Real-time water environment assessment method based on fully-mixed model

Abstract

The invention provides a real-time water environment assessment method based on a fully-mixed model, and relates to the technical field of water environment protection, and the method comprises the steps: obtaining the flow and each water environment index concentration of a to-be-assessed cross section and an upstream cross section in real time, thereby predicting each water environment index concentration of the to-be-assessed cross section at the current moment after customer water merges; and calculating a Nemerow comprehensive pollution index so as to evaluate the water environment. According to the method, the Neimero comprehensive pollution index of the to-be-evaluated section at each moment is predicted through the section total hybrid model, and the influence of the upstream section is considered in prediction, so that the comprehensive index not only reflects the pollution of a single section, but also implies the pollution transmission information of the upstream and downstream hybrid process. Meanwhile, the Nemerow comprehensive pollution indexes predicted at a plurality of continuous moments can form time sequence association, and the method is more suitable for the actual scene that river pollution dynamically changes along with time. The data timeliness and the section relevance are comprehensively considered, and the real-time performance and the accuracy of water environment quality evaluation are improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of water environment protection, in particular to a real-time water environment evaluation method based on a complete mixing model. BACKGROUND

[0002] Water resources have dual attributes of quality and quantity. Climate change and human activities have a significant impact on the water environment quality of a basin. With the rapid development of urban and rural economies, non-point source pollution caused by rainfall runoff has gradually become the main source of river pollution. A large amount of pollutants directly enters the receiving water body with the runoff during rainfall, which rapidly deteriorates the water quality of the receiving water body. The influx of a large amount of exogenous water breaks the dynamic balance of the geochemical action of the original water body environment, which directly affects the water chemical characteristics of the water environment and ultimately affects the surrounding water environment quality. However, there is a lack of research on the real-time changes of water bodies under mixed dilution in the study of water environment. The application constructs a complete mixing model of water quality by using real-time monitoring data of a section, carries out real-time comprehensive evaluation of the water environment, and analyzes the water environment change characteristics under water mixing.

[0003] At present, water mixing is mainly studied from the aspects of mixing ratio and mixing reasons, and combined with the hydrodynamic action of mixed dilution, there is a lack of analysis on the results of real-time water mixing from the perspective of water chemistry, which leads to low accuracy of water environment evaluation results of the water environment, thereby affecting the accuracy of water environment capacity calculation and water environment vulnerability evaluation. SUMMARY

[0004] The application aims to provide a real-time water environment evaluation method based on a complete mixing model to solve the above problems. In order to achieve the above purpose, the technical scheme adopted by the application is as follows: In a first aspect, the application provides a real-time water environment evaluation method based on a complete mixing model, which comprises: obtaining first information and second information, the first information and the second information being real-time section data of a first section and a second section respectively, the section data comprising section flow and the concentrations of a plurality of water environment indexes of the section, and the second section being a section upstream of the first section; calculating third information based on the first information of the last moment, the second information of the current moment and a preset section complete mixing model, the third information being the concentrations of a plurality of water environment indexes of the first section after the influx of guest water at the current moment; calculating a Nemerow comprehensive pollution index based on the third information and an improved Nemerow index method to obtain the Nemerow comprehensive pollution index; performing water environment evaluation of the first section based on the Nemerow comprehensive pollution index to obtain an evaluation result.

[0005] Furthermore, after obtaining the first information and the second information, the process further includes: Obtain a first time series and a second time series, wherein the first time series includes the first information at multiple first moments within a first time period, and the second time series includes the second information at multiple moments within the first time period; Obtain fourth information, which includes cross-sectional data from previous years that are concurrent with and under the same working conditions as the third cross-section in the first time period, wherein the third cross-section is the first cross-section and the second cross-section. Based on the fourth information, the third time series is preprocessed to obtain the preprocessed third time series, which includes the first time series and the second time series. The third information at multiple first time points is calculated based on the preprocessed third time series to obtain the third information at multiple first time points; The water environment assessment of the first cross section during the first time period is performed based on the third information at multiple first moments.

[0006] Furthermore, the preprocessing of the third time series based on the fourth information includes: Data with excessive missing values ​​in the third time series were removed based on the missing value detection algorithm. The first data in the third time series is removed based on multiple preset threshold intervals. Each threshold interval corresponds to a water environment indicator, and the first data is the water environment indicator that exceeds its corresponding threshold interval. The continuity of the third time series is analyzed using the moving average method or exponential smoothing method, and abrupt data is removed. The third and fourth time series were calculated based on the Pearson correlation coefficient algorithm and the fourth information. The fourth time series The third time series is constructed from historical data of the same period and working conditions, and data within the third time series is removed based on the Pearson correlation coefficient. The third time series is filled by interpolation based on the changing trend of the data within the third time series after data removal.

[0007] Furthermore, the threshold range includes both absolute thresholds and relative thresholds; The absolute threshold is set based on the surface water environmental quality standard; The relative threshold is calculated based on the cross-sectional data of the cross-section corresponding to the relative threshold in the fourth information, which are from the same period and under the same working conditions over the years. The relative threshold is based on the first cross-section. .

[0008] Furthermore, the cross-sectional fully hybrid model is represented as follows: ; in, For prediction The first cross section at time The day after the inflow of passenger water The concentration of each water environment indicator The first cross-section to be monitored exist Cross-sectional flow rate at any given time The first cross-section to be monitored exist The first moment The concentration of each water environment indicator The second cross-section to be monitored exist Cross-sectional flow rate at any given time The second cross-section to be monitored exist The first moment The concentration of each water environment indicator.

[0009] Furthermore, the calculation of the Nemerow comprehensive pollution index based on the third information and the improved Nemerow index method includes: Obtain the standard value corresponding to each of the aforementioned water environment indicators; The single-factor index of the first water environment index is calculated based on the concentration of the first water environment index predicted at the current moment and the standard value corresponding to the first water environment index. Calculate the single-factor index corresponding to each of the water environment indicators to obtain multiple single-factor indices; The Nemerow composite pollution index is calculated based on the average and maximum values ​​of multiple single-factor indices and the Nemerow index method.

[0010] Furthermore, the Nemerow comprehensive pollution index, calculated based on the average and maximum values ​​of multiple single-factor indices and the Nemerow index method, can be expressed as: ; in, for The first section at time 1 The Nemerow Comprehensive Pollution Index, for The first section at time 1 The The single-factor indices corresponding to the aforementioned water environment indicators. for Multiple single-factor indices at time points The average value, for Single-factor indices corresponding to multiple water environment indicators at different times The maximum value.

[0011] Furthermore, the water environment assessment of the first cross-section based on the Nemerow Integrated Pollution Index includes: The water quality level of the first cross section is determined based on the preset water quality classification standards and the Nemerow Comprehensive Pollution Index. The water quality level includes excellent, good, relatively good, poor, and very poor.

[0012] Secondly, this application also provides a real-time water environment assessment system based on a fully mixed model, comprising: The information acquisition module is used to acquire first information and second information, wherein the first information and the second information are real-time cross-sectional data of the first cross-section and the second cross-section, respectively. The cross-sectional data includes cross-sectional flow and the concentration of multiple water environment indicators of the cross-section. The second cross-section is the cross-section upstream of the first cross-section. The first processing module is used to calculate the third information based on the first information at the previous moment, the second information at the current moment, and the preset cross-sectional full-mix model. The third information is the predicted concentration of multiple water environment indicators at the first cross-section at the current moment after the inflow of passenger water. The second processing module is used to calculate the Nemerow Comprehensive Pollution Index based on the third information and the improved Nemerow Index method, thereby obtaining the Nemerow Comprehensive Pollution Index. The third processing module is used to conduct a water environment assessment of the first cross section based on the Nemerow Comprehensive Pollution Index and obtain the assessment results.

[0013] The beneficial effects of this invention are as follows: This invention predicts the Nemerow composite pollution index of the assessed cross-section at each time point using a cross-sectional full-mixing model. The prediction considers the influence of upstream cross-sections, ensuring that the composite index not only reflects pollution at a single cross-section but also implicitly includes pollution transfer information from upstream to downstream mixing processes, thus enhancing the spatial correlation of the assessment. Furthermore, the predicted Nemerow composite pollution indices from multiple consecutive time points can form a time-series correlation, better reflecting the actual scenario of dynamic changes in river pollution over time. By comprehensively considering data timeliness and cross-sectional correlation, this invention enables real-time and accurate assessment of water environmental quality, providing a refined scientific explanation for changes in the water environment and offering a technical basis for water environment early warning.

[0014] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating a real-time water environment assessment method based on a fully mixed model, as described in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0018] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0019] Example 1: This embodiment provides a real-time water environment assessment method based on a fully mixed model.

[0020] See Figure 1 The figure shows that the method includes steps S1, S2, S3 and S4.

[0021] S1. Obtain first information and second information, wherein the first information and the second information are real-time cross-sectional data of the first cross-section and the second cross-section, respectively. The cross-sectional data includes cross-sectional flow and the concentration of multiple water environment indicators of the cross-section. The second cross-section is the cross-section upstream of the first cross-section. Specifically, in this embodiment, relevant equipment is deployed at the section to be evaluated and its upstream section to monitor the cross-sectional flow and various water environment indicators of the section in real time, thereby obtaining real-time data of the section. The main water environment indicators collected at this section include water temperature, turbidity, conductivity, suspended solids (SS), pH value, dissolved oxygen (DO), chemical oxygen demand (COD), biochemical oxygen demand (BOD5), ammonia nitrogen (NH3−N), total nitrogen (TN), total phosphorus (TP), heavy metals (mercury, cadmium), volatile organic compounds (VOCs), suspended solids (SS), and turbidity.

[0022] Furthermore, this embodiment uses historical data to systematically screen real-time data in a comparison system, thereby obtaining more reliable data and improving the accuracy of subsequent Nemerow composite pollution index calculations. Specific steps include: S11. Obtain a first time series and a second time series, wherein the first time series includes the first information at multiple first moments within a first time period, and the second time series includes the second information at multiple moments within the first time period; S12. Obtain fourth information, the fourth information including cross-sectional data from previous years that are concurrent with and under the same working conditions as the third cross-section in the first time period, the third cross-section being the first cross-section and the second cross-section. S13. Based on the fourth information, the third time series is preprocessed to obtain the preprocessed third time series, which includes the first time series and the second time series. S14. Calculate the third information at multiple first time points based on the preprocessed third time series to obtain the third information at multiple first time points; S15. Based on the third information at multiple first moments, conduct a water environment assessment of the first cross section during the first time period.

[0023] Understandably, the process begins by using the inherent physicochemical properties of the aquatic environment (such as the natural range of pH and dissolved oxygen values) to eliminate obviously abnormal real-time data. Then, historical data from the same period and under the same operating conditions are used as a benchmark to verify the rationality and representativeness of the real-time data, removing data that deviates from the normal fluctuation range. Essentially, this leverages the dual constraints of the inherent validity of real-time monitoring data and the referential value of historical data to reduce the unreliability caused by sensor errors and sudden interference, thereby making the final calculated Nemerow composite index for the section to be evaluated more accurate.

[0024] Furthermore, it is understood that in this embodiment, the screening criteria for real-time monitoring data include: (1) Data integrity standard: Real-time data fields are not missing (such as pH, dissolved oxygen, COD, ammonia nitrogen and other key indicators must be collected), data transmission is uninterrupted, and the missing rate is ≤5%.

[0025] (2) Data rationality standard: The value of a single indicator conforms to the natural laws of the water environment, such as pH value 6-9, dissolved oxygen 0-14mg / L, COD≤100mg / L. It is understood that the threshold of the specific data rationality standard can be adjusted according to the surface water / groundwater type.

[0026] (3) Data consistency standard: The real-time data time series of the same monitoring point is smooth, and the fluctuation range of adjacent data is ≤30% (to avoid jumps caused by sudden sensor failure).

[0027] (4) Data correlation standard: The deviation between real-time data and data of the same period in previous years (such as the same season and the same rainfall conditions) is ≤ 2 times the standard deviation, and the trend is consistent, such as COD during the dry season is usually higher than that during the wet season.

[0028] Specifically, in the screening process based on data integrity standards, missing value detection algorithms, such as mean imputation, are used to directly remove data with excessive missing values. A threshold method is also employed, pre-setting normal ranges for each indicator; data exceeding these ranges is marked as abnormal and removed. The methods for determining the normal ranges for each indicator include: 1) Set an absolute threshold, i.e., a baseline threshold, based on national / local water quality standards (such as the "Surface Water Environmental Quality Standard" GB 3838-2002), the physicochemical characteristics of the indicator, and the measurement range of the monitoring instrument. Directly refer to the "limit range" of the indicator in the standard. For example, if the COD of Class III surface water is ≤20mg / L, the absolute threshold can be set to 0-100mg / L to cover normal fluctuations and slight anomalies. The instrument measurement range should be 1.2 times higher than the upper limit of the absolute threshold to avoid misjudgment caused by instrument limitations.

[0029] 2) To calculate the relative threshold, i.e. the fluctuation threshold, it is necessary to filter the valid historical data of the same period in the past 3-5 years (same season, same hydrological conditions, such as dry season and wet season), i.e. the dataset after outliers have been removed.

[0030] For water environment indicators, all data are normally distributed. The relative threshold interval is set using the mean ± k times the standard deviation σ. In this embodiment, k is 2-3, such as k=2 in the normal scenario and k=3 in the strict scenario. For a water environment indicator whose data satisfies a normal distribution, the mean and standard deviation are calculated from the valid historical data for that indicator. For example, a larger k value (e.g., 3) in the strict scenario can broaden the judgment interval for normal data, reduce the probability of valid data being misjudged as abnormal, and suit scenarios with extremely high data reliability requirements. The key reason is based on the characteristics of normal distribution.

[0031] Understandably, different relative threshold ranges are chosen based on different scenario requirements. One reason is the difference in probability coverage. In a normal distribution, a threshold of k=2 covers 95.45% of normal data, while k=3 covers 99.73%. Strict scenarios require minimizing the omission of valid data; k=3 includes almost all regular normal data within the range, excluding only extremely rare outliers (with a probability of only 0.27%). Another reason is the different tolerance requirements. In conventional scenarios (k=2), a small amount of normal data (approximately 4.55%) can be mistakenly deleted to prioritize screening efficiency. Strict scenarios (such as drinking water source monitoring and key river basin early warning) have zero tolerance for misjudgments, requiring a larger k value to reduce the risk of false positives and avoid distorted evaluation results due to data deletion.

[0032] Data impact weight: In strict scenarios, data results are directly related to high-risk decisions (such as water quality safety early warning and pollution control decisions). It is necessary to ensure that all data included is absolutely reliable and normal. A wide range of k=3 can better meet this "conservative and accurate" requirement.

[0033] Furthermore, data from existing monitoring indicators that exhibit skewed distributions due to factors such as sudden pollution events are screened. The core source of skewed data is the distribution changes of existing monitoring indicators under specific scenarios, rather than newly added indicators. The screening targets mainly include the following core indicators that are easily affected by sudden pollution events: conventional pollution indicators including COD, ammonia nitrogen, total phosphorus, and total nitrogen (easily skewed by sudden discharges from sewage outlets and runoff from rainfall, such as a surge in concentration at certain times, resulting in a right-skewed distribution); toxic and hazardous indicators including heavy metals (mercury, cadmium) and volatile organic compounds (VOCs) (sudden industrial leaks can cause a sharp increase in local concentrations, causing the data distribution to deviate from normal); and physical indicators including suspended solids (SS) and turbidity (data skewed by sediment inflow after heavy rain or temporary construction). These indicators are core items in water environment monitoring, but their data are skewed under sudden scenarios. Therefore, the mean ± kσ of the normal distribution needs to be replaced with IQR to avoid extreme values ​​being misjudged as anomalies.

[0034] It is understandable that skewed data is non-normally distributed data. For existing water environment indicators that are prone to skewed distribution, a relative threshold range is set by the interquartile range (IQR), namely Q1-1.5IQR to Q3+1.5IQR, where Q1 is the lower quartile and Q3 is the upper quartile.

[0035] It is worth noting that when selecting indicator data, both absolute threshold ranges and relative threshold ranges must be met.

[0036] 3) Scene calibration for dynamic adjustment. Specifically, corrections are made for special scenarios, such as after rainfall or near sewage outlets. The relative threshold range can be widened by 10%-30%, for example, the k value can be adjusted from 2 to 2.5, to avoid misjudging deviations caused by natural fluctuations or reasonable emissions.

[0037] Furthermore, real-time verification and calibration are performed by incorporating newly generated valid data into the historical database each quarter, making the database larger and more convincing. This updates the historical dataset, recalculates relative thresholds, and adapts to water quality change trends. If data deviations are too large, they are rationalized by correcting or removing data before recalculating, thereby improving the accuracy and effectiveness of this assessment method.

[0038] Understandably, the aforementioned filtering of real-time monitoring data using historical valid data effectively improves the quality of the data used in the assessment. It eliminates abnormal data caused by sensor malfunctions, transmission interference, and extreme, accidental factors, increasing the valid data rate to over 90%. This ensures the accuracy of the assessment, as the filtered data conforms to the actual patterns of the water environment, avoiding false positives or false negatives caused by outliers, such as misjudging water quality as meeting or exceeding standards. Simultaneously, the stratified filtering algorithm has low computational complexity, with a single data set processing time of ≤1 second, not affecting the timeliness of the real-time assessment. Furthermore, it enhances trend stability, as the data considers historical trends, providing a reliable foundation for subsequent prediction of water environment change trends and risk warnings.

[0039] After filtering data through threshold ranges, further verification is performed by comparing with historical data to further screen reliable data. Moving averages or exponential smoothing methods are used to analyze the continuity of real-time data time series, eliminating abrupt changes. Then, the Pearson correlation coefficient algorithm is used to compare the similarity between real-time data and valid historical data from the same period, retaining data with a correlation coefficient ≥ 0.7. The Pearson correlation coefficient calculates the similarity between continuous time series segments of a real-time indicator and continuous time series segments of valid historical data from the same period, rather than data at a single moment. Matching the trend of the same period verifies the reliability of real-time data. For example, the real-time end takes continuous real-time time series segments of a certain indicator (such as COD) (e.g., the last 7 days, 30 monitoring points, denoted as sequence X), which must be valid data without anomalies after screening. The historical end extracts valid time series segments of the same period and hydrological conditions from historical data (e.g., 7 days from the same period last year, also during the dry season, denoted as sequence Y), with the same length as X. The Pearson correlation coefficient is calculated based on sequences X and Y to perform data screening.

[0040] Understandably, interpolation should not be performed immediately after each rejection. Instead, the decision should be made based on the need for data continuity in subsequent processing. Missing markers can be temporarily retained in intermediate screening steps, and interpolation should be performed only at key nodes (before time-series continuous analysis or final model input) to ensure sequence integrity and reduce the accumulation of errors caused by multiple interpolations.

[0041] After completing the above screening, for any minor anomalies that may remain after screening, such as data whose fluctuations slightly exceed the threshold but were not removed, interpolation methods are used for correction, such as linear interpolation and adjacent mean filling, to avoid the influence of individual extreme values. Duplicate data is simultaneously deleted, such as data collected repeatedly at the same timestamp, to ensure data uniqueness. Meanwhile, water environment indicators have large differences in dimensions, such as pH 6-9, COD 0-100 mg / L, and ammonia nitrogen 0-10 mg / L, requiring standardized dimensions. This embodiment uses Z-score standardization, i.e., (data - mean) / standard deviation, to ensure all indicators are within the same numerical range.

[0042] Furthermore, the data is reorganized according to timestamps and multiple indicators. Time is used as the horizontal axis (e.g., 10-minute / 1-hour intervals to match real-time monitoring frequency), and the vertical axis represents each core indicator (pH, COD, ammonia nitrogen, dissolved oxygen, etc.), forming a two-dimensional data table of time and indicators. If a timestamp is missing, such as due to transmission delays or missing values ​​resulting from outlier removal, it is filled with trend interpolation from the previous three timestamps to ensure temporal continuity. The integrity and stationarity of the time series data are checked. For integrity, continuous timestamp coverage is ≥98%. For stationarity, the ADF test is used to avoid model oscillations caused by non-stationary data. If the data is non-stationary, such as water quality indicators fluctuating drastically with the seasons, it is converted into a stationary sequence through differencing, such as first-order differencing, to meet the input data requirements of the cross-sectional fully mixed model. Finally, it is input into the cross-sectional fully mixed model in time series format.

[0043] It is worth noting that this embodiment filters data from the most recent time period, generates a time series, and then calculates the Nemerow Composite Index at multiple points within that time series. The real-time aspect is reflected in the real-time data acquisition, the generation of a time series from the most recent time period, and the calculation of the Nemerow Composite Index at multiple points within that time period.

[0044] S2. Calculate the third information based on the first information from the previous moment, the second information from the current moment, and the preset cross-sectional full-mix model. The third information is the predicted concentration of multiple water environment indicators at the first cross-section after the inflow of passenger water at the current moment. Specifically, the cross-sectional fully hybrid model is represented as follows: (1) in, For prediction The first cross section at time The day after the inflow of passenger water The concentration of each water environment indicator The first cross-section to be monitored exist Cross-sectional flow rate at any given time The first cross-section to be monitored exist The first moment The concentration of each water environment indicator The second cross-section to be monitored exist Cross-sectional flow rate at any given time The second cross-section to be monitored exist The first moment The concentration of each water environment indicator.

[0045] Understandably, in this embodiment, the mixing of various components in the water body during the process of the river flowing from one cross-section to the next is mainly driven by mixing. Assuming no other hydrogeochemical processes occur, the mixing behavior of the water body at the real-time cross-section and the real-time upstream cross-section is considered to construct a full mixing model for the cross-section. This is because data from a single cross-section is susceptible to localized, instantaneous interference (such as temporary sewage discharge along the bank or localized sediment suspension). Furthermore, considering only a single cross-section is a static, single-point monitoring method and cannot match the real-time nature of river flow. Mixing behavior, however, can integrate the "average effect" of upstream and downstream water bodies, eliminate accidental local factors, and reflect the overall stable water quality of the cross-section. Therefore, the cross-sectional full mixing model established in this embodiment considers mixing behavior. The mixing model links upstream and downstream real-time data to dynamically simulate the entire process of water flow into mixing, which is in line with the core requirement of "dynamic tracking" for real-time assessment. Moreover, single cross-sectional data can only reflect "the state at this point at this moment" and cannot establish a causal relationship between upstream and downstream water quality. The mixing model in this embodiment clarifies the contribution ratio of upstream water to downstream water, which is convenient for tracing the source of pollution (when the concentration of downstream water is abnormal, the mixing ratio can be used to infer whether it comes from upstream input).

[0046] Furthermore, it can be understood that the formula of the cross-sectional full mixing model in this embodiment is derived based on the law of conservation of pollutant mass. Its core is to simulate the dynamic mixing process of the existing water quality at the previous moment and the real-time upstream inflow. The calculation results are not used to replace the measured values ​​of the cross-section to be evaluated, but rather to correct the local instantaneous deviations of the measured values, outputting an effective concentration that more closely reflects the overall water quality of the cross-section. This is more suitable for the core needs of real-time evaluation than a single measured value. Essentially, it is a dynamic mixing mass balance, specifically manifested as follows: Rivers are continuous flow systems with cross-sections Water quality is constantly being updated dynamically, both in terms of existing and incremental levels. The formula integrates historical data (i.e.,...) time Measured values ​​of the cross-section) and real-time increments (i.e. upstream of time (The inflow of water from the cross-section) precisely simulates this dynamic process. It adapts to the premise of "mixing only," which is reflected in the fact that the formula does not introduce any parameters such as chemical reactions, degradation, or adsorption; it only considers the product (mass) of flow rate and concentration, perfectly matching the setting of no other hydrogeochemical effects, and the derivation logic has no extra redundancy. The core is "proportional weighted average," meaning that the concentration after mixing is essentially a flow-weighted average of the concentrations from the two source water bodies. The larger the flow rate of the water body, the higher its contribution weight to the mixed concentration, thus perfectly conforming to actual mixing patterns.

[0047] It is worth noting that the output of the cross-sectional full mixing model in this embodiment reflects the overall stable concentration of the cross-section, which can correct for local deviations in measured values. Real-time measured values ​​are local data at a certain point or instant in the cross-section, which may be affected by accidental factors such as temporary sewage discharge on the bank, local sediment suspension, and single-point interference from sensors, and cannot represent the overall water quality of the cross-section. The calculation results integrate the average effects of upstream and downstream water bodies through flow weighting, eliminating local instantaneous interference, and reflecting the average stable concentration of the entire cross-section, which is closer to the true state of the watershed water quality. At the same time, it reflects the causal relationship of the mixing process, avoids static assessment bias, and the calculation results clearly show the effect of upstream water flow on the cross-section. The contribution of water quality explains why the concentration at the cross-section is at that value. Compared to isolated measured values, the calculated results are more traceable and logical, providing a complete logical chain of concentration source and mixing result for real-time assessment. Furthermore, it is adapted to real-time dynamic assessment, balancing timeliness and accuracy. Specifically, the formula only requires two sets of nearest neighbor data from the previous and current moments; the calculation process is simple, without complex model iterations, and has low processing latency (completed in seconds), perfectly matching the rapid output requirements of real-time assessment. The calculated result is a dynamically updated mixed concentration, rather than a static measured value, which can track the continuity of river flow and avoid the drawbacks of the static and one-sided nature of a single measured value.

[0048] Furthermore, it should be noted that the accuracy of measured values ​​refers to the authenticity of a single-point physical quantity, such as a sensor indeed measuring a concentration of X mg / L at that point. However, the assessment requires the authenticity of the overall water quality of the cross-section. Relying solely on single-point measurements may lead to deviations from the overall state due to local interference. In the cross-sectional full mixing model of this embodiment, the accuracy of the calculation results is reflected in the authenticity of the overall pattern. By integrating multi-source data through mass conservation, single-point deviations are corrected. Although there may be differences from the measured values, it better reflects the actual level of water quality at the cross-section and is more valuable for assessment decisions. For example, when determining whether water quality meets standards, the overall average concentration is more convincing than the instantaneous concentration at a single point. Measured values ​​cannot reflect the dynamics of mixing. The measured value at the current moment only reflects the state of a certain point in the cross-section at this moment, but this state should be the result of the previous moment's stock and the real-time upstream influx of mixed data. Ignoring the data from the previous moment makes it impossible to separate the influence of local interference from the true mixing. The data from the previous moment is the basic stock of water quality at the cross-section. The water body will not be in The moment suddenly disappears, the previous moment cross-section The water quality is the current basis for mixing; however, this data is lacking, so only the upstream inflow, i.e., the cross-section at the previous moment, is used. The water quality and current measured values ​​cannot distinguish between the base water quality and the influence of the inflow, leading to a distortion in the mixing ratio calculation. Therefore, the core purpose of the cross-sectional full mixing model in this embodiment is to correct rather than replace. It is a prediction recalculated based on the data from the previous moment. It uses dynamic mixing logic to correct local deviations in the measured values ​​and ultimately outputs the optimized results of the measured values ​​and the mixing logic, rather than denying the value of the measured data. The measured data is one of the core inputs of the formula.

[0049] S3. Calculate the Nemerow Comprehensive Pollution Index based on the third information and the improved Nemerow Index method to obtain the Nemerow Comprehensive Pollution Index; Specifically, the concentration of water quality indicators calculated using the fully mixed model. Data on standard values ​​in the "Surface Water Environmental Quality Standard GB3838-2002" An improved single-factor evaluation method and an improved Nemerow comprehensive evaluation method were used to conduct a comprehensive water environment evaluation, as specifically expressed as follows: (2) (3) (4) (5) in, For prediction The first cross section at time The day after the inflow of passenger water The concentration of each water environment indicator For the first The standard values ​​corresponding to each water environment indicator The number of water environment indicators. for Cross-section of time The The single-factor index corresponding to each water environment indicator for Multiple single-factor indices at time points The average value, for Single-factor indices corresponding to multiple water environment indicators at different times The maximum value, for Cross-section of time The Nemerow Comprehensive Pollution Index.

[0050] Understandably, the Nemerow index method is one of the most commonly used methods for calculating the comprehensive pollution index of rivers both domestically and internationally. For any evaluation area, by calculating its comprehensive index and comparing it with the corresponding classification standards, the overall environmental quality conditions of an environmental factor can be assessed. The comprehensive pollution index method uses the "Surface Water Environmental Quality Standard" (GB 3838-2002) as the evaluation standard. It first calculates the single-factor pollution index, then calculates the comprehensive pollution index, and finally determines the degree of water pollution. The traditional Nemerow index method is a weighted multi-factor environmental quality assessment method that emphasizes the maximum value, highlighting the impact and role of the pollutant with the highest pollution index on environmental quality, reflecting the degree of water pollution.

[0051] The improved Nemerow pollution index method, based on the traditional Nemerow pollution index method, introduces the concept of weight for each pollutant factor, considering the weight of each evaluation factor in groundwater quality. This improves the evaluation of certain toxicological indicators with low concentrations but high hazard to some extent, making it suitable for detecting toxic and harmful substances in groundwater. It provides a more objective and comprehensive evaluation of groundwater quality than the traditional Nemerow pollution index method. The improved Nemerow index method incorporates the weight of each pollutant factor in the overall impact on water bodies, resulting in more reliable calculation results.

[0052] Unlike the single-factor water quality labeling index method, which was one of the earliest methods applied to water quality assessment and is simple to calculate but cumbersome, requiring comparison of the measured values ​​of each factor with standard values ​​to select the worst parameter to represent the overall water quality, this method has limitations and cannot comprehensively reflect the overall water quality situation. Therefore, it is often used in conjunction with other methods to assess water pollution. The single-factor evaluation method compares the measured values ​​of each monitoring item in the water body with the standard values ​​of water environmental quality items, selecting the category of the worst single indicator as the comprehensive water quality category of the water body.

[0053] Unlike the traditional Nemerow index, which is calculated solely by taking the square root of the sum of the squares of the average and maximum indices without distinguishing the importance of indicators, the improved method assigns weights to amplify the impact of high-hazard, high-contribution indicators in both single-factor and composite indices (e.g., for low-concentration, highly toxic mercury, a higher weight will significantly increase the composite index). Essentially, it quantifies the actual proportion of each factor's impact on water quality using weights, thus clearly reflecting consideration of the weighting issue.

[0054] Furthermore, it can be understood that formula (2) in this embodiment is an optimization of the data basis of the traditional single-factor index method, without complex mathematical derivation. The core is to adapt to the dynamic characteristics of the fully mixed model: the overall concentration of the cross-section after mixing calculated by the fully mixed model. This method replaces the traditional method's single measured value, solving the problem of single-point data being easily affected by local interference. It uses the actual concentration after mixing divided by the GB3838-2002 standard value. This unifies indicators with different dimensions (such as COD and ammonia nitrogen) into a dimensionless index, allowing for a more intuitive quantification of the pollution level of a single indicator (e.g., <1 indicates that the standard is not exceeded. >1 indicates exceeding the standard). The design logic stems from the relative deviation between concentration and standard, essentially optimizing the fit of the traditional single-factor index method to the previously mentioned fully mixed model, with the input... Overall concentration of the cross-section after mixing It is a mixed concentration that corrects for local biases, rather than a single measured value, so that the single-factor index can better reflect the true risk of exceeding the standard and avoid misjudgment caused by local data.

[0055] Furthermore, it can be understood that formula (3) in this embodiment reflects the cross-section all The overall average pollution level of each water quality indicator. The index of a single indicator cannot represent the overall water quality of the section (e.g., some indicators meet the standards while others exceed them). The pollution level of all indicators is integrated by arithmetic average to avoid generalization. The average result is calculated based on the single-factor index of real-time mixed concentration, which is closer to the overall stable pollution state of the section than the local average of static measured values.

[0056] Furthermore, it can be understood that formula (4) in this embodiment captures the cross-section. The pollution risk of the worst-case indicator is crucial for water quality assessment. Even if most indicators meet the standards, a single severely exceeding indicator (such as heavy metal contamination) can still determine the water quality grade. The bottleneck effect in water quality safety is reflected in the following: the worst-case indicator directly reflects the core pollution problems and ecological risks of a section and must be given priority; the maximum index of all indicators at the current moment should be extracted in real time to meet the dynamic risk capture needs of real-time assessment and quickly locate key exceeding indicators.

[0057] Furthermore, it can be understood that the derivation logic of formula (5) in this embodiment is based on the weighted integration of the square root of the average and the square root of the mean. Essentially, it is a secondary weighting of the average and worst-case levels, highlighting the influence of the worst-case indicator. This is reflected in the fact that the square operation amplifies the weight of extreme values. Specifically, the analysis of formula (5) is as follows: The first step is to analyze the average index. and maximum index Square them separately to strengthen the difference weights between the two (the largest exponent with the most severe over-limit will have a higher weight after being squared).

[0058] The second step is to sum the results and then take the average to balance the overall and local effects.

[0059] The third step is to take the square root and restore it to a comprehensive index of the same magnitude as the single-factor index, ensuring the rationality of the grading standard.

[0060] Understandably, formula (5) is simple to calculate, requiring only two core parameters: average and maximum. This adapts to the rapid computational needs of real-time evaluation and avoids delays caused by complex models. The index value is precisely matched with the grading standards (0.8, 2.50, etc.), amplifying the impact of the worst-performing indicator through squaring, while not ignoring the overall average level, thus solving the evaluation loophole where the average meets the standard but individual indicators seriously exceed the standard.

[0061] S4. Based on the Nemerow Comprehensive Pollution Index, conduct a water environment assessment of the first cross section to obtain the assessment results.

[0062] Specifically, this embodiment uses the Nemerow Comprehensive Pollution Index at the current moment to perform a quantitative classification of water quality assessment, which is intuitive and easy to understand, transforming abstract concentration data into a specific comprehensive index. The system provides clear grading for excellent, good, relatively good, poor, and very poor assessments, with real-time results that can be directly applied (e.g., for early warning and decision-making). In this embodiment, Water quality <0.8 is considered excellent, 0.8 ≤ A water quality index <2.50 indicates good water quality, while an index ≤2.50 indicates good water quality. A water quality of <4.25 is considered good, while 4.25 ≤ <7.20 indicates poor water quality. ≥7.2 Water quality is extremely poor.

[0063] It is understandable that the grading thresholds of 0.8, 2.50, 4.25, and 7.20 are not subjectively set, but are derived from three dimensions: the GB3838-2002 standard, historical watershed data statistics, and an improved Nemerow index calculation logic. This approach aligns with objective water quality laws and adapts to real-time management needs, combining persuasiveness and practicality. Details are as follows: I. Core Basis for Threshold Setting. Anchored to the threshold of GB3838-2002 standard: Using Class III water quality standards as the benchmark (patented core evaluation standard), the standard values ​​of each indicator are substituted into the improved Nemerow formula to calculate the index range corresponding to compliance, slight exceedance, and severe exceedance. For example, when all indicators meet the standard ( When ≤1), the composite index ≤0.8, therefore the excellent threshold is set to 0.8; when 1-2 common indicators slightly exceed the standard ( When ≈1.5), the calculation yields ≈2.50, which serves as the dividing line between good and relatively good, ensuring that the grading is strongly aligned with national water quality standards.

[0064] Based on historical watershed data statistical calibration: Valid monitoring data (outliers removed) from the target watershed over the past three years were collected, and the comprehensive index distribution under different pollution states was calculated. Statistical analysis revealed that 95% of the data indicating "compliance with standards and low ecological risk" was concentrated in... <0.8, 80% of the data for mild pollution that does not require emergency treatment fall in the range of 0.8-2.50, while the data for severe pollution that requires early warning and treatment are mostly ≥7.2. The distribution pattern of the threshold range is highly consistent with the actual water quality conditions.

[0065] To adapt to the calculation characteristics of the Nemerow formula: the formula uses the logic of summing the squares of the average exponent and the squares of the maximum exponent and then taking the square root, which amplifies the impact of the exceeding indicators. Combined with simulation calculations of common river water environment pollution scenarios, such as single indicator exceeding the standard and multi-indicator synergistic pollution, three nodes of 2.50, 4.25, and 7.20 are determined. This not only distinguishes between slight exceeding the standard (good), moderate exceeding the standard (poor), and severe exceeding the standard (very poor), but also avoids management confusion caused by overly dense classification, adapting to the needs of real-time early warning and governance priority ranking.

[0066] II. Explanation of the effect of the threshold. Precisely depicting the gradual change in water quality: Traditional classification methods, such as clean-lightly polluted-heavily polluted, struggle to differentiate between levels that meet standards but pose potential risks, and levels that exceed standards but are manageable. This threshold, however, subdivides the compliant range into excellent and good, and the exceeding range into relatively good, poor, and very poor. It accurately reflects the gradual change in water quality from excellent to poor, providing a basis for refined management. For example, good water quality requires routine monitoring, while poor water quality requires enhanced control.

[0067] Adapting to real-time deployment scenarios: Thresholds are directly linked to actual management needs. <0.8 indicates ecological safety, requiring no intervention; Routine monitoring is required when the pH is between 0.8 and 2.50; pay attention to any changes. For values ​​between 2.50 and 4.25, more frequent monitoring is needed to investigate the source of pollution. For areas between April 25th and July 20th, an early warning system should be activated, and localized remediation measures should be implemented. ≥7.2 An emergency response is required, with comprehensive interception of pollution, so that real-time assessment results can be directly translated into management actions, avoiding a disconnect between evaluation and application.

[0068] It is understandable that the real-time water environment assessment method based on the total mixed model in this embodiment is comprehensive yet highlights key aspects. It reflects overall water quality through an average index while capturing critical exceedance risks through a maximum index, avoiding the evaluation bias of averaging masking severe pollution or using a single indicator to negate the overall picture. Simultaneously, it is adapted for real-time dynamic assessment, relying on real-time calculated mixed concentrations for its calculations. The evaluation results can be based on the Nemeiro composite index at continuous time points. It dynamically tracks changes in water quality after river mixing. The calculation process is simple, with low latency, meeting the core requirement of rapid output for real-time assessment. Furthermore, it adheres to standard guidelines, being designed entirely based on the standard values ​​of GB3838-2002, and the evaluation results are compatible with the national water quality classification system, possessing authority and practicality.

[0069] Compared to the traditional Nemerow index method, the improved Nemerow comprehensive evaluation method in this embodiment solves the problems of susceptibility to random errors, static evaluation lag, and coarse classification in traditional methods by means of multi-sample mean of single-factor indices, systematic selection of the maximum factor, real-time dynamic logic of the comprehensive index, and refinement and coherence of the classification standards. It is more suitable for accurate and dynamic evaluation of complex aquatic environments. Specific advantages are as follows: 1. Reduced interference from random errors, resulting in a more reliable data foundation. Existing Nemerow comprehensive evaluation methods often calculate single-factor indices based on single measured values, making them susceptible to random factors such as instantaneous contamination and sampling operation biases, leading to significant fluctuations in single-factor indices. In the improved method, both the average and maximum single-factor indices are calculated based on multi-sample data from real-time cross-sections. By replacing single-sample data with the mean and maximum values ​​from multiple samplings, the true contamination level of the cross-section can be reflected more stably, reducing the interference of random errors on subsequent comprehensive index calculations and providing stronger data support for the evaluation results.

[0070] 2. The comprehensive index logic is more reasonable, balancing the impact of average and extreme pollution. While the existing Nemerow comprehensive index calculation formula considers both average and maximum pollution, it does not clearly define data timeliness and cross-sectional correlation. This embodiment emphasizes real-time dynamism; the basic data for the comprehensive index calculation are all real-time data, better reflecting the actual scenario of river pollution dynamically changing over time. Furthermore, it strengthens cross-sectional targeting by coupling flow-concentration data between the cross-section to be evaluated and its upstream cross-sections. This allows the comprehensive index to not only reflect the pollution of a single cross-section but also implicitly include pollution transfer information from upstream to downstream mixing processes, improving the spatial correlation of the evaluation.

[0071] 3. The evaluation and grading are more refined and better suited to actual management needs. The existing Nemerow evaluation method is mostly graded into four levels: "clean - slightly polluted - moderately polluted - heavily polluted". The grading thresholds are relatively coarse (e.g., PN<1 is clean, PN≥3 is heavily polluted), which makes it difficult to distinguish subtle changes in water quality that meet the standards but have differences or exceed the standards but to varying degrees. The improved method classifies water quality into five levels with more refined threshold ranges, enabling a more precise depiction of the gradual change in water quality from excellent to poor. For example, the distinction between excellent and good can guide protection strategies for ecologically sensitive areas, while the refinement of poor and very poor provides a clearer basis for prioritizing pollution control, enhancing the practical value of the evaluation results for water environment management.

[0072] 4. Deep coupling with fully mixed models enhances the systematic nature of the evaluation. Existing Nemerow methods are usually independent of water quality prediction models and are based solely on measured data, making it difficult to reflect the spatial migration and transformation of pollution. The improved method directly uses the calculation results of the real-time full-mixing model at the cross-section as input, forming a closed loop between the pollution mixing process and the comprehensive evaluation. This coupling makes the evaluation not only a description of the current situation, but also implicitly includes a quantitative analysis of the impact on upstream water flow, enabling a more systematic identification of pollution sources and providing more comprehensive information for targeted pollution control.

[0073] This embodiment is... Time section The relevant data from the water quality category determination process are illustrated in the table below:

[0074] The diagram shows two adjacent time points. and upstream and downstream sections and Real-time values ​​of flow rate, ammonia nitrogen, total nitrogen, total phosphorus, and nitrate nitrogen were displayed. Time section The concentrations of each index calculated by the cross-sectional full mixture model, and the single-factor indices and predicted cross-sections calculated based on the cross-sectional full mixture model results. Water quality category.

[0075] Example 2: Unlike Example 1, this example provides a real-time water environment assessment system based on a fully mixed model, including: The information acquisition module is used to acquire first information and second information, wherein the first information and the second information are real-time cross-sectional data of the first cross-section and the second cross-section, respectively. The cross-sectional data includes cross-sectional flow and the concentration of multiple water environment indicators of the cross-section. The second cross-section is the cross-section upstream of the first cross-section. The first processing module is used to calculate the third information based on the first information at the previous moment, the second information at the current moment, and the preset cross-sectional full-mix model. The third information is the predicted concentration of multiple water environment indicators at the first cross-section at the current moment after the inflow of passenger water. The second processing module is used to calculate the Nemerow Comprehensive Pollution Index based on the third information and the improved Nemerow Index method, thereby obtaining the Nemerow Comprehensive Pollution Index. The third processing module is used to conduct a water environment assessment of the first cross section based on the Nemerow Comprehensive Pollution Index and obtain the assessment results.

[0076] Example 3: Unlike Examples 1 and 2, this example provides a method for adjusting the screening threshold range of indicators in different scenarios.

[0077] Scenario 1 refers to the period within 12 hours after rainfall, especially moderate to heavy rain; The screening threshold ranges for the indicators suspended solids (SS), turbidity, chemical oxygen demand (COD), ammonia nitrogen, and total phosphorus (TP) have been adjusted as follows: Original threshold and adjusted threshold: Normal distribution indicators (COD, ammonia nitrogen, TP): k value from 2→2.5 (relaxed by 25%); Skewed distribution index (SS, turbidity): IQR range from Q1-1.5IQR~Q3+1.5IQR→Q1-2.0IQR~Q3+2.0IQR (widened by 33%). It is understandable that rainfall runoff washes away surface sediment, organic matter, and nutrients from the soil, leading to a natural increase in physical indicators such as suspended solids (SS) and turbidity, as well as pollution indicators such as COD and ammonia nitrogen. This is a "natural, non-polluting fluctuation," not water quality deterioration. Using the original thresholds would misjudge these reasonable fluctuations as abnormal data and exclude them, resulting in distorted assessment results. By selectively relaxing the thresholds for indicators significantly affected by runoff, we can preserve accurate natural fluctuation data without compromising the accuracy of indicators such as pH and dissolved oxygen (DO), which are minimally affected by rainfall.

[0078] In scenario 2, the area is within 500 meters downstream of the sewage outlet, which is applicable to municipal sewage outlets and compliant sewage outlets in industrial parks. Specifically, the indicators for ammonia nitrogen, total nitrogen (TN), total phosphorus (TP), and biochemical oxygen demand (BOD) are... The corresponding filtering threshold range is adjusted as follows: Original threshold and adjusted threshold: Normally distributed indices (ammonia nitrogen, TN, TP, ...) ): k value from 2 → 2.2 (relaxed by 10%); Unbiased distribution indicators (these indicators fluctuate relatively smoothly near compliant sewage outlets). It is understandable that compliant discharge outlets will discharge treated wastewater, which, although meeting emission standards, may still cause slight increases in nitrogen, phosphorus, and organic matter concentrations in downstream local water bodies. These are reasonable and controllable fluctuations. Strictly adhering to the original thresholds might misclassify these compliant but slightly above-background values ​​as abnormal. Targeted relaxation of the thresholds for nitrogen, phosphorus, and organic matter (by only 10% to avoid over-inclusiveness) is appropriate for the water quality characteristics near the discharge outlets while effectively identifying actual pollution exceeding standards (e.g., k=2.2 can still detect abnormal increases beyond the compliant emission range), thus balancing the needs of compliant emission inclusion with pollution control.

[0079] Scenario 3 is a cross-section of an industrial zone, involving the area surrounding enterprises that emit heavy metals and VOCs; Specifically, for the indicators mercury (Hg), cadmium (Cd), and chromium (… The screening threshold ranges for volatile organic compounds (VOCs) have been adjusted as follows: Original threshold and adjusted threshold: Normal distribution indicators (heavy metals, VOCs): k value changed from 2 to 1.8 (tightened by 10%); for strict scenarios (original k=3), it was adjusted to 2.7 (tightened by 10%). Unbiased distribution indicators (these indicators are highly toxic and have small concentration fluctuations); It is understandable that industrial zones pose a potential risk of leaks of toxic and hazardous substances such as heavy metals and VOCs. These indicators, though at low concentrations, pose significant risks; even minor anomalies can trigger ecological hazards. Tightening their thresholds can more sensitively detect slight fluctuations in these highly hazardous indicators, providing early warnings of potential leaks and preventing the omission of critical pollution signals due to overly broad thresholds. Conventional indicators (such as COD and SS) do not exhibit specific fluctuation patterns in industrial zones; therefore, maintaining their original thresholds ensures a comprehensive assessment.

Claims

1. A real-time water environment assessment method based on a fully mixed model, characterized in that, include: First information and second information are obtained, the first information and the second information are real-time cross-sectional data of the first cross-section and the second cross-section, respectively. The cross-sectional data includes cross-sectional flow and the concentration of multiple water environment indicators of the cross-section. The second cross-section is the cross-section upstream of the first cross-section. The third information is calculated based on the first information from the previous moment, the second information from the current moment, and the preset cross-sectional full-mix model. The third information is the predicted concentration of multiple water environment indicators at the first cross-section after the inflow of foreign water at the current moment. The Nemerow Comprehensive Pollution Index is calculated based on the third information and the improved Nemerow index method to obtain the Nemerow Comprehensive Pollution Index; The water environment assessment of the first cross section was conducted based on the Nemerow Integrated Pollution Index, and the assessment results were obtained.

2. The real-time water environment assessment method based on a fully mixed model according to claim 1, characterized in that... After obtaining the first information and the second information, the process further includes: Obtain a first time series and a second time series, wherein the first time series includes the first information at multiple first moments within a first time period, and the second time series includes the second information at multiple moments within the first time period; Obtain fourth information, which includes cross-sectional data from previous years that are concurrent with and under the same working conditions as the third cross-section in the first time period, wherein the third cross-section is the first cross-section and the second cross-section. Based on the fourth information, the third time series is preprocessed to obtain the preprocessed third time series, which includes the first time series and the second time series. The third information at multiple first time points is calculated based on the preprocessed third time series to obtain the third information at multiple first time points; The water environment assessment of the first cross section during the first time period is performed based on the third information at multiple first moments.

3. The real-time water environment assessment method based on a fully mixed model according to claim 2, characterized in that... The preprocessing of the third time series based on the fourth information includes: Data with excessive missing values ​​in the third time series were removed based on the missing value detection algorithm. The first data in the third time series is removed based on multiple preset threshold intervals. Each threshold interval corresponds to a water environment indicator, and the first data is the water environment indicator that exceeds its corresponding threshold interval. The continuity of the third time series is analyzed using the moving average method or exponential smoothing method, and abrupt data is removed. The third and fourth time series were calculated based on the Pearson correlation coefficient algorithm and the fourth information. The fourth time series The third time series is constructed from historical data of the same period and working conditions, and data within the third time series is removed based on the Pearson correlation coefficient. The third time series is filled by interpolation based on the changing trend of the data within the third time series after data removal.

4. The real-time water environment assessment method based on a fully mixed model according to claim 3, characterized in that... The threshold range includes absolute thresholds and relative thresholds; The absolute threshold is set based on the surface water environmental quality standard; The relative threshold is calculated based on the cross-sectional data of the cross-section corresponding to the relative threshold in the fourth information, which are from the same period and under the same working conditions over the years. The relative threshold is based on the first cross-section. .

5. The real-time water environment assessment method based on a fully mixed model according to claim 1, characterized in that... The cross-sectional fully hybrid model is represented as follows: ； in, For prediction The first cross section at time After the inflow of passenger water The concentration of each water environment indicator The first cross-section to be monitored exist Cross-sectional flow rate at any given time The first cross-section to be monitored exist The first moment The concentration of each water environment indicator The second cross-section to be monitored exist Cross-sectional flow rate at any given time The second cross-section to be monitored exist The first moment The concentration of each water environment indicator.

6. The real-time water environment assessment method based on a fully mixed model according to claim 1, characterized in that... The calculation of the Nemerow comprehensive pollution index based on the third information and the improved Nemerow index method includes: Obtain the standard value corresponding to each of the aforementioned water environment indicators; The single-factor index of the first water environment index is calculated based on the concentration of the first water environment index predicted at the current moment and the standard value corresponding to the first water environment index. Calculate the single-factor index corresponding to each of the water environment indicators to obtain multiple single-factor indices; The Nemerow composite pollution index is calculated based on the average and maximum values ​​of multiple single-factor indices and the Nemerow index method.

7. The real-time water environment assessment method based on a fully mixed model according to claim 6, characterized in that... The Nemerow composite pollution index, calculated based on the average and maximum values ​​of multiple single-factor indices and the Nemerow index method, can be expressed as: ； in, for The first section at time 1 The Nemerow Composite Pollution Index, for The first section at time 1 The The single-factor indices corresponding to the aforementioned water environment indicators. for Multiple single-factor indices at time points The average value, for Single-factor indices corresponding to multiple water environment indicators at different times The maximum value.

8. The real-time water environment assessment method based on a fully mixed model according to claim 7, characterized in that... The water environment assessment of the first cross-section based on the Nemerow Integrated Pollution Index includes: The water quality level of the first cross section is determined based on the preset water quality classification standards and the Nemerow Comprehensive Pollution Index. The water quality level includes excellent, good, relatively good, poor, and very poor.

9. A real-time water environment assessment system based on a fully mixed model, characterized in that... ,include: The information acquisition module is used to acquire first information and second information, which are respectively real-time cross-sectional data of the first cross-section and the second cross-section. The cross-sectional data includes cross-sectional flow and the concentration of multiple water environment indicators of the cross-section. The second cross-section is the cross-section upstream of the first cross-section. The first processing module is used to calculate the third information based on the first information at the previous moment, the second information at the current moment, and the preset cross-sectional full-mix model. The third information is the predicted concentration of multiple water environment indicators at the first cross-section at the current moment after the inflow of passenger water. The second processing module is used to calculate the Nemerow Comprehensive Pollution Index based on the third information and the improved Nemerow Index method, thereby obtaining the Nemerow Comprehensive Pollution Index. The third processing module is used to conduct a water environment assessment of the first cross section based on the Nemerow Comprehensive Pollution Index and obtain the assessment results.

Images