Regional water environment risk assessment system considering water ecology and resident life influence

Abstract

The present application relates to the technical field of water quality monitoring, in particular to a regional water environment risk assessment system considering the influence of water ecology and resident life, which comprises a water temperature and flow velocity module, an ecological analysis module, a pollution analysis module, a risk assessment module and an optimization assessment module.In the present application, water temperature, flow velocity and resident water consumption are collected and subjected to periodic characteristic separation, the synchronous change rate of temperature and water consumption behavior is analyzed to calculate the thermal disturbance intensity, the water ecological thermal response characteristic set is generated, the concentrations of dissolved oxygen, dissolved nitrogen, dissolved phosphorus and chlorophyll a are collected for correlation analysis, the ecological energy metabolism coordination coefficient is calculated to generate the ecological stability sequence, the residual chlorine and soluble organic carbon concentrations in the water supply path are detected to calculate the pollutant concentration fluctuation rate, the thermal disturbance intensity is combined to generate the pollution activation sequence through probability classification, the pollution migration rate is deduced and the risk time sequence data is fitted to generate the regional water environment risk optimization result.

Description

Technical Field

[0001] This invention relates to the field of water quality monitoring technology, and in particular to a regional water environment risk assessment system that takes into account the impact of aquatic ecology and residents' lives. Background Technology

[0002] The field of water quality monitoring technology encompasses the acquisition, analysis, and assessment of physicochemical and biological indicators in water bodies to identify the status and trends of the water environment. Its core content involves the systematic monitoring of pollutant concentrations, aquatic ecological indicators, and related environmental factors in different types of water bodies, such as rivers, lakes, groundwater, and urban water supply systems, through methods such as sensor acquisition, sample testing, and data modeling. Overall, this technology field covers monitoring data acquisition devices, water quality parameter detection methods, monitoring data analysis systems, and water environment risk assessment technologies, providing fundamental technical support for water resource protection and pollution prevention and control.

[0003] The regional water environment risk assessment system, which considers the impact on aquatic ecosystems and residents' lives, is a comprehensive risk identification and analysis system established at a regional scale for aquatic ecosystems and residents' water safety. It primarily assesses technical aspects such as regional water pollution status, aquatic ecosystem health status, and risks to domestic water use. This is achieved by establishing risk factor calculation models based on monitoring data, correlation analysis methods for aquatic ecological characteristic parameters, and risk classification rules for domestic water use, thus enabling a comprehensive determination of regional water environment risks. The system employs methods such as quantifying water quality indicators, determining the weights of aquatic ecological elements, and correlating risk factors related to domestic water use to complete the overall assessment process.

[0004] Existing technologies primarily rely on single-indicator data collection and static analysis, lacking a systematic understanding of time-series changes and the coupling relationships between multiple parameters. This results in monitoring results that lag behind in reflecting complex ecological dynamics. Current risk assessment models typically use pollutant concentration or aquatic ecosystem status as a single evaluation benchmark, neglecting the combined effects of temperature fluctuations, flow velocity changes, and residential water use behavior, making it difficult to comprehensively depict the migration process of pollution along water supply paths and within ecosystems. Most systems lack dynamic fitting and probabilistic correlation mechanisms among multiple factors during data processing, easily leading to discretization in pollution risk level determination and reducing the continuity and reliability of assessments. In regional-scale applications, existing systems are often limited by the layout of monitoring points and the frequency of data updates, failing to reflect the pollution diffusion trends of local water bodies in a timely manner, thus affecting water resource regulation and emergency response decisions. For example, in urban water supply systems, when sudden temperature increases or flow velocity changes trigger pollutant activation, existing technologies often fail to immediately identify risk growth trends, leading to delayed risk warnings and reduced response efficiency. Summary of the Invention

[0005] To address the technical problems existing in the prior art, embodiments of the present invention provide a regional water environment risk assessment system that considers the impact on aquatic ecology and residents' lives. The technical solution is as follows:

[0006] On the one hand, a regional water environment risk assessment system that considers the impact on aquatic ecology and residents' lives is provided. This system includes: The water temperature and flow rate module collects the water temperature, flow rate, and water consumption of residents in the ecological water body, separates the periodic characteristics of water temperature and flow rate changes, analyzes the synchronous change rate of temperature and water use behavior, calculates the intensity of thermal disturbance, generates a water ecological thermal response feature set, and transmits it to the ecological analysis module. The ecological analysis module, based on the water ecological thermal response feature set, collects dissolved oxygen, dissolved nitrogen, dissolved phosphorus and chlorophyll a concentration and normalizes the correlation, calculates the ecological energy metabolism synergy coefficient, generates the ecological stability sequence and transmits it to the pollution analysis module. The pollution analysis module detects the residual chlorine concentration and soluble organic carbon concentration in the water supply path based on the ecological stability sequence, calculates the pollutant concentration fluctuation rate by normalization and performs probability classification in combination with the thermal disturbance intensity, generates the residential water pollution activation sequence and transmits it to the risk assessment module. The risk assessment module calculates the probability of pollution propagation between water supply and aquatic ecology through the activation sequence of pollution in residential water use, derives the pollution migration rate sequence and pollution distribution change sequence, and analyzes the risk time series data in combination with the ecological stability sequence to generate the preliminary water environment risk assessment results. The optimization assessment module, based on the preliminary water environment risk assessment results, smooths and corrects the risk time series data, and performs time fitting and smoothing optimization on the pollution migration rate and pollution distribution changes, generating optimized water environment risk results.

[0007] As a further aspect of the present invention, the aquatic ecological thermal response feature set includes water temperature periodic characteristics, flow velocity periodic characteristics, and thermal disturbance intensity; the ecological stability sequence includes dissolved oxygen correlation, dissolved nitrogen correlation, and dissolved phosphorus correlation; the residential water pollution activation sequence includes residual chlorine concentration fluctuation rate, soluble organic carbon concentration fluctuation rate, and pollutant concentration probability distribution; the preliminary water environment risk assessment results include pollution propagation probability, pollution migration rate sequence, and pollution distribution change sequence; and the water environment risk optimization results include risk time series smoothing data, weighted correction data, and pollution migration and distribution fitting accuracy optimization.

[0008] As a further aspect of the present invention, the water temperature and flow rate module includes: The data analysis submodule collects water temperature and flow rate, records residents' water consumption meters, and performs synchronous matching of the three types of data. It arranges the water temperature, flow rate and water consumption values ​​of multiple time periods in a time series, calculates the average value and rate of change within the continuous sampling interval, and generates a synchronized sequence set of water temperature and flow rate. The feature separation submodule calls the water temperature and flow rate synchronization sequence set, extracts the fluctuation period of water temperature and flow rate, calculates the difference between the mean values ​​of the periods and filters out anomalies, analyzes the daily variation pattern of water consumption to match and correct the periodic features, extracts the correlation rate between temperature change amplitude and water consumption behavior, and generates a temperature and flow periodic feature set. The thermal response generation submodule, based on the temperature flow cycle feature set, calculates the synchronous change rate of temperature change amplitude and water use behavior correlation rate and the thermal disturbance intensity under flow velocity change, analyzes the energy distribution intensity range of thermal disturbance in multiple cycles, and generates a water ecological thermal response feature set.

[0009] As a further aspect of the present invention, the ecological analysis module includes: The element acquisition submodule collects and matches the concentrations of dissolved oxygen, dissolved nitrogen, dissolved phosphorus, and chlorophyll a, calculates the average value and rate of change within adjacent sampling periods, removes outliers, and generates a set of dissolved element concentration sequences. The feature normalization submodule, based on the dissolved element concentration sequence set, calculates multiple normalization ratios for the numerical range of multiple elements, adjusts the abnormal deviation range according to the consistency of the time series, calculates the synchronous change ratio among multiple elements, and generates a normalized feature set of ecological elements. The ecological assessment submodule calls the normalized feature set of ecological elements and the feature set of water ecological thermal response, performs correlation calculation on the ratio of energy change rate to element concentration in the two feature sequences, calculates the metabolic synergy ratio among multiple elements, analyzes the ecological energy metabolism synergy coefficient, and generates an ecological stability sequence.

[0010] As a further aspect of the present invention, the pollution analysis module includes: The residual chlorine monitoring submodule, based on the ecological stability sequence, detects the residual chlorine concentration in multiple sections of the water supply path, calculates the concentration change rate within adjacent sampling periods, removes abnormal drift points and corrects data discontinuities, and establishes a residual chlorine concentration change sequence. The fluctuation calculation submodule, based on the residual chlorine concentration change sequence and soluble organic carbon concentration value, performs differential normalization for the same time period, calculates the relative change amplitude between the two types of concentrations, analyzes the synchronous fluctuation intensity value of organic carbon in multiple water supply sections, and generates a pollutant concentration fluctuation rate set. The pollution classification submodule calls the pollutant concentration fluctuation set and the ecological stability sequence, performs a probability distribution comparison between pollutant fluctuation and thermal disturbance intensity, calculates the pollution trigger probability interval for multiple time periods, divides multiple risk levels according to the classification threshold, and generates a residential water pollution activation sequence.

[0011] As a further aspect of the present invention, the classification threshold is obtained by calculating the joint probability density function of pollutant volatility and thermal disturbance intensity to obtain the distribution overlap rate of the two in multiple time intervals, and then the risk boundary is set according to the quantile interval of the overlap rate.

[0012] As a further aspect of the present invention, the risk assessment module includes: The pollution propagation submodule extracts water supply flow, pollutant concentration and time series data based on the residential water pollution activation sequence, calculates the concentration transfer coefficient of pollutants between the water supply network and the aquatic ecosystem, analyzes the pollution propagation path and ratio as pollution diffusion characteristic quantity, and generates a pollution propagation probability sequence. The migration rate submodule, based on the pollution propagation probability sequence, selects velocity gradient and spatial concentration difference data within the same time period, calculates the concentration change ratio and velocity coupling coefficient of pollutants between adjacent areas, compares the concentration gradient change trend over consecutive time periods, and generates a pollution migration rate sequence. The risk analysis submodule calls the pollution migration rate sequence and combines it with the ecological stability sequence to calculate the time correlation coefficient between pollution migration rate and ecological stability, extract the risk change trend, compare it with the risk benchmark threshold range to classify the risk level, and generate the preliminary water environment risk assessment results.

[0013] As a further aspect of the present invention, the risk benchmark threshold is set by calculating the time-related distribution between the pollution propagation probability and the ecological stability sequence, extracting the overlapping density interval of the two in a continuous time series, and then setting it according to the quantile value of the distribution density.

[0014] As a further aspect of the present invention, the optimization evaluation module includes: The smoothing correction submodule, based on the preliminary water environment risk assessment results, extracts the risk level and pollution migration rate of adjacent time periods in the risk time series data, calculates the average offset of the risk level difference and migration rate difference between adjacent time series, performs a weighted average of the risk levels of continuous time series, and generates a risk smoothing sequence. The weight adjustment submodule calls the risk smoothing sequence, combines the pollution migration rate and the pollution distribution change value in the pollution distribution change sequence, calculates the response sensitivity coefficient of the pollution migration rate to risk change, performs weighted correction on the time series risk value, and generates risk weight correction result. The fitting optimization submodule extracts paired samples of pollution migration rate and pollution distribution change value based on the risk weight correction result, and performs time fitting and smoothing optimization on pollution migration rate and pollution distribution change to obtain water environment risk optimization result.

[0015] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: By simultaneously collecting and separating the periodic characteristics of water temperature, flow velocity, and residential water consumption, a dynamic data structure reflecting the coupling relationship between thermal disturbance and water use behavior was formed, enabling the monitoring process to possess sensitivity and discriminability to temporal characteristics. Through the simultaneous calculation of temperature change rate and water use behavior, a thermal disturbance intensity parameter was established, effectively revealing the immediate impact of environmental physical characteristics on ecological responses. Combined with normalized correlation analysis of dissolved oxygen, dissolved nitrogen, dissolved phosphorus, and chlorophyll a, a quantitative characterization of the synergistic relationship of ecological energy metabolism was formed, enabling ecological stability analysis to possess multi-dimensional synergistic characteristics. The volatility of pollutant concentrations, after incorporating thermal disturbance intensity, was probabilistically classified, improving the dynamic response expression of pollution activation state under different environmental conditions and enhancing the predictability of pollution propagation patterns. Fitting the pollution propagation probability and migration rate sequence enables risk assessment to possess dual characteristics of temporal evolution and spatial diffusion, ultimately forming a dynamic and quantitative judgment of water environment risk. The overall processing logic establishes a multi-dimensional interactive risk identification link by temporal correlation, coupled calculation and multi-factor fitting of physical, chemical and ecological parameters, realizing a continuous connection from water body change perception to risk quantification assessment, and achieving significant improvements in data correlation depth, risk prediction accuracy and assessment timeliness. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the system of the present invention; Figure 2 This is a schematic diagram of the system framework of the present invention; Figure 3 This is a flowchart of the water temperature and flow rate module in this invention; Figure 4 This is a flowchart of the ecological analysis module in this invention; Figure 5 This is a flowchart of the pollution analysis module in this invention; Figure 6 This is a flowchart of the risk assessment module in this invention; Figure 7 This is a flowchart of the optimization evaluation module in this invention. Detailed Implementation

[0018] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0019] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0020] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0021] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0022] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0023] This invention provides a regional water environment risk assessment system that considers the impact on aquatic ecology and residents' lives, such as... Figure 1-2 The diagram shown illustrates a regional water environment risk assessment system that considers the impacts on aquatic ecology and residents' lives. The system includes: The water temperature and flow rate module collects the water temperature, flow rate, and water consumption of residents in the ecological water body, separates the periodic characteristics of water temperature and flow rate changes, analyzes the synchronous change rate of temperature and water use behavior, calculates the intensity of thermal disturbance, generates a water ecological thermal response feature set, and transmits it to the ecological analysis module. The ecological analysis module, based on the water ecological thermal response feature set, collects dissolved oxygen, dissolved nitrogen, dissolved phosphorus and chlorophyll a concentration and normalizes the correlation, calculates the ecological energy metabolism synergy coefficient, generates the ecological stability sequence and transmits it to the pollution analysis module. The pollution analysis module detects the residual chlorine concentration and soluble organic carbon concentration in the water supply path based on the ecological stability sequence, calculates the pollutant concentration fluctuation rate by normalization and performs probability classification in combination with the intensity of thermal disturbance, generates the activation sequence of water pollution in residential water and transmits it to the risk assessment module. The risk assessment module calculates the probability of pollution transmission between water supply and aquatic ecosystem by using the activation sequence of pollution in residential water use, derives the pollution migration rate sequence and pollution distribution change sequence, and combines the ecological stability sequence to analyze the risk time series data and generate the preliminary water environment risk assessment results. The optimization assessment module, based on the initial assessment results of water environment risk, smooths and corrects the risk time series data, and optimizes the pollution migration rate and pollution distribution changes through time fitting and smoothing, generating optimized water environment risk results.

[0024] The water ecological thermal response feature set includes water temperature periodic characteristics, flow velocity periodic characteristics, and thermal disturbance intensity. The ecological stability sequence includes dissolved oxygen correlation, dissolved nitrogen correlation, and dissolved phosphorus correlation. The residential water pollution activation sequence includes residual chlorine concentration fluctuation rate, soluble organic carbon concentration fluctuation rate, and pollutant concentration probability distribution. The preliminary water environment risk assessment results include pollution propagation probability, pollution migration rate sequence, and pollution distribution change sequence. The water environment risk optimization results include risk time series smoothing data, weight correction data, and optimization of the fitting accuracy of pollution migration and distribution.

[0025] Specifically, such as Figure 2 , 3 As shown, the water temperature and flow rate module includes: The data analysis submodule collects water temperature and flow rate, records residents' water consumption meters, and performs synchronous matching of the three types of data. It arranges the water temperature, flow rate and water consumption values ​​of multiple time periods in a time series, calculates the average value and rate of change within the continuous sampling interval, and generates a synchronized sequence set of water temperature and flow rate. First, data is collected from specific monitoring nodes deployed in the residential water supply network. Water temperature data is acquired using a clamp-on PT100 platinum resistance temperature sensor, with a data refresh interval of 1 minute and an accuracy of 0.1 degrees Celsius. Flow velocity data is acquired using a pipe-type electromagnetic flow meter, also with a data refresh interval of 1 minute and an accuracy of 0.01 meters per second. Residential water consumption data comes from a concentrator connected to the smart water meters in the area. This concentrator reports the cumulative water consumption reading for that time period hourly, in cubic meters. After acquiring these three types of raw data, they are immediately synchronized and matched. Since water temperature and flow velocity data are in minute increments, while water consumption data is in hourly increments, the water consumption data needs to be downscaled. For example, for the period from 14:00 to 15:00, if the cumulative reading at 14:00 is 105.2 cubic meters and the cumulative reading at 15:00 is 105.8 cubic meters, then the total water consumption for that hour is 0.6 cubic meters. This 0.6 cubic meter water consumption is linearly allocated to 60 minute points within that hour, that is, the water consumption allocated to each minute point from 14:01 to 15:00 is [amount missing]. Cubic meters. This operation is applied to all hourly data to generate a minute-level water consumption sequence corresponding to the water temperature and flow rate data at the timestamp. Subsequently, the synchronized time series data is defined as a continuous sampling interval of 10 minutes. For the first 10-minute interval (e.g., 08:00 to 08:10), 10 water temperature values ​​within that interval are retrieved, specifically [16.2, 16.2, 16.3, 16.3, 16.3, 16.4, 16.4, 16.5, 16.5, 16.5] degrees Celsius, and the arithmetic mean is calculated. Degrees Celsius. Similarly, retrieve the corresponding 10 flow rate values ​​[0.95, 0.98, 1.02, 1.05, 1.08, 1.10, 1.09, 1.06, 1.04, 1.01] and calculate their average value. At the same time, the water consumption values ​​for 10 minutes within this interval are accumulated, i.e. cubic meters. Next, the rate of change between consecutive sampling intervals is calculated. For the second 10-minute interval (08:10 to 08:20), the average water temperature was calculated to be 16.82 degrees Celsius, and the average flow velocity was 1.152. The rate of change in water temperature is calculated by dividing the difference between the average water temperatures of the two intervals by the interval length (10 minutes), i.e. The calculation process for the rate of change of flow velocity is the same, that is... All calculated average water temperature, average flow rate, total water consumption over 10 minutes, rate of change of water temperature, and rate of change of flow rate are arranged in order of time intervals to generate a synchronized sequence set of water temperature and flow rate.

[0026] The feature separation submodule calls the water temperature and flow rate synchronization sequence set, extracts the fluctuation period of water temperature and flow rate, calculates the difference of the mean period and filters out anomalies, analyzes the daily variation pattern of water consumption to match and correct the periodic features, extracts the correlation rate between temperature change amplitude and water consumption behavior, and generates a temperature and flow periodic feature set. The generated synchronized water temperature and flow rate sequence set is used, with a focus on analyzing the time series data of average flow rate and average water temperature. First, fluctuation cycle extraction is performed. Taking the average flow rate sequence as an example, a sliding window extreme value detection method is used, with the window size set to 5 data points (i.e., 50 minutes). Point-by-point scanning is performed; when the flow rate value of a data point (e.g., 1.45 at 07:30) is greater than all values ​​of the five data points before and after it (06:40 to 07:20 and 07:40 to 08:20), that point is marked as a local maximum. The scan continues, and the next local maximum (1.51) is found at 09:10. The time span (100 minutes) between these two maximums is defined as a flow rate fluctuation cycle. The same operation is performed on the water temperature sequence. Subsequently, the cycle mean difference is calculated and outliers are filtered out. For the newly identified 07:30 to 09:10 cycle, the mean of the average flow rate for all 10-minute intervals within it is calculated, yielding 1.28. The average flow rate of the previous period (e.g., 05:00 to 07:30) is 0.90. The difference between the period averages is... To filter out outliers, a mean difference threshold was set. This threshold was determined experimentally using historical 30-day data: the standard deviation of the mean difference across all historical periods was calculated, yielding a value of 0.12. The threshold was then set to three times this standard deviation. Since the calculated value of 0.38 is greater than 0.36, this period (07:30 to 09:10) is judged as an abnormal fluctuation and is not included in subsequent calculations. Next, the daily variation pattern of water consumption is analyzed. Data on "total water consumption in 10 minutes" from the synchronous sequence set is retrieved, and 72 hours (3 days) of data are superimposed on it in 24-hour format. The average water consumption for each 10-minute interval over 3 days is calculated, and a daily variation curve is plotted. This curve shows fixed water consumption peaks between 06:30 and 08:30 and between 18:00 and 20:00. This pattern is used to match and correct the periodic characteristics. For example, a normal period identified by flow rate is from 06:10 to 08:50, and its time range overlaps with the water consumption peak (06:30 to 08:30) by more than 80%. Therefore, the start and end times of the flow rate period are forcibly corrected to 06:30 and 08:30. Finally, the correlation rate between temperature change amplitude and water consumption behavior is extracted. For each corrected period (e.g., 06:30 to 08:30), the maximum average water temperature (16.5 degrees Celsius) and minimum average water temperature (14.8 degrees Celsius) within that period are extracted from the sequence set, and the temperature variation range is calculated. The temperature was set to 1.5 degrees Celsius. Simultaneously, the total water consumption within this period was accumulated, resulting in 0.75 cubic meters. A threshold of 1.5 degrees Celsius for temperature variation and 0.6 cubic meters for water consumption were set (both thresholds were set based on the 85th percentile of historical experimental data). In this period, if 1.7 degrees Celsius was greater than 1.5 degrees Celsius and 0.75 cubic meters was greater than 0.6 cubic meters, both were considered met, and this was recorded as a "correlation event." All (e.g., 130) corrected periods within 72 hours were iterated over, and the number of correlation events (e.g., 104 times) was counted. The correlation rate was calculated as follows: The features of all the corrected cycles (start and end times, average flow rate, temperature change range, and total water consumption) and the calculated total correlation coefficient (0.80) are integrated to generate a temperature flow cycle feature set.

[0027] The thermal response generation submodule, based on the temperature flow cycle feature set, calculates the synchronous change rate of temperature change amplitude and water use behavior correlation rate and the thermal disturbance intensity under flow velocity change, analyzes the energy distribution intensity range of thermal disturbance in multiple cycles, and generates a water ecological thermal response feature set. The generated temperature flow cycle feature set is retrieved. This feature set includes the start and end times, average flow rate, temperature change amplitude, total water consumption, and a global correlation rate (0.80) for all effective cycles. First, the synchronous change rate of temperature change amplitude and water consumption behavior (quantified here as the total water consumption of the cycle) is calculated. This calculation compares two consecutive cycles one by one. Data for cycle 1 (06:30-08:30) is extracted from the feature set: temperature change amplitude 1.7 degrees Celsius, total water consumption 0.75 cubic meters. Data for the immediately following cycle 2 (08:30-10:00) is then extracted: temperature change amplitude 1.4 degrees Celsius, total water consumption 0.55 cubic meters. The amplitude change from cycle 1 to cycle 2 is as follows: Temperature decreased by degrees Celsius. The water consumption change from cycle 1 to cycle 2 was as follows: Cubic meters (decreasing). Since the two indicators change in the same direction (both negative), this is marked as "synchronous". Iterate through all (e.g., 129) adjacent cycle transitions and count the number of "synchronous" occurrences (e.g., 101 times). The synchronous change rate is then calculated as... Next, the thermal disturbance intensity under the change in flow velocity is calculated for each cycle. The thermal disturbance intensity is calculated by multiplying the average flow velocity within the cycle by the temperature change amplitude of that cycle. Taking cycle 1 (06:30-08:30) as an example, its average flow velocity is 1.12, and the temperature change amplitude is 1.7 degrees Celsius. The thermal disturbance intensity for this cycle is... For period 2 (08:30-10:00), the average flow velocity is 0.98, the temperature variation is 1.4 degrees Celsius, and the thermal disturbance intensity is... This calculation is applied to all 130 cycles, generating a time series of thermal disturbance intensity. Finally, the energy distribution intensity range of the thermal disturbance over multiple cycles is analyzed. This step requires setting a baseline value for the intensity range. Through statistical analysis of tens of thousands of thermal disturbance intensity values ​​in the historical database, the percentile method is used to determine the interval boundaries: low intensity range (0.0 to 1.0, approximately 40% of historical data), medium intensity range (1.0 to 2.5, approximately 45% of historical data), and high intensity range (greater than 2.5, approximately 15% of historical data). The 130 thermal disturbance intensity values ​​calculated from this 72-hour monitoring (including 1.904, 1.372, etc.) are compared with the above intervals. The statistical results are: 50 values ​​fall into the low intensity range, 68 values ​​fall into the medium intensity range, and 12 values ​​fall into the high intensity range. Based on this, the energy distribution intensity percentage is calculated: low intensity percentage... Medium intensity percentage High intensity ratio The calculated synchronous change rate (0.783), thermal disturbance intensity sequence, and energy distribution intensity interval proportions (38.5%, 52.3%, 9.2%) were compiled to generate a water ecological thermal response feature set.

[0028] Specifically, such as Figure 2 , 4 As shown, the ecological analysis module includes: The element acquisition submodule collects and matches the concentrations of dissolved oxygen, dissolved nitrogen, dissolved phosphorus, and chlorophyll a, calculates the average value and rate of change within adjacent sampling periods, removes outliers, and generates a set of dissolved element concentration sequences. A multi-parameter online water quality analyzer deployed at the water monitoring section was activated, simultaneously collecting concentration data of four dissolved elements at one-hour intervals. Dissolved oxygen concentration was measured using a fluorescence quenching sensor, in milligrams per liter (mg / L); dissolved nitrogen concentration was measured using an ion-selective electrode sensor, in milligrams per liter; dissolved phosphorus concentration was measured using a wet-oxidation ammonium molybdate spectrophotometric analyzer, in milligrams per liter; and chlorophyll a concentration was measured using a three-color excitation fluorescence sensor, in micrograms per liter (µg / L). All collected instantaneous data were timestamped to construct a dataset with a unified time reference. For example, at 10:00 AM, a matching set of data was recorded as follows: dissolved oxygen 7.5 mg / L, dissolved nitrogen 1.82 mg / L, dissolved phosphorus 0.11 mg / L, and chlorophyll a concentration 15.8 µg / L. After data acquisition, the system processed the data in continuous four-hour sampling periods. Taking the period from 08:00 to 12:00 as an example, four dissolved oxygen concentration values ​​were retrieved within this period: [7.8, 7.6, 7.5, 7.4] mg / L. These four values ​​were added together to obtain 30.3, and then divided by the number of samples (4) to calculate the average dissolved oxygen concentration for this period as 7.575 mg / L. The same summation and averaging operations were performed on the dissolved nitrogen, dissolved phosphorus, and chlorophyll a data for the same period. Next, the rate of change of the average value between adjacent sampling periods was calculated. If the average dissolved oxygen concentration calculated for the previous 4-hour period (04:00 to 08:00) was 7.950 mg / L, the rate of change of dissolved oxygen concentration between the two periods was obtained by dividing the difference between the two by the period length (4 hours), i.e., 7.575 - 7.950 = -0.375. Dividing this difference by 4 hours yielded -0.09375 mg / L per hour. Before calculating the average value, outlier removal was performed on the original collected data. The criteria for identifying outliers were determined through experimental verification using historical data. Taking dissolved phosphorus as an example, the experiment retrieved hourly dissolved phosphorus concentration data from the past 30 consecutive days, totaling 720 data points. The arithmetic mean of these 720 data points was calculated to be 0.09 mg / L, and the standard deviation was 0.02 mg / L. The upper and lower limits for outlier identification were set as the mean plus or minus three times the standard deviation, resulting in a normal range of 0.03 mg / L to 0.15 mg / L. If, during this data collection, the dissolved phosphorus reading at a certain moment was 0.17 mg / L, it was considered an outlier because it exceeded the upper limit of 0.15. This outlier was replaced by a replacement value, calculated by adding the normal value of 0.11 mg / L from the previous hour and the normal value of 0.12 mg / L from the following hour, and then dividing by two, resulting in 0.115 mg / L. The same process is applied to the data for dissolved oxygen, dissolved nitrogen, and chlorophyll a, with each threshold determined by its own historical data experimental verification process.After all data has undergone outlier processing and the calculation of time-period averages and rates of change, the average concentration values ​​of the four elements and their respective rates of change for each 4-hour period are organized in chronological order to generate a set of dissolved element concentration sequences.

[0029] The feature normalization submodule, based on the set of dissolved element concentration sequences, calculates multiple normalization ratios for the numerical range of multiple elements, adjusts the abnormal deviation range according to the consistency of time series, calculates the synchronous change ratio among multiple elements, and generates a normalized feature set of ecological elements. The module receives a generated set of dissolved element concentration sequences, containing average concentrations of four ecological elements organized over four-hour periods. Due to differences in the numerical ranges and units of each element, the module first normalizes the data, specifically calculating the concentration ratios of dissolved nitrogen to dissolved phosphorus, and chlorophyll a to dissolved phosphorus. Taking the data from 08:00 to 12:00 as an example, the average dissolved nitrogen concentration during this period is 1.75 mg / L, the average dissolved phosphorus concentration is 0.10 mg / L, and the average chlorophyll a concentration is 16.2 μg / L. Therefore, the nitrogen-phosphorus ratio is calculated by dividing 1.75 by 0.10, resulting in 17.5. The chlorophyll-phosphorus ratio is calculated as 16.2 / 0.10 = 162.0. This ratio calculation is applied to all four-hour periods in the sequence set. Next, adjustments are made to the calculated ratio sequences for abnormal deviations based on the consistency of the time series. This process requires setting a consistency deviation threshold, which is determined based on experimental statistical analysis of historical nitrogen-phosphorus ratio data. The experiment involves retrieving nitrogen-phosphorus ratio data from the past 30 consecutive days (180 four-hour time intervals), calculating the difference between the ratios of all adjacent time intervals in the sequence, forming a difference sequence. Then, the standard deviation of this difference sequence is calculated, yielding a value of 2.0. The consistency deviation threshold is set to 2.5 times this standard deviation, i.e., 5.0. When processing the generated nitrogen-phosphorus ratio sequence, if the ratio for the period from 08:00 to 12:00 is 17.5, while the ratio for the next period from 12:00 to 16:00 is 23.8, the absolute value of the difference is 6.3, which is greater than the threshold of 5.0. Therefore, the nitrogen-phosphorus ratio for the period from 12:00 to 16:00 is considered to have an abnormal deviation. This value will be adjusted by adding the deviation threshold to the value of the previous period, i.e., 17.5 plus 5.0, correcting it to 22.5. The chlorophyll-to-phosphorus ratio sequence was also processed using the same experimental and adjustment procedures. Finally, the synchronous change ratio among multiple elements was calculated; dissolved oxygen and chlorophyll a were selected for analysis. The direction of change of the average values ​​of these two elements in adjacent time periods was compared. For example, from time period A to time period B, the average dissolved oxygen value increased from 7.5 mg / L to 7.7 mg / L (positive change), and the average chlorophyll a value increased from 16.0 μg / L to 16.8 μg / L (positive change), with both showing the same direction of change. From time period B to time period C, dissolved oxygen decreased to 7.4 mg / L (negative change), and chlorophyll a increased to 17.1 μg / L (positive change), with both showing different directions of change. Traversing all time period transitions within the entire 72-hour period, if there were 17 transitions in total, and 12 of them showed the same direction of change, then the synchronous change ratio was calculated as 12 / 17 = 0.706. The adjusted nitrogen-to-phosphorus ratios and chlorophyll-to-phosphorus ratios for all time periods, along with the calculated synchronous change ratios, were integrated to generate a normalized feature set of ecological elements.

[0030] The ecological assessment submodule calls the ecological element normalized feature set and the water ecological thermal response feature set, performs correlation calculation on the ratio of energy change rate to element concentration in the two types of feature sequences, calculates the metabolic synergy ratio among multiple elements, analyzes the ecological energy metabolism synergy coefficient, and generates an ecological stability sequence. The generated ecological element normalized feature set and aquatic ecological thermal response feature set are invoked. The aquatic ecological thermal response feature set contains a thermal disturbance intensity sequence arranged with an approximately 2-hour period, with values ​​such as [1.904, 1.372, 2.150, 1.880], in meters multiplied by degrees Celsius per second. The ecological element normalized feature set has a time resolution of 4 hours. To unify the analysis benchmark, the thermal disturbance intensity sequence is first time-aligned by calculating the arithmetic mean of all thermal disturbance intensity values ​​within each 4-hour period. For example, if the period from 04:00 to 08:00 contains two thermal disturbance intensity values, 1.904 and 1.372, then the average thermal disturbance intensity for that period is (1.904 + 1.372) / 2 = 1.638. This operation is applied to all 4-hour periods to generate an energy sequence time-aligned with the element feature set. Subsequently, correlation calculations are performed on the energy change rate and the element concentration ratio (using the nitrogen-phosphorus ratio). First, the energy change rate sequence is obtained by calculating the difference between values ​​in adjacent time periods of the energy sequence. Simultaneously, an adjusted nitrogen-phosphorus ratio sequence is extracted from the normalized feature set of ecological elements. Then, the Pearson correlation coefficient between these two paired sequences is calculated. This process is not directly written into the formula, but rather its operational steps are described in words: the average values ​​of the energy change rate sequence and the nitrogen-phosphorus ratio sequence are calculated separately; then, the deviation of each data point from the average value of its respective sequence is calculated; the deviations of each pair of paired data points are multiplied; all these products are summed to obtain the covariance; finally, the covariance is divided by the product of the standard deviations of the two sequences. The calculated correlation coefficient is -0.52. Next, the metabolic synergy ratio among multiple elements is calculated, defined as the ratio of the average change rate of dissolved oxygen to the average change rate of dissolved phosphorus within the same 4-hour time period. Data was retrieved from a set of dissolved element concentration sequences. For example, if the dissolved oxygen change rate was -0.09375 mg / L / h and the dissolved phosphorus change rate was -0.00125 mg / L / h, then the metabolic synergy ratio would be -0.09375 / (-0.00125) = 75.0. Finally, the ecological energy metabolism synergy coefficient was analyzed. This coefficient is a weighted combination of the correlation calculation result and the metabolic synergy ratio. The weight values ​​were determined through an independent experimental verification process: 100 sets of sample data with known biodiversity indices (as a reference for ecological stability) were selected from a historical database. Through multiple linear regression analysis, it was determined that when the weight of the correlation coefficient was 0.6 and the weight of the metabolic synergy ratio was 0.4, the calculated synergy coefficient showed the highest good of fit with the biodiversity index. Therefore, the weights were set to 0.6 and 0.4. Before the weighted calculation, the metabolic synergy ratio needed to be normalized. The normalization baseline was determined through statistical historical data, with its 95th percentile being 120.0. The normalized value of the metabolic synergy ratio of 75.0 for the current period is 75.0 / 120.0 = 0.625. The ecological energy metabolism synergy coefficient for this period is 0.6. (-0.52) + 0.4 0.625 = (-0.312) + 0.25, which equals -0.062. This coefficient represents the ecological stability for that time period. By performing this calculation for each of the 4-hour periods, an ecological stability sequence is generated.

[0031] Specifically, such as Figure 2 , 5 As shown, the pollution analysis module includes: The residual chlorine monitoring submodule, based on the ecological stability sequence, detects the residual chlorine concentration in multiple sections of the water supply path, calculates the concentration change rate within adjacent sampling periods, removes abnormal drift points and corrects data discontinuities, and establishes a residual chlorine concentration change sequence. The process begins upon receiving the generated ecological stability sequence. This sequence is used as the basis for monitoring and scheduling: when a period in the sequence has an ecological stability value below -0.05 (e.g., -0.062 calculated previously), that period is marked as an "unstable period," and the residual chlorine monitoring frequency for the corresponding water supply path segment is increased from the standard once per hour to once every 30 minutes. In this embodiment, online current-type residual chlorine sensors are deployed in three key water supply segments (Segment A: water plant outlet, Section B: midpoint of the pipeline network, Section C: end of the pipeline network) to collect residual chlorine concentration data at 30-minute intervals, in milligrams per liter (mg / L). At 10:00, the collected data are: Section A 1.15 mg / L, Section B 0.85 mg / L, Section C 0.45 mg / L. At 10:30, the collected data are: Section A 1.12 mg / L, Section B 0.82 mg / L, Section C 0.41 mg / L. Subsequently, the concentration change rate within adjacent sampling periods is calculated. Taking segment B as an example, the calculation process for the rate of change at 10:30 is as follows: subtract the concentration at 10:00 from the concentration at 10:30 (0.82), resulting in a difference of -0.03. Divide this difference by the time interval of 30 minutes to obtain -0.001 mg / L / min. Next, abnormal drift point removal is performed. The threshold for judging abnormal drift is determined through a 72-hour stability experiment on the sensor. Experimental procedure: A calibrated sensor is placed in a chlorine solution of constant concentration (0.80 mg / L), and readings are recorded every 30 minutes. The differences between all adjacent readings are calculated, and the standard deviation of this difference sequence is calculated to be 0.015 mg / L. The "abnormal drift threshold" is set to three times this standard deviation, i.e., 0.045 mg / L. In actual monitoring, any point where the absolute value of the concentration change within a 30-minute interval is greater than 0.045 mg / L is judged as an abnormal drift point. For example, if segment B has a reading of 0.80 mg / L at 11:00, and then abruptly changes to 0.95 mg / L at 11:30, the absolute value of the change is 0.15 mg / L, which is greater than 0.045 mg / L. Therefore, the 11:30 reading of 0.95 is marked as an anomaly. After removing the anomaly, data gaps are corrected. Data gaps are defined as missing data points caused by anomaly removal or sensor communication interruption. Correction uses linear interpolation. Continuing the example, the 11:30 reading of 0.95 is removed. The previous normal reading (0.80 mg / L at 11:00) and the next normal reading (0.78 mg / L at 12:00) are found. By calculating (0.80 + 0.78) / 2 = 0.79, this value is used to replace the abnormal data at 11:30. This process is applied to all data points in all monitoring segments. After completing the above processing, the corrected residual chlorine concentration values ​​of the three sections A, B, and C and their corresponding concentration change rates are arranged in order according to a 30-minute time interval to generate a residual chlorine concentration change sequence.

[0032] The fluctuation calculation submodule, based on the residual chlorine concentration change sequence and soluble organic carbon concentration value, performs differential normalization for the same time period, calculates the relative change amplitude between the two types of concentrations, analyzes the synchronous fluctuation intensity value of organic carbon in multiple water supply sections, and generates a pollutant concentration fluctuation rate set. The generated residual chlorine concentration change sequence was retrieved, and soluble organic carbon (DOC) concentration values, in milligrams per liter, were simultaneously obtained from online ultraviolet absorption (UV254) analyzers deployed in the same three sections (A, B, and C). The DOC sampling interval was also 30 minutes, precisely aligned with the residual chlorine data in terms of timestamps. Taking the data from section B at 10:30 as an example, the corrected concentration at that moment was retrieved from the residual chlorine concentration change sequence as 0.82 mg / L, and the concentration at the previous moment (10:00) as 0.85 mg / L. Simultaneously, the DOC concentration in section B at 10:30 was 2.3 mg / L, and at 10:00 it was 2.1 mg / L. First, differential normalization was performed on these two types of concentration data, i.e., the relative change rate was calculated. The relative change rate of residual chlorine in section B at 10:30 was calculated as: (0.82-0.85) / 0.85 = -0.0353. The relative change rate of DOC in section B at 10:30 is calculated as: (2.3-2.1) / 2.1 = 0.0952. Next, the relative change amplitude between the two concentrations is calculated. This amplitude is defined as the absolute difference between the relative change rate of DOC and the relative change rate of residual chlorine at the same time. Taking section B at 10:30 as an example, the relative change amplitude is 0.0952 minus the absolute value of -0.0353, which is 0.1305. The purpose of this calculation is to quantify the dynamic imbalance between the substances that consume residual chlorine (DOC) and the residual chlorine itself in the water. The advantage of this method is that by performing differential normalization on residual chlorine and organic carbon, and calculating their relative change amplitudes and synchronous fluctuations between sections, the instantaneous dynamics of organic matter intrusion events (manifested as an increase in DOC and a decrease in residual chlorine) can be quantified, eliminating the interference of baseline concentrations in different sections. Subsequently, the synchronous fluctuation intensity of organic carbon across multiple water supply sections is analyzed. This step compares the consistency of the relative rate of change of DOC in space (e.g., between segments A and B). At the same time point of 10:30, the relative rate of change of DOC in segment A is 0.0210 (rising from 2.05 mg / L to 2.09 mg / L). The relative rate of change of DOC in segment B is 0.0952. The calculation process for the synchronous fluctuation intensity value between segments A and B is as follows: subtract the rate of change of segment A from 1; the absolute value of the difference between 0.0210 and 0.0952 is 0.0742. Then divide by the sum of the absolute values ​​of the two: 0.0210 + 0.0952 = 0.1162. The calculation result is 1 - (0.0742 / 0.1162) = 0.3614. This value is low, indicating that the DOC fluctuations in segments A and B are not synchronized. The relative change rates of residual chlorine and DOC at all 30-minute time points and in all sections, along with their relative magnitudes and the synchronous fluctuation intensity of DOC between sections, are aggregated to generate a pollutant concentration fluctuation rate set.

[0033] The pollution classification submodule calls the pollutant concentration fluctuation set and the ecological stability sequence, performs probability distribution comparison between pollutant fluctuation and thermal disturbance intensity, calculates the pollution trigger probability interval for multiple time periods, divides multiple risk levels according to the classification threshold, and generates a residential water pollution activation sequence. The generated pollutant concentration fluctuation set is retrieved, and simultaneously, the generated ecological stability sequence (-0.062, -0.070, 4-hour resolution) and the thermal disturbance intensity sequence (1.638, 1.965, 4-hour resolution) from the aquatic ecological thermal response feature set are retrieved. First, the data is time-aligned, and the pollutant concentration fluctuation set at 30-minute resolution (taking "relative change amplitude" as an example) is arithmetically averaged over a 4-hour period. For example, in the period from 08:00 to 12:00, there are a total of 8 "relative change amplitude" values; summing these values ​​and dividing by 8 yields the result for that time period. The average volatility of the segment is set, for example, 0.115. Next, a probability distribution comparison is performed between the aligned pollutant volatility (average relative change) and the thermal disturbance intensity. This process first requires defining the criteria for high-risk events. The threshold for high pollutant volatility is determined experimentally using historical data: the 90th percentile of the average volatility over 1000 historical 4-hour periods is 0.150. Therefore, the "high volatility threshold" is set to 0.150. The threshold for high thermal disturbance intensity is set to 2.5 m·°C based on the cluster analysis results from the previous process. / second. Then, calculate the probability interval for pollution triggering across multiple time periods. Using a 24-hour window (six 4-hour periods), calculate the number of "double-high" periods within the window that simultaneously meet both the conditions of "average volatility greater than 0.150" and "thermal disturbance intensity greater than 2.5". For example, in the first 24-hour window, two out of the six periods simultaneously meet the "double-high" condition. Therefore, the pollution triggering probability for this window is 2 / 6 = 0.333. Slide the window for four hours to calculate the probability for the second 24-hour window, for example, 0.167. Iterate through all 72-hour periods. The data yields a probability sequence [0.333, 0.167, ...], with a probability interval of [0.167, 0.333]. Finally, multiple risk levels are determined based on classification thresholds. This step introduces the ecological stability sequence as a correction factor to calculate the final "pollution activation index." The calculation process is as follows: the pollution activation index equals the pollution trigger probability multiplied by (1 plus the absolute value of the ecological stability value). Taking the first 24-hour window as an example, its probability is 0.333, and the corresponding average ecological stability for this period is -0.065. The pollution activation index is 0.333. (1 + 0.065) = 0.333 1.065 = 0.355. The risk level classification thresholds were determined based on historical data backtesting experiments: 100 calculated activation indices from the past were compared with water quality complaint events during the same period to set thresholds. Experimental results showed that: indices below 0.15 indicated no complaints (low risk); indices between 0.15 and 0.30 indicated sporadic odor complaints (medium risk); and indices above 0.30 indicated concentrated bacterial colony exceedances or odor complaints (high risk). Based on this, the medium risk threshold was set at 0.15, and the high risk threshold at 0.30. In this example, the calculated value of 0.355 is greater than 0.30, therefore this 24-hour window was classified as high risk.

[0034] Table 1. Examples of Pollution Activation Risk Level Classification

[0035] As shown in Table 1, this table lists the risk level classification process and results of 72-hour monitoring data (sliding in 24-hour windows). The pollution activation index and risk level corresponding to all time windows are arranged in chronological order to generate a residential water pollution activation sequence.

[0036] Specifically, such as Figure 2 , 6 As shown, the risk assessment module includes: The pollution propagation submodule extracts water supply flow, pollutant concentration and time series data based on the activation sequence of pollution in residential water supply, calculates the concentration transfer coefficient of pollutants between the water supply network and the aquatic ecosystem, analyzes the propagation path and ratio of pollution and uses it as a characteristic quantity of pollution diffusion, and generates a pollution propagation probability sequence. Based on the aforementioned residential water pollution activation sequence, the pollution activation index for the first 24-hour time window in the sequence was 0.355, indicating a high-risk level. This 24-hour window was then locked, and synchronous monitoring data from the end of the water supply network (section C) and the adjacent aquatic ecological monitoring point (point E1) were extracted during this period. The water flow rate of section C was measured using a pipeline electromagnetic flowmeter, with an average value of 1500 cubic meters per hour. Soluble organic carbon (DOC) concentration data was used; data for section C was obtained from an in-pipeline ultraviolet absorption analyzer, while data for monitoring point E1 was obtained from a soil infiltration water sampling analyzer located below it. To calculate the concentration transfer coefficient of pollutants between the water supply network and the aquatic ecosystem, a time delay parameter needed to be determined. This parameter was obtained through on-site tracer experiments: tracers were injected upstream of section C, and the times when peak concentrations were detected at monitoring points C and E1 were recorded. The average time difference across multiple experiments was 4 hours, which is the transmission delay. During calculation, the time series data for monitoring point E1 was shifted forward by 4 hours. Taking 08:00 within the high-risk window as an example, the DOC concentration of segment C at 08:00 was 2.8 mg / L, and the concentration 4 hours earlier (04:00) was 2.2 mg / L. Simultaneously, the DOC concentration of monitoring point E1 at 12:00 (4 hours later) was 0.45 mg / L, and the concentration at the current time (08:00) was 0.21 mg / L. The concentration transfer coefficient was calculated as follows: the concentration increment of monitoring point E1 in the corresponding time period was calculated as 0.45 - 0.21 = 0.24; then the concentration increment of segment C in the corresponding time period was calculated as 2.8 - 2.2 = 0.6; finally, the concentration transfer coefficient at that moment was 0.40, calculated by dividing the concentration increment of E1 (0.24) by the concentration increment of segment C (0.6). This coefficient is the pollution diffusion characteristic quantity, and its value is directly used in subsequent analysis. This calculation is applied to every hour within the high-risk window. By analyzing the transmission paths and ratios of pollution, the concentration transfer coefficients calculated for all hours are combined in chronological order to generate a pollution transmission probability sequence.

[0037] The migration rate submodule, based on the pollution propagation probability sequence, selects velocity gradient and spatial concentration difference data within the same time period, calculates the concentration change ratio of pollutants between adjacent areas and the velocity coupling coefficient, compares the concentration gradient change trend over consecutive time periods, and generates a pollution migration rate sequence. Based on the pollution propagation probability sequence, time periods with a concentration transfer coefficient greater than 0.35 are selected for analysis. This threshold was determined through backtracking analysis of data from 50 historical pipeline leakage events. When the transfer coefficient exceeds this value, 95% of the events ultimately lead to environmental pollution requiring intervention. According to the calculation in the previous module, the coefficient at 08:00 is 0.40, exceeding the threshold, thus locking this time period. Velocity and concentration data from monitoring points E2 (located 5 meters downstream of monitoring point E1) and E1 are retrieved within the same time period. Velocity is measured using an acoustic Doppler current meter, and DOC concentration is measured using an ultraviolet absorption analyzer. At 08:00, the groundwater flow velocity at point E1 was 0.005 m / s, and the DOC concentration was 0.21 mg / L; the flow velocity at point E2 was 0.002 m / s, and the DOC concentration was 0.08 mg / L. First, the concentration change rate of pollutants between adjacent areas was calculated: subtracting the concentration of E2 (0.08 mg / L) from the E1 concentration (0.21 mg / L) yielded a spatial concentration difference of 0.13 mg / L. This difference was then divided by the distance between the two points (5 meters), resulting in a concentration change rate of 0.026 mg / L / m. Next, the velocity coupling coefficient was calculated. This coefficient calculation relied on a coupling parameter determined through a laboratory soil column simulation experiment. The experiment, by fitting DOC migration time data under different flow velocities, determined the coupling parameter value to be 50. The calculation process for the velocity coupling coefficient was as follows: first, the velocity gradient between points E1 and E2 was calculated, i.e., (0.005 - 0.002) / 5 = 0.0006 m / s. Then, the negative power of the natural constant e (coupling parameter 50 multiplied by the velocity gradient 0.0006) is calculated from 1, i.e., 1 minus e to the power of -0.03, yielding 0.0296. Subsequently, the concentration gradient trend over consecutive time periods is compared. The concentration value at 09:00 is retrieved, and the new concentration gradient is calculated to be 0.028 mg / L / m, greater than the 0.026 at 08:00, indicating an increasing gradient trend. The pollution migration rate is calculated as follows: [The formula is incomplete and requires further context to translate accurately.] 0.0296 1.1 = 0.0008466. Arrange the rate values ​​calculated for all analysis periods in chronological order to generate a pollution migration rate sequence.

[0038] The risk analysis submodule calls the pollution migration rate sequence and combines it with the ecological stability sequence to calculate the time correlation coefficient between pollution migration rate and ecological stability, extract the risk change trend, compare it with the risk benchmark threshold range to classify the risk level, and generate the preliminary water environment risk assessment results. Upon receiving the pollution migration rate sequence, an ecological stability sequence with a time resolution aligned to 1 hour is simultaneously retrieved. Taking 08:00 as an example, the migration rate is 0.0008466, and the ecological stability is -0.068. First, the time correlation coefficient between the pollution migration rate and ecological stability over the entire 72-hour monitoring period is calculated. The calculation process involves constructing two paired sequences containing 72 data points, calculating the mean and standard deviation of each sequence, then calculating the covariance, and finally dividing the covariance by the product of the two standard deviations, yielding a time correlation coefficient of -0.75. Next, the risk change trend is extracted. This trend is quantified by calculating the slope of the linear regression of the pollution migration rate sequence over 72 hours. The calculated slope is positive (0.000015), indicating that the pollution migration rate is generally accelerating. Finally, risk levels are classified by comparing the risk baseline threshold range. This step is accomplished by calculating a comprehensive "water environment risk index." The index is calculated as follows: the absolute value of the time correlation coefficient (0.75) is multiplied by the maximum value of the migration rate sequence (0.0015), and then multiplied by a trend correction factor. The trend correction factor is set according to the sign of the regression slope; for a positive slope, the factor is 1.2. The calculated risk index is 0.75. 0.0015 1.2 = 0.00135. The risk level classification threshold was determined through a retrospective analysis experiment of historical environmental events. The experiment correlated the consequences of historical events with the calculated index, classifying a risk index below 0.0005 as low risk, 0.0005 to 0.0012 as medium risk, and above 0.0012 as high risk. In this embodiment, the calculated risk index of 0.00135 is higher than the high-risk threshold of 0.0012. Therefore, the module classifies the final result of this assessment as "high risk," which is the generated preliminary water environment risk assessment result.

[0039] Specifically, such as Figure 2 , 7 As shown, the optimization evaluation module includes: The smoothing correction submodule, based on the preliminary assessment results of water environment risk, extracts the risk level and pollution migration rate of adjacent time periods in the risk time series data, calculates the average offset of the risk level difference and migration rate difference between adjacent time series, performs a weighted average of the risk levels of continuous time series, and generates a risk smoothing sequence. After receiving the preliminary water environment risk assessment results, which are a risk level sequence based on a 24-hour rolling window, such as [high risk, medium risk, high risk], the risk levels are first quantified: "low risk" is set to 1, "medium risk" to 2, and "high risk" to 3. Accordingly, the aforementioned sequence is converted into a numerical sequence [3, 2, 3]. The module extracts the risk levels of adjacent time periods and the pollution migration rates generated by the previous process. Taking the first and second time periods as examples, the extracted risk levels are 3 and 2, respectively, with corresponding pollution migration rates of 0.00135 and 0.00090. Next, the difference in risk levels and migration rates between adjacent time periods is calculated. The risk level difference is 3 minus 2, resulting in 1. The migration rate difference is 0.00135 - 0.00090 = 0.00045. Subsequently, the average offset is calculated, which is defined as the arithmetic mean of the ratio of the migration rate difference to the risk level difference for all adjacent time periods. For the first and second time series, the ratio is 0.00045 / 1 = 0.00045. The same calculation is performed on the second and third time series (risk levels 2 and 3, migration rates 0.00090 and 0.00120), yielding a ratio of 0.00030. The average offset is the sum of these two ratios (0.00045 and 0.00030) divided by 2, resulting in 0.000375. Finally, a weighted average of the risk levels for consecutive time series is calculated. The weights are set based on the pollution migration rate, specifically calculated as follows: the weight of the current time period equals the migration rate of the current time period divided by the sum of the migration rates of the current and previous time periods. Taking the second time period as an example, its weight is 0.00090 / (0.00090+0.00135) = 0.4. The smoothed risk value of the first time period is its initial quantification value of 3. The smoothed risk level of the second time period is calculated using the weight of the second time period, 0.4. 2 + (1 - 0.4) 3 = 2.6. Arrange the smoothed risk values ​​calculated for all time periods in chronological order to generate a risk smoothing sequence.

[0040] The weight adjustment submodule calls the risk smoothing sequence, combines the pollution migration rate with the pollution distribution change value in the pollution distribution change sequence, calculates the response sensitivity coefficient of the pollution migration rate to risk change, performs weighted correction on the time series risk value, and generates risk weight correction results. The generated risk smoothing sequence, such as [3, 2.6], is invoked, and the pollution migration rate sequence [0.00135, 0.00090] is retrieved simultaneously. A new parameter, the pollution distribution change rate, is introduced. This parameter is calculated by comparing the spatial distribution maps of pollutant concentrations at consecutive time points. Specifically, it is the difference between the area of ​​the region exceeding the standard (more than 0.1 mg / L) at the later time point and the area at the previous time point, divided by the area at the previous time point. For example, if the area exceeding the standard is 500 square meters at 08:00 and 520 square meters at 09:00, then the pollution distribution change rate at 09:00 is (520-500) / 500 = 0.04. Next, the sensitivity coefficient of the pollution migration rate to risk changes is calculated. This coefficient is defined as the ratio of the change in the smoothed risk value to the change in the pollution migration rate between adjacent time points. Taking the second time period as an example, the change in smoothed risk value is 2.6 - 3 = -0.4; the change in migration rate is 0.00090 - 0.00135 = -0.00045. The response sensitivity coefficient is -0.4 / -0.00045 = 888.9. The advantage of this method is that by calculating the response sensitivity of risk changes to the pollution migration rate and weighting it in conjunction with the rate of change in the spatial distribution of pollution, the risk assessment results not only reflect the current level of risk but also its potential for dynamic deterioration. Subsequently, the time-series risk value is weighted and corrected. The correction process is as follows: a correction factor is calculated, which is equal to 1 plus the product of the response sensitivity coefficient and the rate of change in pollution distribution. Setting the rate of change in pollution distribution for the second time period to 0.05, the correction factor is 1 + 888.9. 0.05 = 45.445. The risk weighting correction value is: 2.6, the smoothed risk value for this period. 45.445 = 118.157. Arrange the calculation results of all time periods in chronological order to generate the risk weight correction result.

[0041] The fitting optimization submodule extracts paired samples of pollution migration rate and pollution distribution change value based on the risk weight correction result, and performs time fitting and smoothing optimization on pollution migration rate and pollution distribution change to obtain water environment risk optimization result; The process begins based on the generated risk weight correction results. Paired samples of pollution migration rate and pollution distribution change rate are extracted to form a time series dataset. Taking the tenth time period as an example, its pollution migration rate is 0.00110, and the pollution distribution change rate is 0.06. First, time fitting and smoothing optimization are performed on the pollution migration rate and pollution distribution change rate. The fitting process for the pollution migration rate is as follows: for the tenth time period, the rate values ​​of the two time periods before and after it (i.e., the eighth, ninth, eleventh, and twelfth time periods) are extracted, which are [0.00095, 0.00102, 0.00115, 0.00118], respectively. The fitted value for the tenth time period is predicted by calculating the slope and intercept of the trend line formed by these four data points. The calculated fitted value is 0.00108. The smoothing optimization process for the pollution distribution change rate is as follows: a five-point weighted average method is used. The weight settings were determined through experiments. In a backtest of 100 historical data sets, the smoothing result showed the highest fit to the actual trend when the weight of the center point was 0.4, the weight of adjacent points was 0.2, and the weight of outer points was 0.1. Therefore, the weights were set to [0.1, 0.2, 0.4, 0.2, 0.1]. The distribution change rate for periods eight to twelve was [0.04, 0.05, 0.06, 0.065, 0.07], and the smoothing value for period ten was 0.1. 0.04 + 0.2 0.05 + 0.4 0.06 + 0.2 0.065 + 0.1 0.07 = 0.058. Finally, the risk value is optimized. The optimization calculation process is as follows: the obtained risk weight correction value is multiplied by an optimization coefficient composed of smoothed parameters. This optimization coefficient is equal to 1 plus the smoothed distribution change rate of 0.058, then multiplied by the fitted migration rate of 0.00108. The calculated optimization coefficient is 1.00006264. If the risk weight correction value for the tenth time period is 125.0, then the optimized value for water environment risk is 125.0. 1.00006264 = 125.00783. The calculation results for all time periods are aggregated to obtain the optimized water environment risk results.

[0042] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives, characterized in that, The system includes: The water temperature and flow rate module collects the water temperature, flow rate, and water consumption of residents in the ecological water body, separates the periodic characteristics of water temperature and flow rate changes, analyzes the synchronous change rate of temperature and water use behavior, calculates the intensity of thermal disturbance, generates a water ecological thermal response feature set, and transmits it to the ecological analysis module. The ecological analysis module, based on the water ecological thermal response feature set, collects dissolved oxygen, dissolved nitrogen, dissolved phosphorus and chlorophyll a concentration and normalizes the correlation, calculates the ecological energy metabolism synergy coefficient, generates the ecological stability sequence and transmits it to the pollution analysis module. The pollution analysis module detects the residual chlorine concentration and soluble organic carbon concentration in the water supply path based on the ecological stability sequence, calculates the pollutant concentration fluctuation rate by normalization and performs probability classification in combination with the thermal disturbance intensity, generates the residential water pollution activation sequence and transmits it to the risk assessment module. The risk assessment module calculates the probability of pollution propagation between water supply and aquatic ecosystem through the activation sequence of pollution in residential water use, derives the pollution migration rate sequence and pollution distribution change sequence, and analyzes the risk time series data in conjunction with the ecological stability sequence to generate the preliminary water environment risk assessment results.

2. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives as described in claim 1, characterized in that: The water ecological thermal response feature set includes water temperature periodic characteristics, flow velocity periodic characteristics, and thermal disturbance intensity. The ecological stability sequence includes dissolved oxygen correlation, dissolved nitrogen correlation, and dissolved phosphorus correlation. The residential water pollution activation sequence includes residual chlorine concentration fluctuation rate, soluble organic carbon concentration fluctuation rate, and pollutant concentration probability distribution. The preliminary water environment risk assessment results include pollution propagation probability, pollution migration rate sequence, and pollution distribution change sequence.

3. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives as described in claim 1, characterized in that, The water temperature and flow rate module includes: The data analysis submodule collects water temperature and flow rate, records residents' water consumption meters, and performs synchronous matching of the three types of data. It arranges the water temperature, flow rate and water consumption values ​​of multiple time periods in a time series, calculates the average value and rate of change within the continuous sampling interval, and generates a synchronized sequence set of water temperature and flow rate. The feature separation submodule calls the water temperature and flow rate synchronization sequence set, extracts the fluctuation period of water temperature and flow rate, calculates the difference between the mean values ​​of the periods and filters out anomalies, analyzes the daily variation pattern of water consumption to match and correct the periodic features, extracts the correlation rate between temperature change amplitude and water consumption behavior, and generates a temperature and flow periodic feature set. The thermal response generation submodule, based on the temperature flow cycle feature set, calculates the synchronous change rate of temperature change amplitude and water use behavior correlation rate and the thermal disturbance intensity under flow velocity change, analyzes the energy distribution intensity range of thermal disturbance in multiple cycles, and generates a water ecological thermal response feature set.

4. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives as described in claim 1, characterized in that, The ecological analysis module includes: The element acquisition submodule collects and matches the concentrations of dissolved oxygen, dissolved nitrogen, dissolved phosphorus, and chlorophyll a, calculates the average value and rate of change within adjacent sampling periods, removes outliers, and generates a set of dissolved element concentration sequences. The feature normalization submodule, based on the dissolved element concentration sequence set, calculates multiple normalization ratios for the numerical range of multiple elements, adjusts the abnormal deviation range according to the consistency of the time series, calculates the synchronous change ratio among multiple elements, and generates a normalized feature set of ecological elements. The ecological assessment submodule calls the normalized feature set of ecological elements and the feature set of water ecological thermal response, performs correlation calculation on the ratio of energy change rate to element concentration in the two feature sequences, calculates the metabolic synergy ratio among multiple elements, analyzes the ecological energy metabolism synergy coefficient, and generates an ecological stability sequence.

5. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives as described in claim 1, characterized in that, The pollution analysis module includes: The residual chlorine monitoring submodule, based on the ecological stability sequence, detects the residual chlorine concentration in multiple sections of the water supply path, calculates the concentration change rate within adjacent sampling periods, removes abnormal drift points and corrects data discontinuities, and establishes a residual chlorine concentration change sequence. The fluctuation calculation submodule, based on the residual chlorine concentration change sequence and soluble organic carbon concentration value, performs differential normalization for the same time period, calculates the relative change amplitude between the two types of concentrations, analyzes the synchronous fluctuation intensity value of organic carbon in multiple water supply sections, and generates a pollutant concentration fluctuation rate set. The pollution classification submodule calls the pollutant concentration fluctuation set and the ecological stability sequence, performs a probability distribution comparison between pollutant fluctuation and thermal disturbance intensity, calculates the pollution trigger probability interval for multiple time periods, divides multiple risk levels according to the classification threshold, and generates a residential water pollution activation sequence.

6. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives as described in claim 5, characterized in that, The classification threshold is determined by calculating the joint probability density function of pollutant volatility and thermal disturbance intensity to obtain the distribution overlap rate of the two in multiple time intervals, and then the risk boundary is set according to the quantile interval of the overlap rate.

7. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives as described in claim 1, characterized in that, The risk assessment module includes: The pollution propagation submodule extracts water supply flow, pollutant concentration and time series data based on the residential water pollution activation sequence, calculates the concentration transfer coefficient of pollutants between the water supply network and the aquatic ecosystem, analyzes the pollution propagation path and ratio as pollution diffusion characteristic quantity, and generates a pollution propagation probability sequence. The migration rate submodule, based on the pollution propagation probability sequence, selects velocity gradient and spatial concentration difference data within the same time period, calculates the concentration change ratio and velocity coupling coefficient of pollutants between adjacent areas, compares the concentration gradient change trend over consecutive time periods, and generates a pollution migration rate sequence. The risk analysis submodule calls the pollution migration rate sequence and combines it with the ecological stability sequence to calculate the time correlation coefficient between pollution migration rate and ecological stability, extract the risk change trend, compare it with the risk benchmark threshold range to classify the risk level, and generate the preliminary water environment risk assessment results.

8. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives according to claim 7, characterized in that, The risk benchmark threshold is set by calculating the time correlation distribution between the pollution transmission probability and the ecological stability sequence, extracting the overlapping density interval of the two in continuous time series, and then setting it according to the quantile value of the distribution density.

9. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives as described in claim 1, characterized in that, The system also includes: The optimization assessment module, based on the preliminary water environment risk assessment results, smooths and corrects the risk time series data, and performs time fitting and smoothing optimization on the pollution migration rate and pollution distribution changes to generate water environment risk optimization results. The water environment risk optimization results include risk time-series smoothed data, weighted corrected data, and optimized fitting accuracy of pollution migration and distribution.

10. The regional water environment risk assessment system considering the impact on aquatic ecology and residents' lives according to claim 9, characterized in that, The optimization evaluation module includes: The smoothing correction submodule, based on the preliminary water environment risk assessment results, extracts the risk level and pollution migration rate of adjacent time periods in the risk time series data, calculates the average offset of the risk level difference and migration rate difference between adjacent time series, performs a weighted average of the risk levels of continuous time series, and generates a risk smoothing sequence. The weight adjustment submodule calls the risk smoothing sequence, combines the pollution migration rate and the pollution distribution change value in the pollution distribution change sequence, calculates the response sensitivity coefficient of the pollution migration rate to risk change, performs weighted correction on the time series risk value, and generates risk weight correction result. The fitting optimization submodule extracts paired samples of pollution migration rate and pollution distribution change value based on the risk weight correction result, and performs time fitting and smoothing optimization on pollution migration rate and pollution distribution change to obtain water environment risk optimization result.

Images