perform the aforementioned multi-parameter linkage water quality sampling method.

[0015] The above technical solution firstly acquires hydrological data, main transport routes, and enterprise emission data, which clarifies the main transport channels and potential impact areas, avoiding blind sampling and ineffective sampling, and improving the targeting and interpretability of the solution. Setting water sampling points based on these three types of data ensures that the sampling points effectively cover the water areas potentially affected by pollution. Simultaneously setting surface and vertical sampling points creates a three-dimensional monitoring layout, which helps to understand the vertical and horizontal distribution patterns of pollutants in the water body, improving the accuracy of water quality monitoring. Water quality sampling at both surface and vertical sampling points yields abundant sampling data, capturing changes in water quality over time, providing strong support for analyzing the occurrence and development of pollution events. Validating the sampling point data allows for the selection of reliable water quality data, ensuring that subsequent analysis is based on accurate data. The validity of surface and vertical sampling points is determined based on the validity verification results, helping to promptly identify whether the sampling point settings are reasonable, improving overall data reliability and the accuracy of subsequent analysis, and avoiding misjudgments or omissions due to invalid samples. Subsequently, by calculating the pollution source linkage index based on water quality monitoring data, the relationship between pollutant emission sources and water quality changes can be quantified. This enables the identification and classification of pollution events, improving the accuracy of event identification and the timeliness of response. Defining the temporal stages of pollution events allows for a dynamic depiction of the pollution evolution process, facilitating subsequent adjustments to sampling strategies or conducting event tracing and risk assessments at different stages. Combining the temporal stages with the determination of the corresponding water quality sampling frequency and volume for each surface sampling point and vertical sampling point in the water area allows for the dynamic optimization of sampling resource allocation, balancing data richness and sampling cost efficiency. Furthermore, setting sampling volume ratios based on different stratigraphic levels improves the collection coverage of key polluted areas.

[0016] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0017] Figure 1 A schematic flowchart of a multi-parameter linkage water quality sampling method provided in this application embodiment; Figure 2This is a schematic diagram of the structure of a water sampling point for setting a target monitoring water area, provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0019] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0020] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0021] Figure 1 The illustration shows a schematic flowchart of a multi-parameter linkage water quality sampling method according to an embodiment of this application. Figure 1 As shown in the figure, this application provides a water quality sampling method with multi-parameter linkage, which may include the following steps.

[0022] S110. Obtain hydrological data and main transport route data of the target monitored water area, as well as pollutant emission data of the target monitored enterprises; S120. Based on hydrological data, pollutant discharge data and main transport path data, set water sampling points for the target monitoring water area. Water sampling points include water surface sampling points and water vertical sampling points. S130. Water quality samples were collected at both the surface sampling point and the vertical sampling point of the water area to obtain sampling point data; S140. Verify the validity of the sampling point data to obtain the validity verification results, and determine the validity of the water surface sampling points and the water vertical sampling points based on the validity verification results; S150. When the water surface sampling point and the water vertical sampling point are valid sampling points, acquire the water quality monitoring data of the water surface sampling point and the water vertical sampling point. S160. Calculate the pollution source linkage index based on water quality monitoring data, and determine the type of pollution discharge event based on the pollution source linkage index; S170. Based on the type of pollution discharge event and water quality monitoring data, classify pollution events and determine the temporal stage of the pollution event. S180. Based on the time sequence stage definition, determine the water quality sampling frequency and water quality sampling volume corresponding to each surface sampling point and vertical sampling point of the water area.

[0023] First, hydrological data for the target monitoring area is obtained through hydrological monitoring stations. This data includes information such as flow rate, velocity, water level, depth, temperature, and water exchange patterns, characterizing the hydrodynamic features of the target monitoring area. Second, the main transport pathway data for the target monitoring area is acquired. This pathway determines the primary transport and movement routes of substances (including water itself, dissolved substances, and suspended particles) within the water body. The main transport pathway data can include the path direction, describing the direction in which pollutants primarily move; for example, the main path from upstream to downstream in a river, or a specific circulation route in ocean currents. Finally, pollutant emission data from the target monitoring enterprises within the target monitoring area is obtained. This data includes the geographical location of the emission outlet, emission flow rate, emission frequency, and the type and concentration of pollutants, serving as input for pollution source terms. By acquiring this data, basic environmental and pollution source information for the target monitoring area can be obtained before sampling point deployment and monitoring strategy formulation, providing a basis for the scientific selection of subsequent water quality sampling points.

[0024] Secondly, water sampling points for the target monitoring water area are set. In this embodiment, water sampling points include surface sampling points and vertical sampling points. The setting of water sampling points for the target monitoring water area is achieved through multi-factor comprehensive analysis. First, hydrodynamic characteristics and water stratification characteristics are obtained based on the hydrological data of the target monitoring water area. The spatial location, emission time, and emission intensity of the pollution source are determined using pollutant emission data, and the possible migration direction of pollutants in the water area is analyzed in conjunction with the main transport path data. Subsequently, based on the above information, surface sampling points covering key areas are set to reflect the distribution characteristics of pollutants on the water surface. Surface sampling points refer to the locations on the surface area of ​​the water body designated for collecting water samples for analysis. Then, combining the vertical stratification structure of the water body and the vertical distribution characteristics of pollutants in the water body, vertical sampling points are set up in the vertical direction corresponding to each surface sampling point, thereby forming a scientific and comprehensive water sampling layout covering different layers of the water body to ensure that the collected water quality data can accurately reflect the spatial distribution characteristics of pollutants in the target monitoring water area. Vertical sampling points in water bodies refer to locations set vertically in the target monitoring water area for collecting water quality information. These points are deployed along the vertical direction of each surface sampling point, covering different depth layers of the water body. The deployment of vertical sampling points is based on the vertical stratification characteristics of the water body, such as temperature strata and density strata, the possible vertical distribution of pollutants, and hydrological conditions. The aim is to scientifically and comprehensively reflect the concentration distribution of pollutants and the patterns of water quality changes in the vertical direction of the water body.

[0025] Next, water quality sampling operations will be performed at the designated surface sampling points and vertical sampling points of the water body to obtain the corresponding sampling point data. The surface sampling points are used to reflect the water quality status of the surface layer of the water body, while the vertical sampling points are used to reflect the water quality status at different depths of the water body. The sampling point data can be the original water sample test data, such as COD, ammonia nitrogen, heavy metal content, etc.

[0026] To ensure the accuracy and representativeness of the data collected from the water sampling points, the validity of the water sampling points is verified. First, the sampling point data includes data at a first time point and data at a second time point. Sampling is performed at any water sampling point at both the first and second preset time points to obtain the corresponding first and second time point data. The time point of the first time point data is not equal to the time point of the second time point data. Next, the relative deviation value is calculated based on the two sets of data. The relative deviation value can be normalized using the absolute difference method to determine the first validity of the sampling point, thus evaluating its stability over time. Further, for any water sampling point, the original concentration of the water sample to be tested is obtained, and a target pollutant standard solution of known concentration is added to the water sample. In this embodiment, the target pollutant standard solution of known concentration refers to a pollutant standard solution prepared or purchased according to a preset standard method, with a clearly defined and traceable concentration value. The concentration value of the target pollutant contained in it is a pre-calibrated and determined value, which can be used as a reference for spiking during the analysis and detection process to ensure the accuracy and comparability of the concentration calculation and validity verification, thereby obtaining the target pollutant spiking amount. The spiked water sample is tested to obtain the spiked concentration of the target pollutant. A second validity is calculated based on the spiked concentration, the amount of the target pollutant spiked, and the original concentration to evaluate the sampling point's performance in terms of detection accuracy and method reliability. Then, the comprehensive validity of the sampling point is calculated by combining the first and second validitys, and compared with a preset validity threshold, which is set through statistical analysis of a large amount of historical monitoring data. When the comprehensive validity is greater than or equal to the preset validity threshold, the surface sampling point and the vertical sampling point are deemed valid, thus ensuring the scientific validity and reliability of the data provided by the sampling points.

[0027] Water quality monitoring is only performed at the sampling locations after both the surface sampling point and the vertical sampling point have been verified as valid, in order to obtain the corresponding water quality monitoring data. Water quality monitoring data includes, but is not limited to, pollutant concentrations, physical parameters such as water temperature, pH, conductivity, turbidity, and dissolved oxygen, and chemical parameters such as salinity, nutrients, chemical oxygen demand (COD), and biochemical oxygen demand (BOD). Various types of sensors can be used to acquire the relevant water quality monitoring data.

[0028] Next, the pollution source linkage index is calculated based on the water quality monitoring data, and the type of pollution discharge event is determined based on the pollution source linkage index. A database can be pre-established to record information such as the typical pollutant composition, proportion, and characteristic pollutants emitted by each potential pollution source. For example, this includes discharge outlets from different factories, different types of ship leaks, and different types of agricultural runoff, much like each pollution source has a unique DNA. The currently monitored water quality data is then compared and matched with the fingerprint of each pollution source in the database. Correlation analysis methods (such as the Pearson correlation coefficient) can be used to calculate the correlation between the monitoring data and each source fingerprint. The specific expression for the Pearson correlation coefficient is shown below:

[0029] Where r represents the Pearson correlation coefficient; xi and yi represent the observed values ​​of the i-th sample in the two variable dimensions, respectively; and represents the mean of the two sample variables respectively; n represents the number of samples.

[0030] The correlation between the two is calculated using the Pearson correlation coefficient formula, known as the pollution source linkage index. The higher the index, the stronger the correlation between the monitored pollution characteristics and the specific pollution source.

[0031] Based on pollution discharge event types and water quality monitoring data, pollution events are classified and their temporal stages are defined. According to the preset pollution discharge event types, the corresponding type parameter group is selected from the event type library. The type parameter group includes rising hysteresis threshold, falling hysteresis threshold, forgetting factor, and minimum residence time. The rising hysteresis threshold is the trigger value used to determine whether an event has started or whether a certain state has changed from "low" to "high". The falling hysteresis threshold is the release value used to determine whether an event has ended or whether a certain state has changed from "high" to "low". The forgetting factor is used to characterize the memory length of event energy for historical data and the rate of weight decay. That is, when calculating a certain indicator or state at the current moment, different weights are assigned to historical data. The more recent the data, the greater the weight, and the more distant the data, the smaller the weight. Moreover, this weight decay is exponential. The pre-defined pollution emission event types are obtained through analysis of historical pollution emission events, experimental research, and environmental management needs. Different emission patterns and their characteristic parameters can be compiled into an event type library, including typical concentration change curves, durations, and fluctuation amplitudes. This event type library can be used for subsequent pollution event identification, linkage index calculation, and time-series stage division. Then, based on the type parameter group, a pre-defined multi-dimensional fusion model is used to calculate the type linkage index, which characterizes the degree of matching between the pollution emission event type and the target pollution type. The pre-defined multi-dimensional fusion model is a complex computational framework or algorithm that has been designed, trained, and validated before practical application, and configured with all parameters. It can integrate multi-dimensional features such as pollutant concentration, concentration change rate, and concentration anomalies for comprehensive analysis and output a unified evaluation result. Based on a machine learning model, detailed monitoring data of past actual pollution events and their corresponding real pollution types can be collected, allowing the model to learn how to map input features to the type linkage index from the prepared data. The extracted multi-dimensional data is used as the model input, and the corresponding real pollution type or matching degree is used as the model output label. By optimizing algorithms (such as gradient descent) and adjusting the model's internal parameters (such as the weights and biases of the neural network), the predicted type linkage index is made as close as possible to the actual situation. Furthermore, a weighted fusion calculation is performed using pollutant concentration, concentration change rate, and outliers to obtain a comprehensive anomaly score. Subsequently, the comprehensive anomaly score and forgetting factor are used to calculate the event energy using a pre-defined event energy formula, and the event's trigger and end times are determined based on rising and falling hysteresis thresholds. Finally, within the trigger and end times, the event energy and type linkage index are divided based on the minimum dwell time to determine the transient, transitional, and steady-state phases of the pollution emission event, thus completing the temporal stage definition of the pollution event.

[0032] For each time-series stage of a pollution event, the water quality sampling frequency and volume for surface and vertical sampling points can be determined based on the surface disturbance level and vertical stratification level corresponding to each time-series stage. When the time-series stage is defined as a transient period and the surface disturbance level is high, surface sampling is performed using a first preset frequency and a third preset volume to determine the sampling frequency and volume of the surface sampling points; when the time-series stage is defined as a transient period and the surface disturbance level is low, surface sampling is performed using a second preset frequency and a second preset volume; when the time-series stage is defined as a steady-state period, surface sampling is performed using a fourth preset frequency and a first preset volume; when the time-series stage is defined as a transient period and the vertical stratification level is high, vertical sampling is performed using a first preset frequency and a third preset volume to determine the sampling frequency and volume of the vertical sampling points; when... When the time series phase is defined as a transient period and the vertical stratification level is low, vertical sampling of the water body is conducted using a second preset frequency and a second preset volume. When the time series phase is defined as a steady-state period and the vertical stratification level is high, vertical sampling of the water body is conducted using a third preset frequency and a first preset volume. When the time series phase is defined as a steady-state period and the vertical stratification level is low, vertical sampling of the water body is conducted using a fourth preset frequency and a first preset volume. The first preset frequency is greater than the second preset frequency, the second preset frequency is greater than the third preset frequency, and the third preset frequency is greater than the fourth preset frequency; the first preset volume is greater than the second preset volume, and the second preset volume is greater than the third preset volume. Through these steps, the sampling frequency and volume of surface and vertical sampling points can be dynamically adjusted according to the water body disturbance and vertical stratification characteristics of the pollution event's time series phase. This allows water quality sampling to accurately reflect the pollutant change characteristics of the water body at different time periods and different vertical levels, while improving sampling efficiency and optimizing resource utilization.

[0033] From acquiring hydrological and pollutant discharge data, to setting sampling points, conducting water quality sampling, verifying data validity, determining pollution event types and time series stages, and then dynamically adjusting sampling frequency and volume, a complete monitoring process has been formed, forming a systematic and comprehensive water quality monitoring system to ensure all-round control over water pollution. Sampling points are scientifically set based on hydrological, pollutant discharge, and main transport path data, and sampling frequency and volume are dynamically adjusted in conjunction with surface disturbance index and stratification intensity index at different time series stages. This allows for accurate capture of water quality changes while avoiding oversampling, rationally optimizing monitoring resource input, improving resource utilization efficiency, and achieving precise sampling and resource optimization. Strict validity verification is performed on sampling point data; only data from sampling points deemed valid are used for subsequent analysis, ensuring the accuracy and reliability of water quality monitoring data from the source and providing a solid data foundation for subsequent pollution event analysis and decision-making. Accurately identifying pollution discharge event types by calculating pollution source linkage indices and finely segmenting the temporal stages of pollution events based on event types and water quality monitoring data helps to quickly and accurately identify the occurrence, development, and changes of pollution events. This provides crucial support for timely and targeted pollution prevention and control measures, minimizing the harm of pollution events to the aquatic environment. Timely adjustment of sampling strategies based on the real-time situation and temporal stage of pollution events provides environmental management departments with timely, accurate, and dynamic decision-making basis for formulating scientific and reasonable aquatic protection policies, pollution control plans, and emergency response measures, thereby improving the scientific nature and effectiveness of aquatic environment management.

[0034] In one embodiment of this invention, water sampling points for the target monitoring water area are set based on hydrological data, pollutant discharge data, and main transport path data, including: S210. Obtain wind field data and surface flow field data through hydrological data of the target monitoring water area; S220. By monitoring the pollutant emission data of the target enterprises, the location of the pollution source, emission time, emission flow rate, and emission concentration can be obtained. S230: Convert emission flow rate and emission concentration into pollutant release flux, and convert wind field data into wind stress according to the preset wind stress conversion rate; S240, Set the simulation time window in conjunction with the emission time; S250. Taking the location of the pollution source as the starting point of diffusion, a migration and diffusion model is constructed within the simulation time window by combining data on pollutant release flux, wind stress, and surface flow field. S260. Based on the migration and diffusion model, extract the pollutant isochrones and combine the pollutant isochrones with the main transport path data to extract the target intersection point; S270. Calculate the surface diffusion boundary and concentration gradient distribution of pollutants based on pollutant release flux, surface flow field data, and wind stress. S280. Set water surface sampling points by combining target intersections, surface diffusion boundaries, and concentration gradient distribution; S290. Based on surface sampling points, hydrological data, and pollutant discharge data, set vertical sampling points for the target monitoring water area.

[0035] Figure 2 This application provides a schematic diagram of the structure of a water sampling point for setting a target monitoring water area, as shown in the embodiment of the present application. Figure 2 As shown, this includes surface sampling points and vertical sampling points in the water area.

[0036] First, by monitoring hydrological data and surface flow field data of the target monitoring water area, wind field data and surface flow field data of that water area can be obtained. Wind field data can include various data such as water level, flow velocity, flow rate, water temperature, and water quality. Surface flow field data can include relevant information about the surface water flow of the water area, such as the speed and direction of the surface water flow.

[0037] Secondly, monitoring the pollutant emission data of the target enterprise can reveal the location of the pollution source, the specific time of emission, the flow rate, and the concentration of pollutants. Monitoring the pollutant emission data of the target enterprise can be achieved by detecting and recording the emissions of pollutants such as waste gas and wastewater. The location of the pollution source can be specified, such as which workshop or discharge pipe in the factory is emitting pollutants; the specific time of emission can include when the emission began and for how long; the flow rate can be the amount of pollutants emitted per unit time, such as how many cubic meters of wastewater are emitted per hour; and the concentration of pollutants can be the concentration of a certain harmful liquid, such as milligrams per cubic meter.

[0038] Subsequently, the pollutant release flux is calculated by multiplying the emission flow rate and emission concentration. Pollutant release flux represents the amount of pollutants passing through a unit area per unit time. By converting it to pollutant release flux, the actual rate and intensity of pollutants entering the environment can be more intuitively measured. For example, if the emission flow rate is 100 cubic meters per hour and the emission concentration is 2 milligrams per cubic meter, then the pollutant release flux is 100 × 2 = 200 milligrams per cubic meter per hour. Then, wind speed and direction distribution characteristics are obtained from wind field data. In hydrodynamics, the effect of wind on the water surface is usually characterized by wind stress. Therefore, by using a preset wind stress conversion rate formula, wind speed data is converted into wind stress data to quantitatively characterize the driving effect of wind on surface water. The preset wind stress conversion rate is a conversion relationship determined based on a certain physical model (such as aerodynamic theory) or empirical formula. The wind stress calculation formula is:

[0039] in, Indicates wind stress; This represents the drag coefficient, which can be considered as part of the preset wind stress conversion rate and can be set by factors such as wind speed and atmospheric stability. It represents air density; U represents wind speed, derived from wind field data.

[0040] By substituting information such as wind speed from the wind field data into the formula above, wind stress can be obtained. In this embodiment, pollutant release flux is used to characterize the total amount of pollutants released from the pollution source into the target water area per unit time, reflecting the intensity change of the pollution process; pollutant emission type is used to describe the category of pollutant components released by the pollution source during the emission process, which is a static parameter in the composition dimension, including inorganic nitrogen, phosphate, heavy metals, organic pollutants, etc.; and the pollution source fingerprint is a set of feature vectors constructed based on the above pollutant emission types and their release flux distribution characteristics, used to uniquely characterize the emission characteristics of a specific pollution source.

[0041] In one embodiment, a simulation time window for the target monitoring water area is set based on the emission time of the pollution source. This simulation time window covers the start time, duration, and end time of pollutant emission, ensuring that subsequent simulations of pollutant transport, diffusion, and concentration changes in the water area accurately reflect the entire emission event. Here, emission time refers to the specific time period between the start and end of pollution source emission. The simulation time window is the time range that needs to be clearly defined in water quality or hydrodynamic simulations, i.e., the simulation start and end times.

[0042] Next, taking the pollution source location as the diffusion starting point, a migration and diffusion model is constructed within the simulation time window, incorporating pollutant release flux, wind stress, and surface flow field data. Specifically, the initial release point where pollutants begin to be released is first determined and used as the diffusion starting point. Using the previously set simulation time window, and combining pollutant release flux from the pollution source, wind stress data of the target monitoring water area, and surface flow field data, a migration and diffusion model of pollutants in the target monitoring water area is constructed to simulate the spatial distribution and concentration evolution of pollutants in the water. The migration and diffusion model can be constructed based on a convolutional neural network framework. Historical water quality monitoring data and hydrological and meteorological data are used to train the neural network model to learn the diffusion patterns of pollutants under different hydrodynamic conditions. Subsequently, pollutant release flux, wind stress, and surface flow field data from the pollution source are used as input features, and the concentration distribution of pollutants at different locations in the water area is used as the output target. The model parameters are optimized through iterative training, enabling the model to accurately predict the spatial migration and concentration changes of pollutants in the water area. During the model prediction phase, inputting new pollution source release fluxes and wind and flow field data will yield corresponding pollutant migration and diffusion results, which can be used for subsequent sampling point deployment or pollution emergency assessment.

[0043] Based on the aforementioned pollutant migration and diffusion model for the target monitoring water area, isochrones of pollutants are extracted to characterize the time required for pollutants to diffuse from the pollution source to various spatial locations within the water area. First, the simulation results of the aforementioned pollutant migration and diffusion model for the target monitoring water area are obtained, including data on the concentration changes of pollutants at various spatial locations within the water area over time. Based on the simulation results, the time required for pollutants to reach various spatial locations within the water area from the pollution source is determined. Lines are drawn connecting spatial locations that arrive at the same time to generate an isochrone distribution map of pollutant diffusion. After obtaining the pollutant isochrones, the target intersection point between the main transport path data and the pollutant isochrones is determined. Since the main transport path of ships is the most likely path for pollutant diffusion, the key area for pollutant diffusion is determined by identifying the target intersection point. The main transport path data can describe the main migration direction or flow channel of pollutants in the water area and can be derived from surface flow field data or historical water quality data. By overlaying the pollutant isochrones with the main transport path, we can find the intersections between the isochrones and the main transport path. These intersections are usually the key locations where pollutants are likely to arrive most quickly and in the most concentrated manner, indicating the specific geographical location and time point along this main transmission path where the pollutants will arrive.

[0044] Based on pollutant release flux, surface flow field data, and wind stress data from pollution sources in the target monitored water area, the diffusion boundary and concentration gradient distribution of pollutants on the water surface are calculated to characterize the spatial diffusion range and concentration variation patterns of pollutants on the water surface, providing a basis for subsequent water quality monitoring sampling point deployment or pollution emergency response. First, based on pollutant release flux data, the total amount of pollutants entering the water body per unit time is determined. Combined with surface flow field data and wind stress data, a mathematical model based on fluid mechanics and mass diffusion theory can be used to simulate the migration and diffusion process of pollutants on the water surface. The diffusion boundary of pollutants on the water surface is obtained from the simulation results, i.e., the maximum spatial range that pollutants may reach, thus obtaining the surface diffusion boundary of pollutants, which refers to the boundary of the range reached by pollutants in the environment. The concentration data of pollutants on the water surface can also be determined based on the simulation results of the migration and diffusion model, i.e., the change of pollutant concentration over time at each spatial grid or sampling point. Specifically, the water surface can be divided into a two-dimensional grid. For each grid cell, the concentration difference between neighboring cells is calculated and divided by the spatial distance to obtain the local concentration gradient vector. This vector includes the direction of concentration change (gradient direction) and the magnitude of change (gradient magnitude), thus yielding the concentration gradient distribution. The concentration gradient distribution of pollutants refers to the spatial variation of pollutant concentration, i.e., the difference in pollutant concentration at different locations. The above steps provide the surface diffusion boundary and concentration gradient distribution of pollutants.

[0045] After obtaining the target intersection points, surface diffusion boundaries, and concentration gradient distribution, sampling points on the water surface are established based on these factors. These sampling points refer to locations on the surface of the water body specifically designated for collecting water samples for analysis. The layout of these sampling points covers both the key target intersection points for pollutant diffusion and the surface diffusion boundaries, prioritizing areas with large or significant pollutant concentration gradients to ensure that water quality monitoring scientifically and comprehensively reflects the distribution characteristics of pollutants on the water surface. In practice, the water surface is first divided into predefined areas. Then, grid cells covering the target intersection points and surface diffusion boundaries are selected as sampling units. The sampling point density is increased in areas with large or significant pollutant concentration gradients, thus generating a scientific, comprehensive, and focused water surface sampling point layout scheme to ensure that the collected water quality data accurately reflects the distribution characteristics and diffusion patterns of pollutants on the water surface.

[0046] Finally, based on surface sampling points, hydrological data, and pollutant discharge data, vertical sampling points for the target monitoring water area were established. Specifically, temperature and density strata in the water body were identified using water temperature and density profile data to determine the water's stratification characteristics. Subsequently, a vertical profile was constructed for each surface sampling point, and combined with the water stratification characteristics, a set of vertical candidate sampling points corresponding to each surface sampling point was determined. Further, the candidate sampling point set was filtered based on the maximum information gain criterion to determine the priority sampling layers. Then, the vertical distribution characteristics of pollutants were determined by combining pollutant discharge data and combined with the priority sampling layers to finally determine the vertical sampling points for the target monitoring water area. These vertical sampling points are locations set in the vertical direction of the target monitoring water area for collecting water quality information, and are arranged vertically along the direction of each surface sampling point, covering different depth layers of the water body. The layout of vertical sampling points in a water body is based on the vertical stratification characteristics of the water body, such as temperature strata and density strata, the possible vertical distribution of pollutants, and hydrological conditions. This aims to scientifically and comprehensively reflect the concentration distribution of pollutants and the patterns of water quality changes along the vertical axis of the water body. Using this method, each surface sampling point corresponds to multiple key vertical layers, thus forming a vertical sampling point layout that covers different layers of the water body and conforms to the distribution patterns of pollutants, ensuring a scientific and comprehensive reflection of the water quality distribution characteristics of the water body.

[0047] By setting target sampling points for monitoring water areas, the layout of these points can be more scientific and rational, better reflecting the actual migration and diffusion of pollutants in the water, thus significantly improving the accuracy of water pollutant monitoring. Furthermore, it allows for faster and more targeted identification of key sampling locations, reducing unnecessary sampling point setups and avoiding resource waste.

[0048] In one embodiment of this invention, vertical sampling points for the target monitoring water area are set based on surface sampling points, hydrological data, and pollutant discharge data, including: S310. Obtain water temperature profile data and water density profile data through hydrological data of the target monitoring water area; S320. Identify the water temperature gradient and water density gradient of the target monitoring water area based on water temperature profile data and water density profile data. S330. Combining the water temperature jump layer and the water density jump layer, the target monitoring water area is divided into an upper uniform layer, a gradient abrupt layer and a lower uniform layer. S340. For any water surface sampling point, construct a surface vertical profile, and combine the surface vertical profile, upper uniform layer, gradient abrupt layer and lower uniform layer to determine the set of vertical candidate sampling points corresponding to each water surface sampling point. S350. Based on the maximum information gain criterion, each set of vertical candidate sampling points is screened to determine the priority vertical sampling layer. S360. Determine the types of pollutants emitted through pollutant emission data, and determine the vertical distribution characteristics corresponding to the types of pollutants emitted in a preset database; S370. Based on the vertical distribution characteristics and the priority vertical sampling layers, set the vertical sampling points for the target monitoring water area.

[0049] By acquiring hydrological data of the target monitoring water area, including water depth, water temperature, flow velocity, flow direction, and related meteorological conditions, the temperature and density distribution of the water area in the vertical direction is analyzed to obtain water temperature profile data and water density profile data. The water temperature profile data represents the temperature distribution of the water area from the surface to the bottom at different depths, and the water density profile data represents the density distribution of the water area at different depths. The temperature profile data and density profile data can be used to identify temperature and density strata in the water body, thereby providing a basis for subsequent water stratification and sampling point layout.

[0050] Next, based on the water temperature profile and density profile data of the target monitoring area, the vertical temperature and density changes of the water body are analyzed to identify temperature and density gradients. A temperature profile is a measurement of the change in water temperature with depth from the surface to the bottom; a density profile is a measurement of the change in water density with depth from the surface to the bottom. First, the temperature and density gradients, i.e., the rates of change of temperature and density with depth, are calculated for the temperature and density profiles in the vertical direction. The gradient method can be used to calculate these rates. Then, the gradients are compared with preset thresholds. Depth segments where the temperature or density gradient exceeds the preset threshold are identified as temperature or density gradients. These preset thresholds can be set based on the required monitoring accuracy. A temperature gradient is a vertical layer in the water body where the water temperature changes abruptly with depth; a density gradient is a vertical layer in the water body where the water density increases abruptly with depth, representing the region with the largest vertical density gradient. Furthermore, by segmenting and fitting temperature and density profiles, we can identify layers with significant abrupt changes in slope as thermoclines. The temperature thermocline is a layer in the water body where the temperature changes significantly with depth, and the density thermocline is a layer in the water body where the density changes significantly with depth. By identifying the temperature thermocline and the density thermocline, we can determine the vertical stratification structure of the water body, providing a basis for subsequent vertical sampling point layout and pollutant vertical distribution analysis.

[0051] Based on the temperature and density strata of the target monitoring area, the water body is vertically divided into an upper homogeneous layer, a gradient abrupt change layer, and a lower homogeneous layer. The upper homogeneous layer is located from the water surface to the upper edge of the temperature or density strata, characterized by slow changes in temperature and density and homogeneous water mixing. The gradient abrupt change layer is located in the region of the temperature or density strata, characterized by abrupt changes in temperature or density with depth, and is the most significant intermediate layer in terms of water stratification. The lower homogeneous layer is located below the gradient abrupt change layer to the bottom, characterized by slow changes in temperature and density and relatively stable water. By calculating the temperature gradient from the temperature profile and comparing it with a preset temperature gradient threshold, layers where the temperature changes significantly with depth are identified as temperature jump layers. The preset temperature gradient threshold can be set based on historical observation data. Simultaneously, by calculating the density gradient from the density profile and comparing it with a preset density gradient threshold, layers where the density changes significantly with depth are identified as density jump layers. The preset density gradient threshold can also be set based on historical observation data. Then, the starting and ending depths of the temperature and density jump layers are integrated to determine the comprehensive jump layer interval. Next, using the comprehensive jump layer interval as the boundary, the water body is divided from top to bottom into an upper homogeneous layer, a gradient abrupt change layer, and a lower homogeneous layer. The upper homogeneous layer is located from the water surface to the upper edge of the jump layer, characterized by slow temperature and density changes and homogeneous water mixing. The gradient abrupt change layer is located within the jump layer interval, characterized by rapid temperature or density changes with depth. The lower homogeneous layer is located from the lower edge of the jump layer to the bottom, characterized by slow temperature and density changes and relatively stable water.

[0052] For any surface sampling point in the target monitoring area, a vertical profile is constructed from the water surface downwards. This profile represents the depth variation range of the sampling point in the water body. Combining the division into an upper homogeneous layer, a gradient abrupt change layer, and a lower homogeneous layer, the depth range of each layer on the vertical profile is analyzed to determine the key depths within each layer as vertical candidate sampling points. These vertical candidate sampling points can include the middle or bottom of the upper homogeneous layer, the middle of the gradient abrupt change layer, and the middle or top of the lower homogeneous layer. Because the upper homogeneous layer exhibits slow temperature and density changes and is homogeneously mixed, selecting the middle or bottom section can represent the water quality characteristics of the entire upper homogeneous layer, ensuring that the sampling point accurately reflects the average physicochemical properties of the upper homogeneous layer, while avoiding sampling only extreme points on the surface that are excessively affected by wind, waves, or sunlight. The gradient abrupt change layer is a water layer with rapidly changing temperature or density, representing the most significant water stratification. The middle section can capture the typical characteristics of gradient changes in this layer, which is beneficial for analyzing the vertical migration and distribution patterns of pollutants in the abrupt change layer region. The lower homogeneous layer exhibits slow temperature and density changes, and the bottom water is typically stable. Selecting the middle or top section can represent the water quality of this layer, ensuring sampling coverage of the bottom water. This reflects both the average characteristics of the lower homogeneous layer and captures potential pollutant accumulation at the bottom. Using the above method, a set of vertical candidate sampling points is formed corresponding to each surface sampling point in the water body. This provides a basis for subsequent selection of priority vertical sampling layers based on information gain or the vertical distribution characteristics of pollutants, thus ensuring that sampling points can cover different layers of the water body vertically and scientifically reflect its vertical characteristics.

[0053] Next, the vertical candidate sampling point set is analyzed based on the maximum information gain criterion. The contribution of each candidate sampling point to the vertical structure and pollutant distribution information of the water body is calculated. Information gain can be quantified based on temperature gradients, density gradients, pollutant concentration gradients, or historical observation data. Information gain measures the contribution of a feature to the information of a sample set. For example, in water quality monitoring, suppose we want to understand the vertical distribution of a certain pollutant in a lake. If, at a certain depth, the concentration of this pollutant differs significantly from other depths, and this difference can significantly distinguish different water quality conditions, then the information gain of that depth is large. Specific calculations may involve variance analysis and entropy calculations for samples at different depths. For example, the original entropy is calculated, and then the conditional entropy for judging the lake's water quality after sampling at a certain depth is calculated; the difference between the two is the information gain of that depth. In the candidate sampling point set, selecting the point with the largest information gain as the priority sampling layer ensures that each surface sampling point, when sampling along the vertical direction, covers the layers where water information changes most significantly. By comparing the information gain values ​​of each candidate point, the sampling point with the highest information gain is selected as the priority vertical sampling layer, thereby ensuring that the selected layer can effectively reflect the vertical structure of the water body and the distribution characteristics of pollutants, and providing a scientific basis for the subsequent vertical sampling point layout.

[0054] Based on pollutant discharge data from the target monitoring water area, the specific types of pollutants emitted by the pollution sources are identified. These pollutants include, but are not limited to, chemical oxygen demand (COD), ammonia nitrogen, heavy metals, and volatile organic compounds (VOCs). Subsequently, the vertical distribution characteristics of each pollutant type are queried from a pre-set database. This database records the vertical distribution patterns and potential accumulation layers of different pollutants in the water body. For example, easily soluble pollutants or those uniformly distributed with the water body tend to accumulate in the upper uniform layer or throughout the entire water body; pollutants with higher density or easier sedimentation tend to accumulate in the lower uniform layer; and pollutants affected by temperature or density gradients may concentrate in gradient abrupt change layers. Determining the vertical distribution characteristics of pollutants provides a basis for subsequent vertical sampling point layout and vertical migration analysis of pollutants.

[0055] Finally, the vertical sampling points of the target monitoring water area are set by combining the vertical distribution characteristics and the priority vertical sampling layers. Specifically, firstly, the preset priority vertical sampling layers are compared and comprehensively analyzed with the vertical distribution characteristics of pollutants in the monitoring water area to determine the key layers of pollutant concentration changes in different depth ranges. The preset priority vertical sampling layers are set according to the pollutant distribution characteristics and are realized by identifying the layers where pollutants accumulate at high concentrations based on historical monitoring data or preliminary detection results. Secondly, based on the comparison results, water layers that can both cover the key layers with the greatest information gain and match the key distribution depth of pollutants are selected as vertical sampling points. Furthermore, using the surface sampling points as a reference, sampling points are extended vertically (in the depth direction) from these points. Based on the analysis results of the priority layers and the vertical distribution characteristics of pollutants, specific depth locations are selected as actual sampling points. These may be one point or multiple points, ensuring that the sampling points cover the key water layer structure, including the upper uniform layer, the gradient abrupt layer, and the lower uniform layer. Finally, the vertical sampling points for the target monitoring water area are obtained, ensuring that the deployed vertical sampling points can comprehensively reflect the vertical structural characteristics of the water body and accurately capture the distribution characteristics of different pollutants in the vertical direction, thereby achieving a scientific assessment and representative sampling of the water quality of the monitoring water area.

[0056] By setting vertical sampling points in the target monitoring area, the sampling layout can better align with the water stratification and the actual vertical distribution patterns of pollutants, making the monitoring work more scientific and targeted. Furthermore, it can accurately locate areas of high pollutant concentration or key changing water layers, obtaining more representative samples, thereby improving the accuracy of monitoring data and better reflecting the actual pollution situation.

[0057] In one embodiment of this invention, the sampling point data includes first time point data and second time point data. The validity of the sampling point data is verified to obtain a validity verification result. Based on the validity verification result, the validity of the water surface sampling points and the water vertical sampling points is determined, including: S410. For any water area sampling point, collect data at the first preset time point and data at the second preset time point respectively; S420. Calculate the relative deviation value based on the data at the first time point and the data at the second time point, and determine the first validity based on the relative deviation value; S430. For any water sampling point, obtain the original concentration of the water sample to be tested; S440. Add a standard solution of the target pollutant of known concentration to the water sample to be tested to obtain the spiking amount of the target pollutant. S450. Analyze the water sample to be tested by adding a standard solution of the target pollutant of known concentration to obtain the concentration of the target pollutant after spiking; S460. Calculate the second validity based on the spiked concentration of the target pollutant, the spiking amount of the target pollutant, and the original concentration. S470. Calculate the overall validity based on the first validity and the second validity. S480. If the overall validity is greater than or equal to the preset validity threshold, then the corresponding water surface sampling point and water vertical sampling point are determined to be valid.

[0058] First, for any given water sampling point, sampling is performed at a first preset time point and a second preset time point to obtain corresponding data at the first and second time points. In this embodiment, the first and second preset time points are two different time points. These two sets of data are used for subsequent validity verification to evaluate the stability of the sampling point over time, thereby ensuring the reliability of subsequent water quality monitoring results.

[0059] Based on data from arbitrary water sampling points acquired at first and second preset time points, the relative deviation between the two sets of data is calculated to characterize the data stability of the sampling points over time. The relative deviation can be calculated using the formula shown below:

[0060] in This represents the relative deviation value; This represents data at the first point in time. This indicates data at the second time point.

[0061] Furthermore, the first validity of the sampling point is determined by the relative deviation value. The first validity is used to evaluate the consistency of the data collected at different time points and to provide a basis for subsequent comprehensive validity determination.

[0062] Subsequently, for any water sampling point, the water sample to be tested is collected at the sampling point, and the original concentration of the target pollutant in the water sample is determined to obtain basic water quality data without spiked treatment. The original concentration is used for subsequent spiked experiments and effectiveness verification to evaluate the accuracy of the sampling point test results.

[0063] A pre-prepared or purchased standard pollutant solution with a clearly defined concentration is added to the original water sample collected from the water sampling point to obtain the target pollutant spiking amount. The target pollutant spiking amount is used for subsequent spiking recovery experiments. By comparing the original concentration before spiking and the concentration measured after spiking, the accuracy and reliability of the detection results from the water sampling point are evaluated.

[0064] To analyze water samples to be tested with a standard solution of a target pollutant at a known concentration, atomic absorption spectrometry can be used to obtain the spiked concentration of the target pollutant in the water sample. The spiked concentration is then compared with the original concentration and the known amount of spike, thereby evaluating the accuracy of the detection method at the water sampling point.

[0065] After obtaining the spiked concentration of the target pollutant, the spiking amount, and the original concentration, the second validity is calculated using a preset formula, as shown below:

[0066] The second validity is calculated using the formula above.

[0067] Next, based on the first and second validity of the sampling points, the overall validity is calculated using the weighted average method. The overall validity is used to comprehensively characterize the reliability of the sampling point data in terms of time stability and spike recovery accuracy. Each validity indicator can be assigned a weight, and then their weighted average is calculated. For example, overall validity = W1 × first validity + W2 × second validity, where W1 and W2 are weights, and W1 + W2 = 1. W1 and W2 can be set according to the importance of each indicator in the overall evaluation, and then the overall validity is calculated using the above formula.

[0068] Finally, when the overall validity of the water sampling points is greater than or equal to the preset validity threshold, the corresponding surface sampling points and vertical sampling points are determined to be valid sampling points. The preset validity threshold can be set based on statistical analysis of historical monitoring data or multiple sampling results. This method ensures that only sampling points meeting the requirements for time stability and spike recovery accuracy are used for subsequent water quality monitoring, thereby improving the reliability and accuracy of the obtained water quality monitoring data.

[0069] By determining the validity of surface sampling points and vertical sampling points in the water area, deviations in monitoring results due to unstable sampling data or detection errors can be avoided, thus ensuring the authenticity and reliability of the acquired water quality data and improving data reliability. At the same time, excluding invalid sampling points saves sampling and analysis costs, improves monitoring efficiency, avoids wasting human, material, and experimental resources, and optimizes resource utilization.

[0070] In one embodiment of this invention, the pollution event is classified and its temporal stages are defined based on the type of pollution discharge event and water quality monitoring data, including: S510. Obtain pollutant concentrations, concentration change rates, and abnormal pollutant concentration values ​​through water quality monitoring data; S520. Select the corresponding type parameter group from the preset event type library according to the type of pollution emission event. The type parameter group includes rising hysteresis threshold, falling hysteresis threshold, forgetting factor and minimum residence time. The forgetting factor is used to characterize the memory length of event energy for historical data and the weight decay rate. S530. Calculate the type linkage index based on the type parameter group using a preset multidimensional fusion model. The type linkage index is used to characterize the degree of matching between the pollution emission event type and the target pollution type. S540. A comprehensive anomaly score is obtained by weighted fusion calculation of pollutant concentration, concentration change rate and pollutant concentration anomaly value. S550. Calculate the event energy using the forgetting factor and the comprehensive abnormality score using the preset event energy formula; S560. Combine the rising hysteresis threshold and the falling hysteresis threshold to determine the time trigger time and the time end time; S570. Within the time trigger moment and time end moment, the event energy and type linkage index are combined with the minimum residence time to divide and determine the transient segment, transition segment and steady-state segment of the pollution emission event type, wherein the transient segment, transition segment and steady-state segment are used to characterize the time sequence stage definition.

[0071] Water quality monitoring data obtained from water sampling points can yield pollutant concentrations, concentration change rates, and pollutant concentration anomalies. Pollutant concentration characterizes the absolute amount of target pollutants in the water body; the concentration change rate characterizes the rate of change of pollutant concentration over time, thus reflecting the dynamic characteristics of a pollution event; pollutant concentration anomalies are used to identify significant deviations from historical averages or preset thresholds, indicating the occurrence of a pollution event or abnormal discharge. Pollutant concentration, concentration change rate, and anomalies can serve as key input parameters for classifying pollution events and defining their timing stages.

[0072] Based on the type of pollution emission event, the corresponding type parameter group is selected from a preset event type library. The type parameter group includes an upward hysteresis threshold, a downward hysteresis threshold, a forgetting factor, and a minimum residence time. The upward hysteresis threshold is the trigger value used to determine whether an event has started or whether a certain state has changed from "low" to "high." The downward hysteresis threshold is the release value used to determine whether an event has ended or whether a certain state has changed from "high" to "low." The forgetting factor characterizes the length of time an event's energy is remembered for historical data and the rate of weight decay. In other words, when calculating a certain indicator or state at the current moment, different weights are assigned to historical data; more recent data has a higher weight, and older data has a lower weight, and this weight decay is exponential. The preset pollution emission event types are obtained through the analysis of historical pollution emission events, experimental research, and environmental management needs. Different emission patterns and their characteristic parameters can be organized into an event type library, including typical concentration change curves, durations, fluctuation amplitudes, and other parameters. This event type library can be used for subsequent pollution event identification, linkage index calculation, and time series stage division. By selecting type parameter groups, a quantitative basis can be provided for calculating the event energy and dividing the time sequence of pollution events.

[0073] Based on a set of type parameters, a pre-defined multidimensional fusion model is used to analyze water quality monitoring data to calculate a type linkage index. The type linkage index characterizes the degree of matching between pollution discharge event types and target pollution types. The pre-defined multidimensional fusion model can be a pre-trained neural network model. A pre-trained neural network model is a complex computational framework or algorithm that has been designed, trained, and validated before practical application, with all parameters configured. It can integrate multidimensional features such as pollutant concentration, concentration change rate, and concentration outliers for comprehensive analysis and output a unified evaluation result. Detailed monitoring data of past actual pollution events and their corresponding real pollution types are collected, allowing the neural network model to learn how to map input features to the type linkage index from the prepared data. The extracted multidimensional data is used as the model input, and the corresponding real pollution type or matching degree is used as the model output label. Through optimization algorithms (such as gradient descent), the model's internal parameters (such as the weights and biases of the neural network) are adjusted to make the model's predicted type linkage index as close as possible to the actual situation, quantifying the degree of event type matching and providing a basis for the temporal stage division of pollution events.

[0074] Next, by combining pollutant concentration, concentration change rate, and pollutant concentration anomalies, each characteristic value is weighted and fused according to preset weights to obtain a comprehensive anomaly score. The comprehensive anomaly score is used to characterize the overall degree of anomaly of the current pollution event in the target water area. The weighted fusion calculation can be set according to the importance of each characteristic to the judgment of the pollution event, comprehensively reflecting the pollutant concentration level, concentration change dynamics, and anomaly deviations, providing a quantitative basis for the calculation of pollution event energy and the definition of timing stages.

[0075] The forgetting factor and the comprehensive anomaly score are input into a preset event energy formula to calculate the event energy. Event energy quantifies the intensity or severity of a pollution event in the target water area. The forgetting factor characterizes the rate of decay of historical data's influence on the event energy, while the comprehensive anomaly score reflects the current level of water quality anomaly. Specifically, the comprehensive anomaly score at the current moment is normalized to obtain a normalized dimensionless index. Based on the normalized comprehensive anomaly score at the current moment, the preset event energy formula is as follows:

[0076] Where Et is the event energy at the current moment; S t E represents the overall anomaly score at the current moment. t-1 The energy α of the previous event is used as a forgetting factor to attenuate the influence of historical data. The forgetting factor α can be determined based on the decay rate of the historical impact of the pollution event. For pollution events with a relatively fast historical impact (such as sudden pollution), a larger α value should be selected (such as a value close to 1, the specific value can be set according to actual needs); for pollution events with a long duration of historical impact (such as chronic pollution), a smaller α value should be selected (such as a value close to 0.1, the specific value can be set according to actual needs).

[0077] By calculating the energy of an event, we can comprehensively consider the immediacy of water quality anomalies and the impact of historical data, providing a quantitative basis for the temporal stage division and trigger determination of pollution events.

[0078] By combining the rising hysteresis threshold and the falling hysteresis threshold, the trigger and end times of a pollution event can be determined. Specifically, when the rate of change of pollutant concentration in the monitored water body reaches or exceeds the rising hysteresis threshold, the event is determined to have entered the rising phase, and the event trigger time is recorded. When the rate of change of pollutant concentration drops to or falls below the falling hysteresis threshold, the event is determined to have entered the falling phase, and the event end time is recorded. Thus, the event trigger and end times are determined. The event trigger time refers to the point in time when the pollution event is determined to have started. In water quality monitoring, when the pollutant concentration or rate of change reaches the preset rising hysteresis threshold, the event is considered to have been officially triggered. Essentially, this marks the start time of the pollution event transitioning from a normal state to an abnormal state. The event end time refers to the point in time when the pollution event is determined to have ended. In water quality monitoring, when the pollutant concentration or rate of change drops below the preset falling hysteresis threshold, the event is considered to have ended. Essentially, this marks the end of the abnormal state of the pollution event, and the water quality has returned to or is close to normal. By setting and monitoring thresholds, the time range of pollution events can be effectively defined, avoiding misjudgments due to short-term concentration fluctuations, and providing a basis for dividing pollution events into transient, transitional, and steady-state phases.

[0079] Between the time trigger moment and the time end moment, combined with a preset minimum dwell time, the event energy and type linkage index are analyzed and divided to determine the transient, transitional, and steady-state segments of the pollution emission event type. The transient segment represents the stage where the event energy rises rapidly immediately after triggering; the transitional segment represents the stage where the energy increase slows down; and the steady-state segment represents the stage where the event energy reaches a relatively stable and sustained level. Specifically, starting from the time trigger moment, a continuously rising interval is identified along the event energy curve and this interval is determined as the transient segment. The transient segment represents the stage where the event energy rises rapidly immediately after triggering, and the duration of this interval must be greater than or equal to the minimum dwell time to avoid short-term fluctuations. Subsequently, when the event energy curve reaches a relatively stable level and the fluctuation amplitude is below a preset threshold for a continuous period of time, and the duration meets the minimum dwell time, this interval is divided into the steady-state segment. The steady-state segment represents the stage where the event energy tends to stabilize and persists. Within the time interval between the transient and steady-state segments, the period where the energy increase gradually slows down but has not yet reached a steady-state level is divided into the transitional segment. During the classification process, the type linkage index is used to assist in verifying the degree of matching between each time period and the preset pollution emission event type, ensuring the scientific and reasonable nature of the division into transient, transitional, and steady-state segments. Through this classification, a detailed characterization of pollution emission events in the time dimension can be achieved, providing a quantitative basis for defining the temporal stages of pollution events, event response, and risk assessment.

[0080] By defining the temporal stages of pollution events, we can precisely divide them into transient, transitional, and steady-state phases. This helps us to more accurately grasp the dynamic evolution of pollution events, clarify the pollution characteristics and patterns at different stages, and provide more precise timing for subsequent pollution control and prevention. Furthermore, by combining factors such as comprehensive anomaly scores and event energy, the identification of pollution events becomes more scientific and accurate, effectively avoiding misjudgments and omissions, and ensuring the timely and accurate detection of the occurrence and development of various pollution events.

[0081] In one embodiment of this invention, the pollution emission event types include volatile events and heavy particulate events, and the method further includes: S610. When the pollution emission event type is a volatile event, obtain surface flow velocity data from water surface sampling points; S620. Calculate the average surface velocity based on the surface velocity data, and calculate the sum of squares of velocity deviations based on the average surface velocity. S630, calculate the surface disturbance index by combining the sum of squared velocity deviations and the number of sampling points on the water surface; S640. The rising hysteresis threshold is adjusted by combining the surface disturbance index to obtain the adjusted rising hysteresis threshold. S650. When the pollution emission event type is a heavy particulate event, obtain the heavy particulate concentrations of the upper homogeneous layer, the gradient abrupt layer, and the lower homogeneous layer, respectively. S660. Calculate the absolute value of the heavy particle concentration between any two adjacent layers by combining the heavy particle concentrations of the upper homogeneous layer, the gradient abrupt layer, and the lower homogeneous layer, and add each absolute value together to obtain the stratification intensity index. S670. The descent hysteresis threshold is adjusted by combining the stratification intensity index to obtain the adjusted descent hysteresis threshold.

[0082] When a pollution discharge event is classified as a volatile event, surface velocity data from sampling points on the water surface can be obtained. A volatile event refers to a pollution discharge event occurring in a water body, characterized by the easy volatilization or rapid diffusion of pollutants from the water surface into the air. Specifically, for each water surface sampling point, a velocity measurement device installed on the water surface collects surface velocity information to reflect the flow speed and direction of the water at that sampling point. Surface velocity data can be used to analyze and simulate the migration and diffusion characteristics of volatile pollutants on the water surface, thus providing a basis for pollutant distribution prediction, pollution event response, and related water quality monitoring.

[0083] In one embodiment of this invention, the average surface velocity can be calculated based on surface velocity data from sampling points on the water surface. Specifically, for each water surface sampling point, the average velocity value during the sampling period is obtained by averaging multiple sets of collected surface velocity data. This average velocity value reflects the overall flow level of the water surface. Subsequently, based on the average surface velocity, the deviations of each set of velocity data from the mean are squared and summed to obtain the sum of squared velocity deviations. This sum of squared deviations characterizes the fluctuation range or variation of the surface velocity in the water body. This calculation provides a quantitative basis for subsequent simulation of the migration and diffusion of volatile pollutants, optimized sampling point layout, and water quality monitoring.

[0084] The surface disturbance index is calculated by combining the sum of squared velocity deviations at various water surface sampling points and the number of sampling points. Specifically, for each water surface sampling point, the surface disturbance index is obtained by dividing its sum of squared velocity deviations by the number of sampling points. The surface disturbance index quantifies the intensity or instability of surface flow in water. It reflects the uniformity of flow and the magnitude of disturbance at the water surface and can be used to assess the migration and diffusion characteristics of volatile pollutants at the water surface, thus providing a scientific basis for pollutant distribution simulation, optimized sampling point placement, and water quality monitoring.

[0085] Next, the rise hysteresis threshold in the pollution event classification is adjusted based on the surface disturbance index to obtain the adjusted rise hysteresis threshold. Specifically, the original rise hysteresis threshold of the water surface sampling points is obtained, and an adjustment coefficient function for threshold adjustment is determined based on the surface disturbance index. The surface disturbance index is then substituted into this function to calculate the corresponding adjustment coefficient. The adjustment coefficient function is shown below:

[0086] Where k is the adjustment coefficient, which can be determined by the surface disturbance index. For example, the larger the disturbance index, the larger the k value can be selected to increase the adjustment range; the smaller the disturbance index, the smaller the k value, and the smaller the adjustment range. The adjusted rise hysteresis threshold; D represents the surface disturbance index of the water area.

[0087] When the surface disturbance intensity of the water body is large, the surface disturbance index When the value is high, the system automatically lowers the rise hysteresis threshold, allowing disturbances with relatively low event energy to enter the transient phase, thus enabling timely capture of changes in surface diffuse pollution; when the surface disturbance intensity is low, the surface disturbance index... If the value of the threshold is low, the system will correspondingly increase the rise hysteresis threshold to avoid misidentifying background noise or non-polluting natural fluctuations as pollution events. Through this method, the rise hysteresis threshold can be dynamically and adaptively adjusted according to the actual environment, effectively balancing the sensitivity and accuracy of pollution event identification, and improving its applicability and robustness under complex hydrological conditions.

[0088] Subsequently, the original rise hysteresis threshold is multiplied by an adjustment coefficient to obtain the adjusted rise hysteresis threshold. Since surface disturbances in water bodies cause short-term fluctuations in pollutant concentrations, directly using a fixed threshold may lead to misjudgments or missed detections. Adjusting the rise hysteresis threshold by using a disturbance index allows the pollution event classification method to adapt to the dynamic characteristics of the water body, improving the accuracy and reliability of the judgment. The adjusted rise hysteresis threshold can more accurately determine the upward trend of pollutant concentrations, considering the impact of surface flow disturbances on pollutant concentration changes, thereby improving the scientific rigor and reliability of pollution event classification.

[0089] When a pollution discharge event is detected as a heavy particulate matter (HPM) event, vertical stratified sampling can be performed on the target water body to obtain the distribution characteristics of HPMs in the water. Specifically, the target water body is vertically divided into an upper homogeneous layer, a gradient abrupt change layer, and a lower homogeneous layer. The upper and lower homogeneous layers are stable regions with relatively small changes in particle concentration, while the gradient abrupt change layer is a transitional region where particle concentration changes significantly. Subsequently, water samples are collected from the upper homogeneous layer, the gradient abrupt change layer, and the lower homogeneous layer, and the HPM concentration of each water sample is measured. Through these steps, information on the distribution of HPMs in the water body at different vertical levels can be obtained, thus providing accurate data support for pollution event analysis, pollutant migration pattern research, and water quality management.

[0090] Subsequently, based on the heavy particle concentrations in the upper homogeneous layer, the gradient abrupt change layer, and the lower homogeneous layer, the stratification intensity index of the water body can be further calculated to quantify the vertical distribution characteristics of heavy particles in the water body. First, the heavy particle concentrations in the upper homogeneous layer, the gradient abrupt change layer, and the lower homogeneous layer are obtained, denoted as the upper homogeneous layer concentration, the intermediate gradient abrupt change layer concentration, and the lower homogeneous layer concentration, respectively. Then, the absolute values ​​of the heavy particle concentration differences between any two adjacent layers are calculated, i.e., the absolute values ​​of the concentration differences between the upper homogeneous layer and the gradient abrupt change layer, and the absolute values ​​of the concentration differences between the gradient abrupt change layer and the lower homogeneous layer. Adding these absolute values ​​yields the stratification intensity index of the water body. The stratification intensity index reflects the vertical distribution variation of heavy particles in the water body; a larger value indicates more pronounced stratification, while a smaller value indicates a higher degree of water mixing, thus providing reliable data support for pollution incident analysis, heavy particle migration pattern research, and water quality management.

[0091] After obtaining the stratification intensity index, it is correlated with a preset hysteresis threshold, and the original hysteresis threshold is adjusted according to the magnitude of the stratification intensity index. Specifically, when the stratification intensity index is large, it indicates that the water body is clearly stratified, and the descent rate of pollutants in different vertical layers may be delayed. Therefore, the hysteresis threshold needs to be increased or decreased accordingly to accurately reflect the starting point of pollutant descent. When the stratification intensity index is small, it indicates that the water body is uniformly mixed, and the original hysteresis threshold can be directly applied.

[0092] By dynamically adjusting the hysteresis threshold in conjunction with the stratification intensity index, adaptive judgment of the pollution event regression process can be achieved. When the stratification intensity is high, the system automatically increases the hysteresis threshold, ensuring that the pollution event is only judged to have entered a steady-state phase when the monitored parameters show a significant decline. This avoids misjudging short-term fluctuations caused by the stability of vertical water stratification as pollution regression. When the stratification intensity is low and vertical mixing is enhanced, the system correspondingly decreases the hysteresis threshold to promptly identify the true decline trend of the pollution event. Through this method, the hysteresis threshold can be adaptively adjusted for different stratification intensities, effectively reducing misjudgments and delays, and improving the accuracy and robustness of defining the time sequence stages of pollution events.

[0093] Based on different types of pollution emission events, surface disturbance index and stratification intensity index are calculated separately, and then the rising hysteresis threshold and falling hysteresis threshold are adjusted accordingly. This allows the classification of pollution events to better adapt to the characteristics and changing patterns of different types of pollution events, improves the accuracy and rationality of the classification, avoids the inaccuracies caused by a one-size-fits-all classification method, enhances the adaptability of pollution event classification, and improves the utilization efficiency of monitoring data.

[0094] In one embodiment of this example, the water quality sampling frequency and water quality sampling volume corresponding to each surface sampling point and vertical sampling point of the water body are determined by combining the time sequence stage definition, including: S710. Determine the corresponding surface disturbance index and stratification intensity index for each time series stage; S720. By defining the corresponding surface disturbance index and stratification intensity index for each time series stage, the corresponding surface disturbance level and vertical stratification level are determined, wherein the surface disturbance level is high surface disturbance and low surface disturbance, and the vertical stratification level is high vertical stratification and low vertical stratification. S730. Define the corresponding surface disturbance level and vertical stratification level according to each time series stage, and determine the water quality sampling frequency and water quality sampling volume of the water surface sampling point and the water vertical sampling point respectively.

[0095] First, the time-series stages of the pollution event are defined, and the corresponding surface disturbance index and stratification intensity index are obtained for each stage. This ensures that the degree of surface disturbance and the vertical distribution characteristics of pollutants are accurately reflected at each stage. Dividing the pollution event into different time periods, such as transient, transitional, and steady-state stages, and then calculating the surface disturbance index and stratification intensity index for each stage, provides a reliable basis for subsequent pollution event threshold adjustments, stage determination, and water quality analysis.

[0096] Secondly, for each time-series stage of a pollution event, the surface disturbance level and vertical stratification level of the water body can be determined based on the surface disturbance index and stratification intensity index corresponding to each time-series stage. The surface disturbance level includes two categories: high surface disturbance and low surface disturbance. High surface disturbance indicates strong mixing and significant disturbance at the water surface, while low surface disturbance indicates weak mixing and minimal disturbance at the water surface. The vertical stratification level includes two categories: high vertical stratification and low vertical stratification. High vertical stratification indicates significant vertical stratification of pollutants in the water body with large concentration differences, while low vertical stratification indicates homogeneous mixing and small vertical concentration differences. Through these steps, continuous numerical indicators are transformed into clear level classifications, providing an operational basis for subsequent pollution event threshold adjustments, stage determination, and water quality analysis.

[0097] Finally, for each time-series stage of the pollution event, the water quality sampling frequency and volume for surface and vertical sampling points can be determined based on the surface disturbance level and vertical stratification level corresponding to each time-series stage. Specifically, for time-series stages with higher surface disturbance levels, surface sampling points can be set to higher sampling frequencies and larger sampling volumes to capture rapidly changing pollution characteristics on the water surface; for time-series stages with lower surface disturbance levels, the sampling frequency and volume can be appropriately reduced. Similarly, for time-series stages with higher vertical stratification levels, vertical sampling points can be set to higher sampling frequencies and larger sampling volumes to reflect the obvious stratification characteristics of vertical pollutant concentrations in the water body; for time-series stages with lower vertical stratification levels, the sampling frequency and volume can be appropriately reduced. Through the above steps, the sampling strategy can be dynamically adjusted according to water disturbance and stratification characteristics, enabling the collected water quality data to accurately reflect the changing characteristics of the pollution event at different time periods and different water layers, while improving sampling efficiency and resource utilization.

[0098] Based on the surface disturbance index and stratification intensity index of different time-series stages, different surface disturbance levels and vertical stratification levels are determined, and then the corresponding water quality sampling frequency and volume are determined for each water sampling point. This differentiated strategy can fully consider the pollution change characteristics of different stages and different water layers, making the sampling work more scientific and flexible, avoiding the problems of insufficient or excessive sampling caused by a one-size-fits-all sampling method, realizing a differentiated sampling strategy, and improving sampling efficiency and accuracy.

[0099] In one embodiment of this invention, based on defining the corresponding surface disturbance level and vertical stratification level for each time stage, the water quality sampling frequency and water quality sampling volume for surface sampling points and vertical sampling points in the water area are determined, including: S810 When the time sequence stage is defined as a transient segment and the surface disturbance index is high surface disturbance, the water quality sampling frequency and water quality sampling volume corresponding to the water surface sampling point are determined by sampling using the first preset frequency and the third preset volume. S820 When the time sequence is defined as a transient segment and the surface disturbance index is low, the water quality sampling frequency and water quality sampling volume corresponding to the water surface sampling point are determined by sampling using the second preset frequency and the second preset volume. S830. When the time sequence stage is defined as the steady state stage, the water quality sampling frequency and water quality sampling volume corresponding to the water surface sampling point are determined by sampling using the fourth preset frequency and the first preset volume. S840 When the time sequence stage is defined as a transient segment and the vertical stratification level is high vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the first preset frequency and the third preset volume. S850 When the time sequence stage is defined as a transient segment and the vertical stratification level is low vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the second preset frequency and the second preset volume. S860 When the time sequence stage is defined as a steady state segment and the vertical stratification level is high vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the third preset frequency and the first preset volume. S870 When the time sequence stage is defined as a steady state segment and the vertical stratification level is low vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the fourth preset frequency and the first preset volume. S880, wherein the first preset frequency is greater than the second preset frequency, the second preset frequency is greater than the third preset frequency, and the third preset frequency is greater than the fourth preset frequency, and the first preset volume is greater than the second preset volume, and the second preset volume is greater than the third preset volume.

[0100] When the timeframe of a pollution event is defined as the transient phase, and the surface disturbance index is high, water quality sampling can be conducted at surface sampling points using a first preset frequency and a third preset volume to determine the sampling frequency and volume for that sampling point during the transient high-disturbance phase. The reason for using a higher sampling frequency is that pollutant concentrations change rapidly within a short period during the transient phase. If the sampling frequency is too low, peak concentrations or short-term fluctuations may be missed, resulting in sampling data that does not accurately reflect water quality changes. The reason for using a larger sampling volume is that high surface disturbance indicates intense mixing at the water surface and localized heterogeneity in pollutant distribution; a small single sampling volume may not be sufficient to represent the overall water condition. Through this high-frequency, large-volume sampling, the rapid changes in surface pollutants can be fully captured, improving the representativeness and reliability of the sampling data, thereby providing accurate and reliable data for subsequent pollution event analysis, threshold determination, and water quality management.

[0101] When the time series of a pollution event is defined as a transient phase, and the surface disturbance index is low, water quality sampling can be conducted at surface sampling points using a second preset frequency and a second preset volume to determine the sampling frequency and volume for that sampling point during the transient low-disturbance phase. Since low surface disturbance indicates weak mixing at the water surface and relatively small fluctuations in pollutant concentrations, the sampling frequency and volume can be appropriately reduced compared to the high-disturbance phase. This ensures that the collected data still reflects the changing characteristics of pollutants in the water body, while avoiding oversampling and improving resource utilization efficiency. By employing an appropriate sampling frequency and volume, the representativeness and reliability of the collected water quality data can be guaranteed.

[0102] When the time series of a pollution event is defined as a steady-state phase, water quality samples can be taken from surface sampling points using a fourth preset frequency and a first preset volume to determine the sampling frequency and volume for that sampling point during the steady-state phase. Since the pollutant concentration in the water body changes gradually and water quality fluctuations are small during the steady-state phase, frequent sampling is unnecessary, and the sampling frequency can be set to a relatively low fourth preset frequency. Simultaneously, to ensure the representativeness of the sampled water, a larger sampling volume, i.e., the first preset volume, can be used. This low-frequency, large-volume sampling method conserves resources while ensuring that the collected water quality data accurately reflects the pollution level of the water body.

[0103] When the timeframe of a pollution event is defined as a transient phase, and the vertical stratification level of the water body is high vertical stratification, water quality sampling can be conducted at vertical sampling points in the water body using a first preset frequency and a third preset volume to determine the water quality sampling frequency and volume at that sampling point during the transient high vertical stratification phase. Since high vertical stratification indicates significant differences in pollutant concentrations at different depths of the water body, and pollutant concentrations change rapidly during the transient phase, a higher sampling frequency and a larger sampling volume are required to ensure that a sufficient number of samples are obtained within a short time and to fully reflect the vertical distribution characteristics of pollutants in the water body. This high-frequency, large-volume sampling method ensures the representativeness and reliability of the vertical sampling data, providing an accurate basis for subsequent pollution event analysis, threshold determination, and water quality management.

[0104] When the timeframe of a pollution event is defined as a transient phase, and the vertical stratification level of the water body is low, water quality sampling can be conducted at vertical sampling points using a second preset frequency and a second preset volume to determine the sampling frequency and volume for that sampling point during the transient low vertical stratification phase. Since low vertical stratification indicates relatively small differences in pollutant concentrations at different depths of the water body, and the vertical mixing of the water is relatively uniform, representative water quality data can be obtained without using high-frequency or large-volume sampling. In this phase, a sampling method with moderate frequency and volume can fully reflect the changing characteristics of vertical pollutants while conserving sampling resources and improving sampling efficiency.

[0105] When the time series of a pollution event is defined as a steady-state phase, and the vertical stratification level of the water body is high vertical stratification, water quality sampling can be conducted at vertical sampling points in the water body using a third preset frequency and a first preset volume to determine the water quality sampling frequency and volume at that sampling point during the steady-state high vertical stratification phase. Because the pollutant concentration in the water body changes gradually in the steady-state phase, but vertical stratification still exists significantly, with large differences in pollutant concentrations at different depths, a moderate sampling frequency and a large sampling volume are required to ensure that water samples at each depth accurately reflect their water quality characteristics. Through the above sampling method, the sampling frequency can be appropriately reduced while ensuring the representativeness and reliability of the sampling data, thereby saving sampling resources and improving sampling efficiency.

[0106] When the time series of a pollution event is defined as a steady-state phase, and the vertical stratification level of the water body is low, water quality sampling can be conducted at vertical sampling points in the water body using a fourth preset frequency and a first preset volume to determine the water quality sampling frequency and volume for that sampling point in the steady-state low-vertical stratification phase. Since the pollutant concentration changes gradually in the steady-state phase, and vertical stratification is not obvious, the water body is relatively homogeneous, thus representative water samples can be obtained without high-frequency sampling. By using a lower sampling frequency and a larger sampling volume, it is possible to ensure that the sampling data accurately reflects the vertical water quality characteristics of the water body while conserving sampling resources. When the time series is defined as a steady-state phase, it indicates that the water body is in a relatively balanced and stable state, that is, the surface disturbance index of the water body is stable. At this time, the water body as a whole (surface and deep layers) exhibits high temporal stability and spatial homogeneity, and the water parameters (such as water quality) change slowly with small fluctuations. Under these stable conditions, monitoring surface water quality does not require very high frequencies or large sampling volumes to capture rapid changes. Therefore, a lower preset frequency (fourth preset frequency) and a smaller preset volume (first preset volume) are used for data collection. Similarly, when the time series is defined as a steady-state phase and the vertical stratification level is low, the water quality across the entire water area is relatively uniform and stable. In this case, monitoring vertical water quality also does not require very high frequencies or large sampling volumes. Similar to the steady-state situation of the surface, water quality changes slowly and uniformly; therefore, the same lower frequency (fourth preset frequency) and smaller volume (first preset volume) are sufficient for effective monitoring. Thus, a relatively low sampling frequency and a smaller sampling volume, i.e., the "fourth preset frequency" and the "first preset volume," can be used to save resources while still obtaining effective data.

[0107] In one embodiment of the present invention, there is a decreasing relationship between the sampling frequency and the sampling volume, namely, the first preset frequency is greater than the second preset frequency, the second preset frequency is greater than the third preset frequency, and the third preset frequency is greater than the fourth preset frequency; simultaneously, the first preset volume is greater than the second preset volume, and the second preset volume is greater than the third preset volume. The purpose of this design is to rationally adjust the sampling strategy according to different stages of water pollution events and water disturbance characteristics. When water disturbance is strong or pollutant concentration changes rapidly, high-frequency and large-volume sampling is used to fully capture instantaneous changes in pollutant concentration and ensure the representativeness of the water sample; when water disturbance is low or pollutant concentration changes are gradual, lower-frequency and moderate-volume sampling is used to reduce unnecessary sampling times and resource consumption, while still ensuring the accuracy and reliability of water quality data. Specifically, there is a positively correlated decreasing relationship between the first, second, and third preset volumes and the first, second, third, and fourth preset frequencies, i.e., the higher the sampling frequency, the larger the corresponding single sampling volume; the lower the sampling frequency, the smaller the corresponding single sampling volume. This correspondence is mainly determined based on the characteristics between the intensity of water disturbance and the rate of change of water quality parameters. When the water body is in a transient phase and the surface disturbance or vertical stratification intensity is high, the water quality parameters change rapidly and the local gradient is significant. To ensure the temporal continuity and spatial representativeness of the sampling data, the system needs to use a high sampling frequency and a large single sampling volume. When the water body disturbance is weak or in a steady-state phase, the water quality parameters change gradually, the water body is well mixed, and the parameter change amplitude is small. At this time, a lower sampling frequency and a smaller sampling volume can be used to optimize the energy consumption of data acquisition and suppress sampling redundancy. Therefore, the first preset frequency and the third preset volume correspond to the strong disturbance and high change state, the second preset frequency and the second preset volume correspond to the moderate disturbance state, and the third and fourth preset frequencies and the first preset volume correspond to the steady-state stage, thus constructing a unidirectional hierarchical matching relationship between frequency and volume. In addition, the fourth preset volume is not defined separately in this invention. The technical reason is that when the timing stage is defined as the steady-state segment, the water body disturbance amplitude is weak, and the water quality parameters tend to be stable in both time and space. Further reducing the sampling volume has limited effect on improving data representativeness and will instead increase the complexity of system parameter setting and computational burden. Based on this, the present invention uniformly adopts a first preset volume as the minimum sampling volume configuration in the steady state stage, so as to balance the three aspects of sampling accuracy, energy consumption control and algorithm simplification.

[0108] Based on changes in time series, disturbance index, and stratification level, the sampling frequency and volume of sampling points can be flexibly adjusted. This allows for better adaptation to the occurrence, development, and stabilization stages of different pollution events, ensuring the targetedness and effectiveness of sampling work and making monitoring more efficient. Monitoring resources are concentrated during the critical stage (transient phase) of a pollution event to quickly acquire effective data, while resource input is reduced during the non-critical stage (steady-state phase), achieving an overall improvement in monitoring efficiency.

[0109] This application provides an electronic device, including: The memory is configured to store instructions; and The processor is configured to retrieve instructions from memory and, when executing instructions, implement the aforementioned multi-parameter linkage water quality sampling method.

[0110] This application also provides a machine-readable storage medium storing instructions that cause a machine to perform the above-described multi-parameter linkage water quality sampling method.

[0111] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0112] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0113] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0114] These computer program instructions can also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0115] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0116] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0117] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0118] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0119] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A multi-parameter linkage water quality sampling method, characterized in that, include: Acquire hydrological data and main transport route data of the target monitoring water area, as well as pollutant emission data of the target monitoring enterprises; Based on hydrological data, pollutant discharge data, and main transport path data, water sampling points are set for the target monitoring water area. The water sampling points include water surface sampling points and water vertical sampling points. Water quality samples were collected at both the surface sampling point and the vertical sampling point of the water area to obtain sampling point data; The validity of the sampling point data is verified to obtain the validity verification results, and the validity of the water surface sampling points and the water vertical sampling points is determined based on the validity verification results; When the surface sampling point and the vertical sampling point of the water body are valid sampling points, water quality monitoring data of the surface sampling point and the vertical sampling point of the water body are obtained; The pollution source linkage index is calculated based on water quality monitoring data, and the type of pollution discharge event is determined based on the pollution source linkage index. Pollution events are classified and their temporal stages are defined based on the type of pollution discharge event and water quality monitoring data. Based on the temporal stage definition, the water quality sampling frequency and water quality sampling volume corresponding to each surface sampling point and vertical sampling point of the water area are determined respectively.

2. The method according to claim 1, characterized in that, The water sampling points for setting target monitoring water areas based on hydrological data, pollutant discharge data, and main transport path data include: Wind field data and surface flow field data are obtained from the hydrological data of the target monitoring water area; By monitoring the pollutant emission data of target enterprises, we can obtain the location of pollution sources, emission time, emission flow rate, and emission concentration. Emission flow and emission concentration are converted into pollutant release flux, and wind field data are converted into wind stress according to a preset wind stress conversion rate; The simulation time window is set in conjunction with the emission time; Using the location of the pollution source as the starting point of diffusion, a migration and diffusion model is constructed within the simulation time window by combining data on pollutant release flux, wind stress, and surface flow field. Pollutant isochrones were extracted based on a migration and diffusion model, and target intersections were extracted by combining pollutant isochrones with main transport path data. The surface diffusion boundary and concentration gradient distribution of pollutants are calculated based on pollutant release flux, surface flow field data, and wind stress. Sampling points on the water surface were determined by combining the target intersection point, surface diffusion boundary, and concentration gradient distribution. Vertical sampling points for target monitoring water areas are set based on surface sampling points, hydrological data, and pollutant emission data.

3. The method according to claim 2, characterized in that, The vertical sampling points for setting target monitoring water areas based on water surface sampling points, hydrological data, and pollutant discharge data include: Water temperature profile data and water density profile data are obtained from the hydrological data of the target monitoring water area; Identify the water temperature gradient and water density gradient in the target monitoring area based on water temperature profile data and water density profile data; By combining the water temperature jump layer and the water density jump layer, the target monitoring water area is divided into an upper uniform layer, a gradient abrupt layer and a lower uniform layer. For any water surface sampling point, construct a surface vertical profile, and combine the surface vertical profile, upper uniform layer, gradient abrupt layer and lower uniform layer to determine the set of vertical candidate sampling points corresponding to each water surface sampling point; The priority vertical sampling layer is determined by filtering each set of vertical candidate sampling points based on the maximum information gain criterion. The types of pollutants emitted are determined by pollutant emission data, and the vertical distribution characteristics corresponding to the types of pollutants emitted are determined in a pre-set database; Vertical sampling points for the target monitoring water area are determined by combining vertical distribution characteristics and priority vertical sampling layers.

4. The method according to claim 1, characterized in that, The sampling point data includes data from a first time point and data from a second time point. The process of validating the sampling point data to obtain a validity verification result, and determining the validity of surface sampling points and vertical sampling points based on the validity verification result, includes: For any water area sampling point, data at the first preset time point and data at the second preset time point are collected respectively. The relative deviation value is calculated based on the data from the first time point and the data from the second time point, and the first validity is determined by the relative deviation value. For any given water sampling point, obtain the original concentration of the water sample to be tested; The amount of the target pollutant spiked is obtained by adding a standard solution of the target pollutant of known concentration to the water sample to be tested. The concentration of the target pollutant after spiking is obtained by analyzing the water sample to be tested with a standard solution of the target pollutant at a known concentration. The second validity is calculated based on the spiked concentration of the target pollutant, the amount of the target pollutant spiked, and the original concentration. The overall validity is calculated based on the first and second validity levels; If the overall validity is greater than or equal to the preset validity threshold, then the corresponding water surface sampling point and water vertical sampling point are determined to be valid.

5. The method according to claim 1 or 4, characterized in that, The method of classifying pollution events based on pollution discharge event types and water quality monitoring data to determine the temporal stages of pollution events includes: The concentration of pollutants, the rate of change of concentration, and the abnormal values ​​of pollutant concentration are obtained through water quality monitoring data. Based on the type of pollution emission event, select the corresponding type parameter group from the preset event type library. The type parameter group includes rising hysteresis threshold, falling hysteresis threshold, forgetting factor and minimum residence time. The forgetting factor is used to characterize the memory length of event energy for historical data and the weight decay rate. Based on the type parameter group, the type linkage index is calculated using a preset multidimensional fusion model. The type linkage index is used to characterize the degree of matching between the pollution emission event type and the target pollution type. A comprehensive anomaly score is obtained by weighted fusion calculation combining pollutant concentration, concentration change rate, and pollutant concentration anomalies. The event energy is calculated using a preset event energy formula by combining the forgetting factor and the comprehensive abnormality score. The time trigger moment and the time end moment are determined by combining the rising hysteresis threshold and the falling hysteresis threshold. Within the time trigger moment and time end moment, the event energy and type linkage index are combined with the minimum residence time to divide and determine the transient segment, transition segment and steady-state segment of the pollution emission event type. The transient segment, transition segment and steady-state segment are used to characterize the time sequence stage definition.

6. The method according to claim 5, characterized in that, The pollution emission event types include volatile events and heavy particulate events, and the method further includes: When the pollution emission event type is a volatile event, obtain surface flow velocity data from sampling points on the water surface; The mean surface velocity is calculated based on the surface velocity data, and the sum of squares of velocity deviations is calculated based on the mean surface velocity. The surface disturbance index is calculated by combining the sum of squared velocity deviations and the number of sampling points on the water surface. The adjusted rise hysteresis threshold is obtained by combining the surface disturbance index with the threshold adjustment. When the pollution emission event type is a heavy particulate event, the heavy particulate concentrations of the upper homogeneous layer, gradient abrupt layer and lower homogeneous layer are obtained respectively. The absolute value of the heavy particle concentration between any two adjacent layers is calculated by combining the heavy particle concentrations of the upper homogeneous layer, the gradient abrupt layer, and the lower homogeneous layer, and the absolute values ​​are added together to obtain the stratification intensity index. The adjusted descent hysteresis threshold is obtained by combining the stratification intensity index with the threshold adjustment.

7. The method according to claim 6, characterized in that, The determination of the water quality sampling frequency and water quality sampling volume corresponding to each surface sampling point and vertical sampling point in the water area, based on the combined time-series stage definition, includes: Determine the corresponding surface disturbance index and stratification intensity index for each time series stage; The corresponding surface disturbance index and stratification intensity index are defined for each time series stage to determine the corresponding surface disturbance level and vertical stratification level, where the surface disturbance level is high surface disturbance and low surface disturbance, and the vertical stratification level is high vertical stratification and low vertical stratification. Based on the definition of the corresponding surface disturbance level and vertical stratification level for each time series stage, the water quality sampling frequency and water quality sampling volume for surface sampling points and vertical sampling points in the water area are determined respectively.

8. The method according to claim 7, characterized in that, The process of defining the corresponding surface disturbance level and vertical stratification level according to each time stage, and determining the water quality sampling frequency and water quality sampling volume of surface sampling points and vertical sampling points in the water area, includes: When the timing phase is defined as the transient phase and the surface disturbance index is high, the water quality sampling frequency and water quality sampling volume corresponding to the water surface sampling point are determined by sampling using the first preset frequency and the third preset volume. When the timing phase is defined as the transient phase and the surface disturbance index is low, the water quality sampling frequency and water quality sampling volume corresponding to the water surface sampling point are determined by sampling using the second preset frequency and the second preset volume. When the timing phase is defined as the steady state phase, the water quality sampling frequency and water quality sampling volume corresponding to the sampling point on the water surface are determined by sampling using the fourth preset frequency and the first preset volume. When the timing phase is defined as a transient phase and the vertical stratification level is high vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the first preset frequency and the third preset volume. When the time sequence stage is defined as a transient segment and the vertical stratification level is low vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the second preset frequency and the second preset volume. When the time sequence stage is defined as a steady state and the vertical stratification level is high vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the third preset frequency and the first preset volume. When the time sequence stage is defined as a steady state and the vertical stratification level is low vertical stratification, the water quality sampling frequency and water quality sampling volume corresponding to the vertical sampling point of the water area are determined by sampling using the fourth preset frequency and the first preset volume. Among them, the first preset frequency is greater than the second preset frequency, the second preset frequency is greater than the third preset frequency, and the third preset frequency is greater than the fourth preset frequency; the first preset volume is greater than the second preset volume, and the second preset volume is greater than the third preset volume.

9. An electronic device, characterized in that, include: The memory is configured to store instructions; as well as The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the multi-parameter linkage water quality sampling method according to any one of claims 1 to 8.

10. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing the machine to perform the multi-parameter linkage water quality sampling method according to any one of claims 1 to 8.

Images