An intelligent water quality monitoring method and system

Abstract

The present invention discloses an intelligent water quality monitoring method and system, which relates to the technical field of environmental monitoring. First, water quality information and potential time series information at each position in the target water area are obtained, and an information set is obtained through preprocessing. Then, according to the information set, the charge imbalance situation at each position in the target water area is analyzed to determine the imbalance hot spot area and the imbalance non-hot spot area. The correlation between the charge imbalance situation and each parameter in the water quality information is respectively analyzed in the imbalance hot spot area and the imbalance non-hot spot area. After comparison, the coupling area is determined. Finally, the correlation between the charge imbalance index and the corresponding parameter in the coupling area is verified to obtain a parameter set, and a time stamp for triggering an early warning signal is respectively set for each parameter in the parameter set to formulate corresponding treatment means.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and particularly to an intelligent water quality monitoring method and system. Background Art

[0002] With the acceleration of the industrialization and urbanization processes, the problem of water body pollution has become increasingly complex and diverse. Traditional water quality monitoring mainly focuses on conventional indicators such as chemical substance concentration, pH value, dissolved oxygen, and chemical oxygen demand. However, at the microscopic level, the influence of the charge distribution state in water bodies on the ecosystem has gradually become prominent. Especially in terms of microbial activity and ecosystem stability, an imbalance in charge distribution may lead to abnormal microbial metabolism and changes in community structure, thereby threatening the health of the entire aquatic ecosystem, but this has been long ignored. Therefore, there is an urgent need for an intelligent water quality monitoring method that can accurately monitor the charge distribution and its ecological impact.

[0003] Currently, the existing water quality monitoring means have significant deficiencies in evaluating the impact of charge distribution on the ecosystem. On the one hand, there is a lack of effective monitoring technologies for charge distribution imbalance, making it difficult to accurately obtain the distribution signals of micro-scale charged particles in water bodies and unable to quantify the degree of charge imbalance. On the other hand, the long-term impact of charge distribution imbalance on the ecosystem is ignored. For example, in some eutrophic water bodies, due to the lack of monitoring and analysis of charge distribution, it is difficult to accurately explain the internal relationship between the changes in microbial community structure and the eutrophication process of water bodies, resulting in difficulty in timely warning of the stress risks caused by water quality deterioration to the ecosystem. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides an intelligent water quality monitoring method and system, which solves the problems in the above background art.

[0005] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent water quality monitoring method, including the following steps,

[0006] S1: Obtain the water quality information and potential time series information at each position in the target water area, and after preprocessing, obtain an information set;

[0007] S2: According to the information set, analyze the charge imbalance situation at each position in the target water area to determine the imbalance hot spot area and the non-imbalance hot spot area, and respectively analyze the correlation between the charge imbalance situation and each parameter in the water quality information in the imbalance hot spot area and the non-imbalance hot spot area. After comparison, determine the coupling area;

[0008] S3: Verify the correlation between the charge imbalance index and the corresponding parameters in the coupling area, obtain a parameter set, and set a time stamp for triggering an early warning signal for each parameter in the parameter set to formulate corresponding treatment measures.

[0009] Preferably, sensor group nodes and multiple groups of nanoelectrode arrays are arranged in the target water area according to the specified grid size to obtain water quality information and potential time series information at various positions in the target water area;

[0010] Three-point calibration is performed using standard solutions. The electrodes are placed in three different concentrations of standard solutions respectively, and the corresponding potential values are measured. By recording the potential change amount and ion concentration change amount under different standard solutions, the sensitivity coefficient of each electrode is calculated; the sensitivity coefficient is used to reflect the response ability of the electrode to ion concentration changes.

[0011] By using wavelet threshold denoising and baseline drift correction on the water quality information and potential time series information, a processed information set is obtained. According to the sensitivity coefficient of each electrode, the corresponding potential value in the information set is converted into charge concentration, and after normalization processing, charge time series information is obtained.

[0012] Preferably, according to the charge time series information, the average value and standard deviation of the charge concentration of all electrodes are calculated, and then through the average value and standard deviation of the charge concentration, the charge imbalance index of each electrode array is calculated, expressed as: , is the charge imbalance index of the corresponding electrode array at time t, is the standard deviation of the charge concentration of the corresponding electrode array at time t, is the average value of the charge concentration of the corresponding electrode array at time t, is a positive number;

[0013] Through the charge imbalance index of the corresponding electrode array at time t, a time-varying index sequence of the corresponding electrode array is obtained. The time-varying index sequence is used to record the change of the charge imbalance index of the corresponding electrode array at different times.

[0014] Preferably, the charge imbalance index corresponding to the position coordinates of each electrode array is collected, and Kriging interpolation method is used to reconstruct the charge imbalance index on the grid of the entire target water area. Specifically: according to the grid range of the entire target water area, the points to be interpolated are determined. According to the spatial correlation between the position of each point to be interpolated and the position coordinates of each collected electrode array, the charge imbalance index of the point to be interpolated is obtained, expressed as: , is the charge imbalance index of the corresponding point to be interpolated at time t, i is the number of the electrode array to be collected, n is the total number of the electrode arrays to be collected, is the charge imbalance index of the i-th electrode array at time t, is the weight coefficient of the i-th electrode array;

[0015] After reconstruction, a charge imbalance heat map at each moment is obtained, and threshold segmentation is performed on the charge imbalance heat map;

[0016] The charge imbalance index at each position in the charge imbalance heat map is compared with a preset threshold. If the charge imbalance index exceeds the threshold, the corresponding position is marked as an imbalance hot spot area; otherwise, the corresponding position is marked as a non-hot spot area of imbalance;

[0017] The coordinates of the imbalance hot spot area are extracted to obtain a set of hot spot area coordinates.

[0018] Preferably, based on the set of hot spot area coordinates, a set of non-hot spot area coordinates is determined, and the water quality information of the set of hot spot area coordinates and the set of non-hot spot area coordinates is extracted from the information aggregation, and the charge imbalance index of the set of hot spot area coordinates and the set of non-hot spot area coordinates is extracted from the time-varying index sequence;

[0019] The set of hot spot area coordinates and the set of non-hot spot area coordinates are combined to generate a set of area coordinates;

[0020] In the set of hot spot area coordinates and the set of non-hot spot area coordinates, the correlation coefficients between the charge imbalance index and each parameter in the water quality information are respectively counted. Specifically: each parameter in the time-varying index sequence and the water quality information corresponding to the time is used as an input and substituted into the correlation coefficient calculation function. The formula is: , where is the correlation coefficient between the charge imbalance index and the corresponding parameter, t is the moment number, m is the duration in the time-varying index sequence, is the mean value of the charge imbalance index in the time-varying index sequence, is the value of the corresponding parameter in the water quality information at moment t, is the mean value of the corresponding parameter;

[0021] After taking the absolute value of the correlation coefficients between the charge imbalance index and all parameters in the water quality information and performing mean value calculation, a stress coupling coefficient is constructed. The formula is: , is the stress coupling coefficient of the corresponding set of area coordinates, G is the total number of correlation coefficients in the corresponding set of area coordinates, g is the coordinate number in the corresponding set of area coordinates, K is the total number of parameters, k is the parameter number, is the correlation coefficient between the charge imbalance index at coordinate g in the corresponding set of area coordinates and the k-th parameter B.

[0022] Preferably, the stress coupling coefficients in the set of hot spot area coordinates and the set of non-hot spot area coordinates are respectively determined, and after subtraction calculation of the stress coupling coefficients in the set of hot spot area coordinates and the set of non-hot spot area coordinates, an influence difference value is obtained. If the influence difference value exceeds a preset difference threshold, the imbalance hot spot area is marked as a coupling area; otherwise, it is not a coupling area;

[0023] The coupling region is used to indicate the risk of water quality stress in the corresponding area.

[0024] Preferably, according to the determined coupling region, the correlation between the charge imbalance index and the corresponding parameters in the coupling region is verified, specifically including:

[0025] According to the sampling frequency, the maximum lag order is set. Based on the maximum lag order, a lag autoregressive term using only the corresponding parameters to predict the corresponding parameters and a lag autoregressive term with the charge imbalance index added are constructed respectively, specifically expressed as:

[0026] Using only the lag autoregressive term of the corresponding parameter to predict the corresponding parameter: ;

[0027] Adding the lag autoregressive term of the charge imbalance index at the same time:

[0028] ;

[0029] In the formula, P is the maximum lag order, p is the lag index, is the autoregressive coefficient, is the first residual term, is the regression coefficient of the charge imbalance index at time t - p, is the charge imbalance index at time t - p, is the second residual term, is the value of the corresponding parameter at time t - p;

[0030] Calculate the sum of squared residuals respectively to obtain the sum of squared residuals of the first residual term and the sum of squared residuals of the second residual term;

[0031] Based on the sum of squared residuals of the first residual term and the sum of squared residuals of the second residual term, an F statistic is constructed, and the formula is: , F is the F statistic, is the sum of squared residuals of the first residual term, is the sum of squared residuals of the second residual term, is the sample size;

[0032] If the F statistic exceeds the critical value of the distribution, then reject the null hypothesis that the charge imbalance index does not Granger cause the corresponding parameter, indicating that there is a causal relationship between the charge imbalance index and the corresponding parameter, and the charge imbalance index is in the position of cause in the causal relationship. If the F statistic does not exceed the critical value of the distribution, then it indicates that there is no causal relationship between the charge imbalance index and the corresponding parameter.

[0033] Preferably, according to the maximum lag order, the correlation situation is verified for different lag orders to draw a curve graph of the F statistic varying with the lag order, and according to the change trend in the curve graph, the target F statistic is found from the curve graph.

[0034] Statistically analyze the parameters that have a causal relationship with the charge imbalance index, generate a parameter set, and find the target F statistic associated with each parameter in the parameter set.

[0035] Preferably, the steps for obtaining the target F statistic include:

[0036] Take three different lag orders with an adjacent relationship in the curve graph as a comparison group, and through statistics, obtain several groups of comparison groups;

[0037] Calculate the mean and standard deviation of the F statistics corresponding to the respective lag orders in each comparison group, select the comparison group with the largest mean and the smallest standard deviation as the target group; take the F statistic with the smallest difference from its corresponding mean in the target group as the target F statistic, and take the lag order corresponding to the target F statistic as the optimal time delay for the charge imbalance index to affect the corresponding parameter;

[0038] According to the optimal time delay for the charge imbalance index to affect the corresponding parameter, obtain the time stamp for triggering the warning signal in advance when the corresponding parameter is abnormal, and within the time period of this optimal time delay, formulate corresponding treatment measures.

[0039] An intelligent water quality monitoring system, comprising:

[0040] An information acquisition module is used to acquire the water quality information and potential time series information at each position in the target water area, and through preprocessing, obtain an information set;

[0041] A region identification module is used to analyze the charge imbalance situation at each position in the target water area according to the information set to determine the imbalance hot spot area and the non - imbalance hot spot area, respectively analyze the correlation between the charge imbalance situation and each parameter in the water quality information in the imbalance hot spot area and the non - imbalance hot spot area, and through comparison, determine the coupling area;

[0042] A setting module is used to verify the correlation between the charge imbalance index and the corresponding parameter in the coupling area, obtain a parameter set, and set the time stamp for triggering the warning signal for each parameter in the parameter set respectively to formulate corresponding treatment measures.

[0043] The present invention provides an intelligent water quality monitoring method and system, having the following beneficial effects:

[0044] (1) By performing threshold segmentation on the reconstructed charge imbalance heat map, this method can accurately identify the imbalance hot spots. For example, in lake ecological monitoring, this method has successfully located the charge imbalance hot spots caused by pollutant accumulation in the bay. Compared with the traditional methods relying on manual experience or simple data statistics, this method can quickly and accurately determine the areas that need key attention, extract the coordinate set of the hot spots, provide a clear target for subsequent in-depth research and targeted treatment, and further improve the efficiency and accuracy of the monitoring work. In S2, the coordinate set of the non-hot spots is determined based on the coordinate set of the hot spots, and the two are combined to generate the coordinate set of the area. At the same time, the water quality information and charge imbalance index of the corresponding area are extracted, and the relationship between charge imbalance and various water quality parameters in different areas can be systematically compared and analyzed. By constructing the stress coupling coefficient and comparing the coefficient differences between the hot and non-hot spots, the coupling area is judged based on the influence difference value and difference threshold, and the areas where the water quality is at risk of stress can be accurately identified. For example, in the monitoring scenario of urban inland rivers, the coupling area with strong correlation between charge imbalance and parameters such as dissolved oxygen and ammonia nitrogen caused by industrial sewage discharge can be quickly located. This accurate identification makes the water quality warning no longer blind, and can issue alarms for the areas where there are real risks. The calculation of the stress coupling coefficient quantifies the degree of association between charge imbalance and water quality parameters, making the ecological stress relationship clearer and more definite. At the same time, by determining the coupling area, the key treatment areas and key influencing parameters can be clarified, avoiding the blindness of treatment measures, making the treatment resources more reasonably allocated, and improving the treatment efficiency.

[0045] (2) By constructing the lag autoregressive term of only the corresponding parameter to predict the corresponding parameter and adding the lag autoregressive term of the charge imbalance index at the same time, and using the F statistic for Granger causality test, it can accurately determine whether there is a causal relationship between the charge imbalance index and water quality parameters in the coupling area. Compared with traditional correlation analysis, it can more accurately reveal the essential connection between variables. After determining the causal relationship, a more scientific warning mechanism can be set based on this. When it is clear that the charge imbalance index is the cause of the change of some water quality parameters, when the charge imbalance index shows abnormal fluctuations, the future change trend of the corresponding water quality parameters can be accurately predicted. For example, in river monitoring, if it is verified that the charge imbalance index will cause a significant decrease in dissolved oxygen after 3 hours, then when the charge imbalance index is monitored to increase abnormally, a warning of a decrease in dissolved oxygen can be issued 3 hours in advance, and the accuracy and reliability of the warning are greatly improved. Compared with fuzzy warning, it can reserve sufficient response time for relevant departments and reduce ecological risks.

[0046] (3) According to the maximum lag order, verify the correlation for different lag orders and plot a curve graph showing the change of the F-statistic with the lag order. This step is to comprehensively observe the change in the degree of correlation between the charge imbalance index and the corresponding parameters under different time lags, and visually present the dynamic evolution of the relationship between the two over time through the trend of the curve, providing data visualization support for subsequent analysis. Generate a parameter set for the parameters that have a causal relationship with the charge imbalance index, and find the relevant target F-statistic for each parameter in the parameter set. This step aims to determine all water quality parameters affected by the charge imbalance index and find the F-statistic that can relatively significantly reflect this causal relationship, providing a key basis for subsequent determination of the optimal time delay and warning timestamp. By taking three adjacent different lag orders in the curve graph as comparison groups, calculate the mean and standard deviation of the F-statistics of each comparison group, select the comparison group with the largest mean and the smallest standard deviation as the target group, and then determine the F-statistic with the smallest difference from the mean in the target group as the target F-statistic. The corresponding lag order is the relatively better time delay for the charge imbalance index to affect the corresponding parameter. This series of operations is to accurately screen out the time points when the charge imbalance index has a significant and relatively stable impact on each parameter from among many possible time lags, so as to provide the most accurate time basis for early warning and treatment. Based on the relatively better time delay for the charge imbalance index to affect the corresponding parameter, obtain the timestamp for triggering the early warning signal in advance when the corresponding parameter is abnormal, and formulate corresponding treatment measures within this time period. This step transforms the previous analysis results into practical applications, realizes the early warning and targeted treatment of water quality anomalies, and improves the practicality and effectiveness of the water quality monitoring system. Description of the Drawings

[0047] Figure 1 Schematic flow diagram of an intelligent water quality monitoring method of the present invention;

[0048] Figure 2 Logic diagram of an intelligent water quality monitoring method of the present invention;

[0049] Figure 3 Curve graph showing the change of the F-statistic with the lag order in an intelligent water quality monitoring method of the present invention;

[0050] Figure 4 Block diagram of an intelligent water quality monitoring system of the present invention. Detailed Embodiments

[0051] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0052] Example 1

[0053] Please refer to Figures 1 to 3 , the present invention provides an intelligent water quality monitoring method, including the following steps,

[0054] S1: Obtain the water quality information and potential time series information at each position in the target water area, and after preprocessing, obtain an information set;

[0055] S2: According to the information set, analyze the charge imbalance situation at each position in the target water area to determine the imbalance hot spot area and the imbalance non-hot spot area, and respectively analyze the correlation between the charge imbalance situation and each parameter in the water quality information in the imbalance hot spot area and the imbalance non-hot spot area. After comparison, determine the coupling area;

[0056] S3: Verify the correlation between the charge imbalance index and the corresponding parameters in the coupling area, obtain a parameter set, and set a time stamp for triggering an early warning signal for each parameter in the parameter set to formulate corresponding treatment measures.

[0057] In this embodiment, by obtaining the water quality information and potential time series information at each position in the target water area through step S1 and performing preprocessing to obtain an information set, multi-dimensional data covering physical, chemical, and charge characteristics can be comprehensively collected. For example, in the monitoring of the water area near an industrial park, not only conventional water quality parameters such as pH value and dissolved oxygen are obtained, but also potential time series information is collected through a nanoelectrode array, providing a rich and accurate data basis for subsequent analysis. Compared with traditional single-index monitoring, it can more accurately reflect the true water quality situation.

[0058] Based on the information set obtained in step S1, step S2 analyzes the charge imbalance situation, determines the imbalance hot spot area and the non-hot spot area, and studies the correlation between the charge imbalance and each water quality parameter to determine the coupling area. This analysis can accurately locate the areas that have a significant impact on the ecosystem. For example, during the monitoring of a certain lake, this step successfully identified the bay as the charge imbalance hot spot area, and it was found that there was a strong correlation between the charge imbalance in this area and the growth of algae (characterized by the chlorophyll-a content), that is, the coupling area, indicating the direction for subsequent targeted monitoring and treatment, and further avoiding blindness.

[0059] In step S3, the correlation between the charge imbalance index and corresponding parameters in the coupling area is verified to obtain a parameter set, set a warning timestamp, and formulate control measures. For example, in the monitoring of a certain river, it is determined that parameters such as dissolved oxygen and ammonia nitrogen have a causal relationship with charge imbalance and are included in the parameter set. According to the analysis, the dissolved oxygen will significantly decrease about 6 hours after charge imbalance, so a timestamp is set to trigger a warning signal 5 hours in advance. Once the monitoring data reaches the warning condition, the system automatically starts the oxygenation equipment and adjusts the discharge strategy of the sewage treatment plant, achieving scientific warning and precise control. Compared with traditional lagged control, the control efficiency is improved and the risk of ecological damage is reduced.

[0060] In summary, step S2 is based on S1. Using the information set obtained in S1, it deeply analyzes the charge imbalance situation in the target water area, determines the imbalance hotspots and non-hotspot areas, and further explores the correlation between charge imbalance and water quality parameters, screening out the coupling area. This step is equivalent to sifting through a vast amount of data to find the areas and factors that are crucial for water quality and the ecosystem, determining the targets for subsequent precise research and control. Step S3 then deeply verifies and applies the coupling area determined in S2. By verifying the correlation, a parameter set is obtained, a warning timestamp and control measures are set. This step is the key to transforming the previous analysis results into practical applications, realizing a closed-loop from monitoring, analysis to warning and control, ensuring that the entire water quality monitoring system can effectively play its role and timely respond to problems brought about by water quality changes.

[0061] Embodiment 2

[0062] Please refer to Figure 1 ., specifically: Sensor group nodes and multiple groups of nanoelectrode arrays are arranged in the target water area according to the specified grid size and installed on a floating ball bracket with adjustable depth to ensure sampling at three layers of 0.5m, 1.5m, and 2.5m. Each node is equipped with a microfiltration flow cell. After removing large particle impurities, the potential signal is transmitted to the edge computing platform in real time through optical fiber or wireless link; the system regularly calibrates the sensitivity coefficient automatically using standard solutions. The array deployment not only achieves spatial redundancy and improves data reliability.

[0063] Obtain the water quality information and potential time series information at each location in the target water area; the potential time series represents the position of each electrode array in the target water area and the potential value at the sampling moment;

[0064] Among them, the grid size can be set to a 5-meter by 5-meter grid;

[0065] Each electrode array contains multiple microelectrodes that can simultaneously measure the local current and potential. Through such a deployment method, the distribution signals of charged particles at different positions and micro-scales in the water body can be obtained, forming a comprehensive perception of the charge distribution in the water body.

[0066] Three-point calibration is carried out using standard solutions (known ion concentrations). The electrodes are placed in three different concentrations of standard solutions respectively, and the corresponding potential values are measured. By recording the potential change amounts and ion concentration change amounts under different standard solutions, the sensitivity coefficients of each electrode are calculated. The sensitivity coefficient reflects the response ability of the electrode to the change in ion concentration and is used to convert the measured potential signal into the actual charge concentration.

[0067] The way to obtain the sensitivity coefficient is as follows: By calculating the ratio of the potential change amount to the ion concentration change amount of the corresponding electrode under different standard solutions, and after multiple verifications, the average value of multiple groups of calculation results is taken to obtain the sensitivity coefficient. Through the results of multiple calibrations, the influence of single measurement error is reduced, making the finally obtained sensitivity value more representative and stable.

[0068] By using wavelet threshold denoising and baseline drift correction on the water quality information and potential time series information, a processed information set is obtained. According to the sensitivity coefficients of each electrode, the corresponding potential values in the information set are converted into charge concentrations, and after normalization, charge time series information is obtained. That is to say, the corresponding potential signal is converted into a physical quantity that can directly reflect the charge distribution in the water body, which is convenient for subsequent analysis and processing.

[0069] By converting the corresponding potential values into charge concentrations, a physical quantity that can directly reflect the charge distribution in the water body is obtained, which is convenient for subsequent analysis and processing.

[0070] Wavelet transform can decompose the signal into different frequency scales. By setting a threshold, the wavelet coefficients corresponding to high-frequency noise are set to zero or subjected to shrinkage processing, and then the signal is reconstructed through inverse wavelet transform, thereby removing high-frequency interference. For example, high-frequency noise such as electromagnetic interference that may exist in the water body can be effectively removed by this method, making the potential signal smoother and more accurate.

[0071] According to the charge time series information, the average value and standard deviation of the charge concentrations of all electrodes are calculated. Then, through the average value and standard deviation of the charge concentrations, the charge imbalance index of each electrode array is calculated, expressed as: , is the charge imbalance index of the corresponding electrode array at time t, is the standard deviation of the charge concentration of the corresponding electrode array at time t, is the average value of the charge concentration of the corresponding electrode array at time t, is a positive number, a very small positive number (such as ), which is used to ensure that the denominator is not zero and ensure the stability of numerical calculation;

[0072] The charge imbalance index is used to quantify the degree of charge distribution imbalance in the area covered by the entire electrode array. Among them, the aforementioned standard deviation of charge concentration reflects the degree of dispersion of charge concentration. The larger the standard deviation, the greater the difference in charge concentration between different electrodes. The average value represents the average level of charge concentration in this area. By dividing the standard deviation by the average value, the obtained charge imbalance index can comprehensively reflect the degree of deviation of charge distribution from the uniform state. The larger the index value, the more serious the charge distribution imbalance; the smaller the index value, the relatively more balanced the charge distribution.

[0073] Specifically, the charge imbalance index can reflect the changes in processes such as ion migration and material transformation in water bodies. For example, when the water body is polluted, the ionic components of pollutants will change the original ion distribution in the water body, resulting in a change in the charge imbalance index. By monitoring this index, it is possible to indirectly understand whether the water body is polluted and the degree of pollution.

[0074] Through the charge imbalance index of the corresponding electrode array at time t, a time-varying index sequence of the corresponding electrode array is obtained. The time-varying index sequence is used to record the changes in the charge imbalance index of the corresponding electrode array at different times. Through this sequence, the evolution trend of the degree of charge distribution imbalance over time can be observed.

[0075] In this embodiment, a standard solution is used for three-point calibration to obtain the sensitivity coefficient of each electrode, ensuring the accuracy of electrode measurement. For example, when monitoring the water area near an industrial wastewater discharge outlet, the accurate sensitivity coefficient can accurately convert the potential signal into charge concentration, enabling the monitoring data to truly reflect the charge distribution of the water body and providing a reliable basis for subsequent analysis.

[0076] The node is built-in with a microfiltration flow cell to remove large particle impurities. Combining preprocessing means such as wavelet threshold denoising and baseline drift correction effectively improves the reliability and stability of the data. Taking river monitoring as an example, large particle impurities such as sediment in the water will interfere with electrode measurement, and the microfiltration flow cell can filter impurities in advance; the signal noise and baseline drift caused by environmental factors are eliminated after wavelet threshold denoising and baseline drift correction, making the finally obtained charge time series information more accurate.

[0077] In addition, the system automatically calibrates the sensitivity coefficient using a standard solution at regular intervals, avoiding measurement errors caused by the degradation of electrode performance after long-term use and ensuring the stability and consistency of long-term monitoring data. Based on the processed charge time series information, the charge imbalance index and the time-varying index sequence are calculated, which can quantify the degree of charge distribution imbalance and monitor its dynamic changes. For example, in the monitoring of urban landscape lakes, by analyzing the time-varying index sequence, the evolution trend of the charge imbalance degree of the lake water over time can be grasped in real time. If it is found that the charge imbalance index continues to rise during a certain period, combined with other water quality parameters, it can be timely judged whether there is an increase in pollution or an abnormality in the ecosystem, providing key data support for water quality early warning and treatment.

[0078] Meanwhile, the spatial redundancy brought by the array deployment further improves the reliability of the data. Even if some nodes fail, the continuity and integrity of the monitoring can still be ensured through the data of other nodes.

[0079] Example 3

[0080] Please refer to Figure 1 and Figure 2 , specifically: collect the charge imbalance index corresponding to the position coordinates of each electrode array, and use Kriging interpolation to reconstruct the charge imbalance index on the grid of the entire target water area. Specifically: according to the grid range of the entire target water area, determine the points to be interpolated. According to the spatial correlation between the position of each point to be interpolated and the position coordinates of each collected electrode array, obtain the charge imbalance index of the point to be interpolated, expressed as: , is the charge imbalance index at time t for the corresponding point to be interpolated, i is the number of the electrode array to be collected, n is the total number of the electrode arrays to be collected, is the charge imbalance index of the i-th electrode array at time t, is the weight coefficient of the i-th electrode array, which reflects the degree of spatial correlation between the point to be interpolated and the known points;

[0081] Through this method, the discrete CDII values of the electrode arrays can be made continuous on the entire water area grid, and the estimated value of the charge distribution imbalance index at each position at time t can be obtained.

[0082] After reconstruction, obtain the charge imbalance heat map for each moment, and perform threshold segmentation on the charge imbalance heat map;

[0083] Compare the charge imbalance index at each position in the charge imbalance heat map with a preset threshold. If the charge imbalance index exceeds the threshold, mark the corresponding position as an imbalance hot spot area, otherwise mark the corresponding position as an imbalance non-hot spot area;

[0084] Extract the coordinates of the imbalance hot spot areas to obtain the hot spot area coordinate set.

[0085] Through spatiotemporal heat map reconstruction and hotspot identification, the distribution of charge imbalance in space and time is intuitively displayed, and the areas with high imbalance risk are located, providing information in the spatial dimension for subsequent assessment of ecological impacts.

[0086] In this embodiment, the discrete electrode array charge imbalance index is reconstructed on the entire target water area grid through Kriging interpolation, which can convert the data of limited sampling points into continuous spatial distribution data. For example, in the monitoring of a river with a length of 10 kilometers and a width of 500 meters, only partial point data can be obtained relying on discrete electrode arrays. After interpolation reconstruction, the estimated values of the charge imbalance index for each grid point on the entire river plane can be obtained, forming a continuous charge imbalance heat map. This visual presentation method enables researchers and managers to intuitively understand the spatial distribution of charge imbalance, and it is easier to discover potential water quality problem areas compared with discrete data.

[0087] Threshold segmentation is performed on the reconstructed charge imbalance heat map. Based on the preset threshold, such as the outlier boundary determined through historical data statistical analysis, the charge imbalance hotspot areas can be accurately identified. For example, in the monitoring of a certain lake ecological reserve, a charge imbalance hotspot area caused by sewage discharge in the center of the lake is successfully located through this method. The charge imbalance index in this area far exceeds the threshold, providing accurate location information for the environmental protection department to take targeted treatment measures in a timely manner.

[0088] Extracting the coordinate set of the imbalance hotspot area provides a clear research object for subsequent in-depth analysis of the correlation between charge imbalance and water quality parameters, ecological stress impacts, etc. For example, when studying the impact of charge imbalance on aquatic biological communities, biological samples can be collected for the hotspot area and non-hotspot area respectively, and comparative analysis can be carried out in combination with charge imbalance data, making the research more targeted and scientific. It avoids blind research in the entire water area range, effectively saving manpower, material resources and time costs, and at the same time improving the reliability and application value of research results.

[0089] Example 4

[0090] Please refer to Figure 1 and Figure 2 , specifically: based on the coordinate set of the hotspot area, determine the coordinate set of the non-hotspot area, and extract the water quality information of the coordinate sets of the hotspot area and the non-hotspot area from the information aggregation, and extract the charge imbalance indexes of the coordinate sets of the hotspot area and the non-hotspot area from the time-varying index sequence;

[0091] Combine the coordinate set of the hotspot area and the coordinate set of the non-hotspot area to generate a regional coordinate set;

[0092] In the hot spot area coordinate set and the non-hot spot area coordinate set, the correlation coefficients of the charge imbalance index and each parameter in the water quality information are respectively counted. Specifically: taking the time-varying index sequence and each parameter in the water quality information at the corresponding time as inputs and substituting them into the correlation coefficient calculation function. The formula is: , where is the correlation coefficient between the charge imbalance index and the corresponding parameter, t is the time number, m is the duration in the time-varying index sequence, is the mean value of the charge imbalance index in the time-varying index sequence, is the value of the corresponding parameter in the water quality information at time t, is the mean value of the corresponding parameter;

[0093] For the aforementioned correlation coefficient between the charge imbalance index and the corresponding parameter, where the corresponding parameter refers to various water quality parameters in the water quality information. Specifically, each parameter in the water quality information includes but is not limited to pH value, dissolved oxygen, and plankton count;

[0094] The correlation coefficient between the charge imbalance index and the corresponding parameter is a statistical indicator used to measure the closeness of the linear relationship between the two. Specifically: if the correlation coefficient is greater than 0, it indicates that there is a positive correlation between the charge imbalance index and the corresponding parameter, that is, when the charge imbalance index increases, the corresponding parameter also tends to increase; when the charge imbalance index decreases, the corresponding parameter also tends to decrease. The closer the correlation coefficient is to 1, the stronger the positive correlation between the two.

[0095] When the correlation coefficient is less than 0, it means that there is a negative correlation between the charge imbalance index and the corresponding parameter, that is, when the charge imbalance index increases, the corresponding parameter tends to decrease; when the charge imbalance index decreases, the corresponding parameter tends to increase. The closer the correlation relationship is to -1, the stronger the negative correlation between the two.

[0096] If the correlation coefficient is close to 0, it indicates that there is no obvious linear correlation between the charge imbalance index and the corresponding parameter, and the changes of the two are independent of each other without an obvious pattern to follow.

[0097] By calculating the correlation coefficients of the charge imbalance index and each parameter in the water quality information, it can help analyze the connection between the charge imbalance phenomenon and the changes of water quality parameters, providing an important basis for further studying the potential relationship between the two and judging whether there is stress coupling.

[0098] Among them, the pH value directly measures the acidity and alkalinity of the water body through a pH meter. This parameter reflects the acid-base balance state of the water body and has an important impact on the metabolic activity, enzyme activity, and cell membrane permeability of microorganisms. Different microorganisms have different suitable pH ranges for growth. Dissolved oxygen is measured by the electrochemical probe method or the iodometric method. Dissolved oxygen is an essential substance for the respiration of aquatic organisms, and its content directly affects the metabolic activities of aerobic microorganisms in the water body and the composition and distribution of the aquatic biological community. The plankton count is obtained by regularly collecting water samples and classifying and counting the plankton (including phytoplankton and zooplankton) in them. As an important part of the water ecosystem, the species composition, quantity change, and community structure of plankton can intuitively reflect the health status and stability of the ecosystem and are extremely sensitive to environmental changes.

[0099] After taking the absolute value of the correlation coefficients between the charge imbalance index and all parameters in the water quality information and then calculating the mean value, a stress coupling coefficient is constructed. The formula is: , is the stress coupling coefficient of the corresponding regional coordinate set, G is the total number of correlation coefficients within the corresponding regional coordinate set, g is the coordinate number within the corresponding regional coordinate set, K is the total number of parameters, and k is the parameter number. is the correlation coefficient between the charge imbalance index at coordinate g within the corresponding regional coordinate set and the k-th parameter B.

[0100] In this way, regardless of whether the charge imbalance index is positively or negatively correlated with each parameter, as long as the correlation is significant, it will be reflected in the stress coupling coefficient. The larger the stress coupling coefficient, the stronger the coupling degree between the charge imbalance and the ecological parameter perturbation, that is, the more significant the impact of the unbalanced charge distribution on the ecosystem; conversely, the smaller the stress coupling coefficient, the weaker the correlation between the two, and the relatively smaller the impact of the charge imbalance on the ecosystem.

[0101] The stress coupling coefficient is an index used to comprehensively measure the tightness of the correlation between the charge imbalance index and each parameter in the water quality information. Its construction is based on the correlation coefficients between the charge imbalance index and each parameter in the water quality information. The stress coupling coefficient can comprehensively consider the relationship between the charge imbalance index and multiple water quality parameters, rather than just a single parameter. By processing all the correlation coefficients, a value that can reflect the overall coupling degree is obtained, so as to comprehensively evaluate the comprehensive impact of the charge imbalance phenomenon on the water quality. Of course, it can also be used to compare the stress coupling degrees of different regional coordinate sets. For example, when comparing the stress coupling coefficients of the hot spot regional coordinate set and the non-hot spot regional coordinate set, it can be judged which region has a closer relationship between the charge imbalance index and the water quality parameters, and then analyze the differences in the impact of charge imbalance on the water quality in different regions.

[0102] Determine the stress coupling coefficients in the hot spot area coordinate set and the non-hot spot area coordinate set respectively. After performing subtraction calculation on the stress coupling coefficients in the hot spot area coordinate set and the non-hot spot area coordinate set to obtain the influence difference value, if the influence difference value exceeds the pre-set difference threshold, mark the unbalanced hot spot area as the coupling area, otherwise it is not the coupling area;

[0103] The coupling area is used to indicate the stress risk faced by the water quality in the corresponding area.

[0104] After determining the coupling area through the stress coupling coefficient, the relationship between charge imbalance and water quality parameters in these areas can be studied in depth, providing a scientific basis for formulating more effective water quality governance and protection measures. For example, if it is found that the stress coupling coefficient in a certain area is high, it means that the charge imbalance in this area has a greater impact on water quality, and it is necessary to focus on and take measures to adjust the charge distribution to improve the water quality status.

[0105] In this embodiment, based on the determined hot spot area coordinate set, the non-hot spot area coordinate set is determined, and the water quality information of the corresponding area is extracted from the information set. At the same time, the corresponding charge imbalance index is extracted from the time-varying index sequence. Finally, the hot spot and non-hot spot area coordinate sets are combined into a regional coordinate set. This step mainly provides structured data for subsequent analysis, integrating the key data of different regions together. Compared with the existing water quality monitoring methods, the existing water quality monitoring often only focuses on the overall water quality status or simply analyzes a single area, lacking systematic integration and comparison of data between different regions. By clearly dividing the hot spot and non-hot spot areas and integrating relevant data, this step can analyze the differences in water quality characteristics of different regions more carefully, laying a foundation for in-depth study of the relationship between charge imbalance and water quality parameters.

[0106] The hot spot area coordinate set is a set of coordinates of areas with a relatively high degree of charge imbalance determined through previous analysis. These areas are the key objects of concern in the study of water quality problems.

[0107] The non-hot spot area coordinate set is a set of coordinates of areas with a relatively low degree of charge imbalance, opposite to the hot spot areas, serving as a comparison reference area.

[0108] The regional coordinate set is a set obtained by combining the hot spot area coordinate set and the non-hot spot area coordinate set, covering the coordinate information of all areas to be analyzed in the target water area.

[0109] In the hot spot area coordinate set and the non - hot spot area coordinate set, the correlation coefficients of the charge imbalance index and each parameter in the water quality information are respectively counted. Through the correlation coefficient calculation function, the time - varying index sequence (reflecting the change of the charge imbalance index over time) and the water quality information parameters at the corresponding time are used as inputs to calculate the correlation coefficient between the two, so as to measure the linear correlation degree between the charge imbalance index and each water quality parameter. Compared with the prior art, the prior art usually only analyzes the water quality parameters separately or simply observes the superficial relationship between the parameters, making it difficult to accurately quantify the correlation degree between charge imbalance and other water quality parameters. This step can scientifically reveal the potential relationship between charge imbalance and each water quality parameter by accurately calculating the correlation coefficient, providing a quantitative basis for further analysis.

[0110] The correlation coefficient is an index used to measure the linear correlation degree between two variables, with a value range between - 1 and 1. The closer its absolute value is to 1, the stronger the linear correlation between the two variables; approaching 0 indicates a weak linear correlation. A positive correlation means that the change trends of the two variables are the same, and a negative correlation means that the change trends are opposite.

[0111] The time - varying index sequence is a sequence that records the change of the charge imbalance index of the electrode array at different times. Through this sequence, the evolution trend of the charge distribution imbalance degree over time can be observed, providing data for analyzing the relationship between charge imbalance and other parameters in the time dimension.

[0112] After taking the absolute values of the correlation coefficients between the charge imbalance index and all parameters in the water quality information and calculating the mean value, a stress coupling coefficient is constructed. This coefficient comprehensively reflects the overall correlation degree between the charge imbalance and each parameter in the water quality information, and is used to evaluate the stress degree of the charge imbalance on the water ecosystem. Compared with the prior art, the prior art lacks a quantitative assessment of the comprehensive correlation between multiple water quality parameters and charge imbalance, making it difficult to comprehensively evaluate the stress on the water ecosystem. This step can measure the coupling degree between charge imbalance and water quality parameters as a whole by constructing the stress coupling coefficient, providing a more comprehensive index for judging the water quality ecological risk. For example: in lake monitoring, for a certain area, after calculating the correlation coefficients between the charge imbalance index and multiple water quality parameters such as pH, dissolved oxygen, ammonia nitrogen, etc., the stress coupling coefficient is constructed. If the coefficient value is large, it indicates that there is a strong correlation between the charge imbalance and multiple water quality parameters in this area, meaning that the water quality ecosystem in this area is highly stressed by the charge imbalance and needs to be focused on.

[0113] The stress coupling coefficient is a comprehensive index obtained by processing the correlation coefficients between the charge imbalance index and each parameter in the water quality information. The larger its value, the stronger the coupling degree between the charge imbalance and the ecological parameter disturbance, that is, the more significant the impact of the charge distribution imbalance on the ecosystem, and it is used to quantitatively evaluate the stress degree on the water ecosystem.

[0114] Determine the stress coupling coefficients in the hot spot area coordinate set and the non-hot spot area coordinate set respectively, and calculate the difference between the two to obtain the impact difference value. Compare the impact difference value with a preset difference threshold. If it exceeds the threshold, mark the unbalanced hot spot area as the coupling area, indicating that the water quality in this area faces a relatively high stress risk. Compared with the effects of the prior art, it is difficult for the prior art to accurately identify which areas have a more significant impact on the water quality ecosystem due to charge imbalance, and it is difficult to achieve targeted monitoring and governance. This step provides a clear target for subsequent effective measures by locking high-risk areas. For example: In the monitoring of a certain reservoir, by calculating the stress coupling coefficients of the hot spot area and the non-hot spot area, it is found that the difference between the two is large and exceeds the set threshold, so the hot spot area is marked as the coupling area, which prompts the management personnel that due to the strong coupling relationship between charge imbalance and water quality parameters in this area, it faces a relatively high ecological risk and requires strengthened monitoring and timely treatment measures, such as adjusting the water flow or adding purification substances, etc.

[0115] The coupling area is the area determined by comparing the stress coupling coefficients of the hot spot area and the non-hot spot area. There is a strong correlation between charge imbalance and water quality parameters in this area, indicating that the water quality in the corresponding area faces a stress risk and is an area that needs to be focused on and treated. The impact difference value is the difference between the stress coupling coefficients in the hot spot area coordinate set and the non-hot spot area coordinate set, which is used to measure the difference in the degree of influence of charge imbalance on water quality parameters between the hot spot area and the non-hot spot area, and is an important basis for judging the coupling area.

[0116] Example 5

[0117] Please refer to Figure 1 and Figure 2 Specifically: According to the determined coupling area, verify the correlation between the charge imbalance index and the corresponding parameters in the coupling area, specifically including:

[0118] Set the maximum lag order. For example, according to the sampling frequency and ecological response characteristics, the maximum lag order can be set to 12 or 24;

[0119] Based on the maximum lag order, construct a lag autoregressive term that only uses the corresponding parameter to predict the corresponding parameter and a lag autoregressive term that simultaneously adds the charge imbalance index, specifically expressed as:

[0120] Using only the lag autoregressive term of the corresponding parameter to predict the corresponding parameter: ;

[0121] Simultaneously adding the lag autoregressive term of the charge imbalance index:

[0122] ;

[0123] In the formula, P is the maximum lag order, that is, the maximum time step length to be traced back, p is the lag index, ranging from 1 to P, representing the p-th moment forward, is the autoregressive coefficient, representing the parameter at the p-th moment forward The value is the linear influence weight of the parameter B value at the current moment t, is the first residual term (prediction error), reflecting the remaining unexplained part when predicting only using the lag terms of the corresponding parameter B itself, is the regression coefficient of the charge imbalance index at time t - p, representing the influence weight of the charge imbalance index at the p-th moment forward on the charge imbalance index at the current moment, is the charge imbalance index at time t - p, used to test the additional explanatory power of charge perturbation on the corresponding parameter, is the second residual term, reflecting the part that remains unexplained after considering both the corresponding parameter and the lag terms of the charge imbalance index, is the value of the corresponding parameter at time t - p, which is the autoregressive input to the current value of the corresponding parameter;

[0124] Among them, the lag autoregressive term is a content in time series analysis used to describe the relationship between the past values and the current value of a variable itself. By considering the lag values of the variable, the lag autoregressive term can capture the dynamic change law in time series data. For the corresponding parameter, it can describe the dependence relationship between the parameter itself at different time points, helping to analyze the evolution trend of the parameter over time. For example, if the autoregressive coefficient of the lag autoregressive term of a water quality parameter (such as pH) is large, it indicates that the current value of this parameter is largely affected by the past value, that is, it has a strong time series correlation.

[0125] When adding the lag autoregressive term of the charge imbalance index, the influence of the charge imbalance index on the corresponding parameter can be further explored. By analyzing the magnitude and significance of, the direction and degree of the role of the charge imbalance index on the corresponding parameter at different lag orders can be understood. If at a certain lag order is significantly non-zero, it indicates that the charge imbalance index at that moment has a significant influence on the current corresponding parameter, thus revealing the potential causal relationship between the charge imbalance and the corresponding parameter.

[0126] Calculate the sum of squared residuals respectively to obtain the sum of squared residuals of the first residual term and the sum of squared residuals of the second residual term;

[0127] Based on the sum of squared residuals of the first residual term and the sum of squared residuals of the second residual term, construct the F statistic, and the formula is: , F is the F statistic, is the sum of squared residuals of the first residual term, is the sum of squared residuals of the second residual term, is the sample size, which refers to the total number of time series data points used to establish the autoregressive terms with only the corresponding parameters for prediction and when adding the autoregressive terms of the charge imbalance index simultaneously. The is the adjustment term for degrees of freedom. The role of this adjustment term is to adjust the sample size and model complexity to more accurately reflect the distribution characteristics of the statistic;

[0128] Among them, the numerator part reflects the degree of improvement in the model's explanatory ability after adding the lag term of the charge imbalance index (i.e., the reduction in the sum of squared residuals). Dividing by P is to eliminate the influence of the difference in the number of parameters. The denominator part represents the average residual including both the parameter and the lag term of the charge imbalance index, that is, the average residual of the unrestricted model, which is used to standardize the change in the numerator;

[0129] The F-statistic is used to test whether the charge imbalance index has a significant predictive ability for the corresponding parameter;

[0130] If the F-statistic exceeds the critical value of the distribution, then the null hypothesis that the charge imbalance index does not Granger cause the corresponding parameter is rejected, indicating that there is a causal relationship between the charge imbalance index and the corresponding parameter, and the charge imbalance index is in the position of cause in the causal relationship. If the F-statistic does not exceed the critical value of the distribution, it indicates that there is no causal relationship between the charge imbalance index and the corresponding parameter.

[0131] The critical value of the distribution can be obtained in the following way: First, clarify the degrees of freedom. Here, the F-distribution has two degrees of freedom. The first degree of freedom is P; the second degree of freedom is T - 2p - 1, where T is the sample size. Then, according to the given significance level, such as a significance level of 0.05, look up the corresponding critical value in the F-distribution table. The F-distribution table usually lists the critical values corresponding to different significance levels for different combinations of degrees of freedom. First, find the row and column corresponding to the degrees of freedom P and T - 2p - 1, and the value at their intersection is the critical value of the F-distribution at this significance level.

[0132] Among them, is an F-distribution, where P and T - 2p - 1 are the two degrees of freedom of this distribution;

[0133] In the field of water quality monitoring, traditional methods mostly focus on the correlation analysis between water quality indicators, but correlation is not equivalent to causation. By clarifying the causal relationship and response lag time between the charge imbalance index and each parameter, it is of great significance for early warning of ecological stress and optimizing treatment strategies.

[0134] Of course, the above steps can be repeated, but the roles of the charge imbalance index and the corresponding parameters are swapped, that is, to check whether the corresponding parameter is in the causal position in the causal relationship. If so, it is a bidirectional coupling.

[0135] In this embodiment, the maximum lag order is set according to the sampling frequency. The purpose is to determine the maximum time step of the retrospective data and delimit the time range for subsequent model construction. On this basis, a lag autoregressive term prediction using only the corresponding parameter and a lag autoregressive term with the charge imbalance index added are constructed. The former is used to reflect the influence of the parameter's own historical data on the current value, and the latter explores whether the charge imbalance index can additionally explain the parameter change. For example, when monitoring the dissolved oxygen in a river, if the sampling frequency is once per hour and the maximum lag order is set to 6, the restricted model predicts the current river dissolved oxygen only using the river dissolved oxygen in the past 6 hours, and the unrestricted model adds the contemporaneous charge imbalance index to analyze its influence on the prediction of river dissolved oxygen.

[0136] The sum of squared residuals of the two models is calculated respectively. Its function is to measure the deviation degree between the model prediction value and the actual value. The smaller the value, the better the model fitting effect. In the above example of river dissolved oxygen monitoring, calculating the sum of squared residuals in the prediction model using only the lag autoregressive term of the corresponding parameter can evaluate the prediction accuracy based only on the historical data of dissolved oxygen, and the sum of squared residuals in the model with the lag autoregressive term of the charge imbalance index added reflects the fitting effect of the model after adding the charge imbalance index.

[0137] The sum of squared residuals is an important basic data for constructing the F statistic subsequently. The F statistic is constructed based on the sum of squared residuals of the two models and is used to test whether the charge imbalance index has a significant influence on the corresponding parameter. Comparing the F statistic with the critical value, if the F statistic exceeds the critical value, the original hypothesis is rejected, indicating that there is a causal relationship between the charge imbalance index and the corresponding parameter, that is, the charge imbalance will cause the change of this parameter; otherwise, there is no causal relationship. For example, in the monitoring of the chlorophyll-a content in a lake, if the calculated F statistic is significant, it means that the charge imbalance index is one of the reasons affecting the change of chlorophyll-a content. The F statistic is used to quantify the significance of the causal relationship, and the original hypothesis is the initial hypothesis condition for the test.

[0138] Compared with the prior art, this method scientifically verifies the causal relationship between the charge imbalance index and water quality parameters from the time series dimension through rigorous hypothesis testing and model comparison, avoiding misjudgment caused by relying solely on correlation analysis. For example, traditional methods may find that the pH value of a certain water area is related to the charge imbalance index, but cannot determine the causal direction; while this method can clearly judge whether the charge imbalance affects the pH value or the change of the pH value causes the charge imbalance, providing a more accurate basis for water quality governance and improving the scientificity and effectiveness of water quality monitoring and governance.

[0139] Example 6

[0140] Please refer to Figure 1 and Figure 3 , specifically: according to the maximum lag order, verify the correlation for different lag orders to plot a curve of the F-statistic varying with the lag order, and find the target F-statistic from the curve based on the changing trend in the curve.

[0141] Statistically analyze the parameters that have a causal relationship with the charge imbalance index to generate a parameter set, and find the target F-statistic associated with each parameter in the parameter set.

[0142] The steps for obtaining the target F-statistic include:

[0143] Take three different lag orders with an adjacent relationship in the curve as a comparison group, and through statistics, obtain several groups of comparison groups;

[0144] Calculate the mean and standard deviation of the F-statistics corresponding to the respective lag orders in each comparison group, and select the comparison group with the largest mean and the smallest standard deviation as the target group; take the F-statistic with the smallest difference from its corresponding mean in the target group as the target F-statistic, and take the lag order corresponding to the target F-statistic as the optimal time delay for the charge imbalance index to affect the corresponding parameter;

[0145] Based on the optimal time delay for the charge imbalance index to affect the corresponding parameter, obtain the time stamp for triggering the early warning signal in advance when the corresponding parameter is abnormal, and formulate corresponding control measures within the time period of this optimal time delay.

[0146] According to the lag length, that is, within the time period of this optimal time delay, formulate and deploy control strategies such as oxygenation and drug administration in advance to achieve early intervention in the change of microbial activity.

[0147] Please refer to Figure 3 , in Figure 3 , the lag orders from lag order 1 to lag order 7 are all the number of lag days. By taking three different lag orders with an adjacent relationship in the curve as a comparison group, it can be seen that in Figure 3 , the lag orders from lag order 1 to lag order 3 form a comparison group, the lag orders from lag order 2 to lag order 4 form a comparison group, the lag orders from lag order 3 to lag order 5 form a comparison group, and so on. It can be known that there are five comparison groups. By calculating the mean and standard deviation of the F-statistics corresponding to the respective lag orders in each comparison group, and selecting the comparison group with the largest mean and the smallest standard deviation as the target group, as can be seen from Figure 3 it can be seen that Figure 3 the target group in is the lag orders from lag order 5 to lag order 7. Then take the F-statistic with the smallest difference from its corresponding mean in the target group as the target F-statistic, and take the lag order corresponding to the target F-statistic as the optimal time delay for the charge imbalance index to affect the corresponding parameter.

[0148] In this embodiment, by verifying different lag orders, the correlation between the charge imbalance index and the corresponding parameters at different time delays is comprehensively analyzed. The curve graph of the F statistic varying with the lag order can visually present the changing trend of this relationship. For example, when studying the relationship between the dissolved oxygen parameter in water quality and the charge imbalance index, different lag orders correspond to different days. Through verification, it can be understood how the charge imbalance index a few days ago affects the current dissolved oxygen.

[0149] Statistically analyze the parameters that have a causal relationship with the charge imbalance index to form a parameter set, and clarify which parameters are substantially related to the charge imbalance index, making the subsequent analysis more targeted. For example, among multiple water quality parameters, it is determined that the acidity, dissolved oxygen, etc. have a causal relationship with the charge imbalance index and include them in the parameter set.

[0150] Take three adjacent different lag orders in the curve graph as a comparison group, and calculate the mean and standard deviation of the F statistic in each comparison group. The mean reflects the average level of the F statistic in this group, and the standard deviation reflects the degree of data dispersion. For example, there are multiple comparison groups of three lag orders, and calculate the mean and standard deviation of their F statistics. If the mean of a certain group is large and the standard deviation is small, it means that the overall F statistic of this group is relatively high and relatively stable. Select the comparison group with the largest mean and the smallest standard deviation as the target group, and select the F statistic with the smallest difference from the mean as the target F statistic. A relatively representative and stable F statistic can be found, corresponding to the situation where the charge imbalance index has the most significant impact on the corresponding parameter.

[0151] The lag order corresponding to the target F statistic is the relatively optimal time delay for the charge imbalance index to affect the corresponding parameter, clarifying the specific time interval for the parameter to be affected by the charge imbalance index, providing a key basis for early warning and treatment. For example, if the relatively optimal time delay is 4 days, it means that the abnormality of the corresponding parameter can be predicted 4 days in advance according to the change of the charge imbalance index. According to the relatively optimal time delay, the time stamp for triggering the early warning signal in advance for the abnormality of the corresponding parameter can be obtained, allowing the manager to have enough time to take countermeasures. At the same time, formulating corresponding treatment means within this time period can specifically solve the possible water quality problems, improving the scientificity and effectiveness of water quality management. Compared with the prior art, this method can more accurately determine the causal relationship between the charge imbalance index and water quality parameters and the optimal impact time delay, providing a more accurate basis for water quality monitoring and treatment, helping with early warning and taking effective measures, and reducing the harm caused by water quality deterioration.

[0152] Example 7

[0153] Please refer to Figure 4 , specifically: An intelligent water quality monitoring system, including:

[0154] The information acquisition module is used to acquire water quality information and potential time series information at each position in the target water area. After preprocessing, an information set is obtained.

[0155] The area recognition module is used to analyze the charge imbalance situation at each position in the target water area according to the information set to determine the imbalance hot spot area and the non - imbalance hot spot area, and respectively analyze the correlation between the charge imbalance situation and each parameter in the water quality information in the imbalance hot spot area and the non - imbalance hot spot area. After comparison, the coupling area is determined.

[0156] The setting module is used to verify the correlation between the charge imbalance index and the corresponding parameters in the coupling area, obtain a parameter set, and respectively set the time stamps for triggering warning signals for each parameter in the parameter set to formulate corresponding treatment measures.

[0157] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent water quality monitoring method, characterized in that: including the following steps, S1: Obtain the water quality information and potential time series information at each location in the target water area. After preprocessing, obtain the information set; S2: According to the information set, analyze the charge imbalance situation at each location in the target water area to determine the imbalance hot spot area and the non-hot spot area of imbalance. Analyze the correlation between the charge imbalance situation and each parameter in the water quality information in the imbalance hot spot area and the non-hot spot area of imbalance respectively. After comparison, determine the coupling area; According to the charge timing information, calculate the average value and standard deviation of the charge concentration of all electrodes, and then calculate the charge imbalance index of each electrode array through the average value and standard deviation of the charge concentration, which is expressed as: , is the charge imbalance index of the corresponding electrode array at time t, is the standard deviation of the charge concentration of the corresponding electrode array at time t, is the average value of the charge concentration of the corresponding electrode array at time t, is a positive number; Obtain the time-varying index sequence of the corresponding electrode array through the charge imbalance index of the corresponding electrode array at time t. The time-varying index sequence is used to record the change of the charge imbalance index of the corresponding electrode array at different times; Extract the coordinates of the imbalance hot spot area to obtain the hot spot area coordinate set; Based on the hot spot area coordinate set, determine the non-hot spot area coordinate set, and extract the water quality information of the hot spot area coordinate set and the non-hot spot area coordinate set from the information aggregation, and extract the charge imbalance index of the hot spot area coordinate set and the non-hot spot area coordinate set from the time-varying index sequence; Combine the hot spot area coordinate set and the non-hot spot area coordinate set to generate the area coordinate set; In the hot spot area coordinate set and the non-hot spot area coordinate set, the correlation coefficients of the charge imbalance index and each parameter in the water quality information are respectively counted. Specifically: taking the time-varying index sequence and each parameter in the water quality information at the corresponding time as inputs and substituting them into the correlation coefficient calculation function. The formula is: , where is the correlation coefficient between the charge imbalance index and the corresponding parameter, t is the time number, m is the duration in the time-varying index sequence, is the mean value of the charge imbalance index in the time-varying index sequence, is the value of the corresponding parameter in the water quality information at time t, is the mean value of the corresponding parameter; After taking the absolute value of the correlation coefficient between the charge imbalance index and all parameters in the water quality information and then calculating the mean value, a stress coupling coefficient is constructed. The formula is as follows: , is the stress coupling coefficient of the corresponding regional coordinate set, G is the total number of correlation coefficients within the corresponding regional coordinate set, g is the coordinate number within the corresponding regional coordinate set, K is the total number of parameters, and k is the parameter number. is the correlation coefficient between the charge imbalance index at coordinate g within the corresponding regional coordinate set and the k-th parameter B. Determine the stress coupling coefficients in the hot spot area coordinate set and the non-hot spot area coordinate set respectively, and perform subtraction calculation on the stress coupling coefficients in the hot spot area coordinate set and the non-hot spot area coordinate set to obtain the influence difference value. If the influence difference value exceeds the preset difference threshold, mark the imbalance hot spot area as the coupling area, otherwise it is not the coupling area; The coupling area is used to prompt the stress risk faced by the water quality in the corresponding area; S3: According to the sampling frequency, set the maximum lag order. Based on the maximum lag order, construct a lag autoregressive term using only the corresponding parameter to predict the corresponding parameter and a lag autoregressive term that also includes the charge imbalance index to verify the correlation between the charge imbalance index and the corresponding parameter in the coupling area, obtain the parameter set, find the target F statistic related to each parameter in the parameter set, and use the lag order corresponding to the target F statistic as the best time delay for the charge imbalance index to affect the corresponding parameter, and set the time stamp for triggering the warning signal for each parameter in the parameter set to formulate the corresponding treatment means.

2. An intelligent water quality monitoring method according to claim 1, characterized in that: Deploy sensor group nodes and multiple groups of nanoelectrode arrays within the target water area at a specified grid size to obtain water quality information and potential time series information at various positions in the target water area; Perform three-point calibration using standard solutions, place the electrodes in standard solutions of three different concentrations respectively, measure the corresponding potential values, and calculate the sensitivity coefficient of each electrode by recording the potential change amount and ion concentration change amount under different standard solutions; the sensitivity coefficient is used to reflect the response ability of the electrode to the change of ion concentration; Through wavelet threshold denoising and baseline drift correction of the water quality information and potential time series information, obtain the processed information set, and convert the corresponding potential values into charge concentrations according to the sensitivity coefficients of the electrodes in the potential time series information in the information set, and after normalization processing, obtain the charge time series information.

3. An intelligent water quality monitoring method according to claim 2, characterized in that: Collect the charge imbalance index corresponding to the position coordinates of each electrode array, and use Kriging interpolation to reconstruct the charge imbalance index on the grid of the entire target water area. Specifically: According to the grid range of the entire target water area, determine the points to be interpolated. Based on the spatial correlation between the positions of each point to be interpolated and the position coordinates of each collected electrode array, obtain the charge imbalance index of the points to be interpolated, expressed as: , is the charge imbalance index of the corresponding point to be interpolated at time t, i is the number of the electrode array to be collected, and n is the total number of the electrode arrays to be collected. is the charge imbalance index of the i-th electrode array at time t, is the weight coefficient of the i-th electrode array; After reconstruction, obtain the charge imbalance heat map at each moment, and perform threshold segmentation on the charge imbalance heat map; Compare the charge imbalance index at each position in the charge imbalance heat map with a preset threshold. If the charge imbalance index exceeds the threshold, mark the corresponding position as an imbalance hot spot area; otherwise, mark the corresponding position as a non-imbalance hot spot area.

4. An intelligent water quality monitoring method according to claim 3, characterized in that: Verify the correlation between the charge imbalance index and the corresponding parameters in the coupling area according to the determined coupling area, specifically including: Set the maximum lag order according to the sampling frequency. Based on the maximum lag order, construct a lag autoregressive term using only the corresponding parameter to predict the corresponding parameter and a lag autoregressive term that also includes the charge imbalance index, specifically expressed as: Predict the corresponding parameter using only the lagged autoregressive terms of the corresponding parameter: ; Lag autoregressive term that also includes the charge imbalance index: ； Wherein, P is the maximum lag order, p is the lag index, is the autoregressive coefficient, is the first residual term, is the regression coefficient of the charge imbalance index at time t-p, is the charge imbalance index at time t-p, is the second residual term, is the corresponding parameter value at time t-p; Calculate the sum of squared residuals respectively to obtain the sum of squared residuals of the first residual term and the sum of squared residuals of the second residual term; Based on the sum of squared residuals of the first residual term and the sum of squared residuals of the second residual term, construct the F-statistic. The formula is: , where F is the F-statistic, is the sum of squared residuals of the first residual term, is the sum of squared residuals of the second residual term, is the sample size; If the F-statistic exceeds the critical value of the distribution, the null hypothesis that the charge imbalance index does not Granger cause the corresponding parameter is rejected, indicating that there is a causal relationship between the charge imbalance index and the corresponding parameter, and the charge imbalance index is in the position of cause in the causal relationship. If the F-statistic does not exceed the critical value of the distribution, it indicates that there is no causal relationship between the charge imbalance index and the corresponding parameter.

5. An intelligent water quality monitoring method according to claim 4, characterized in that: Verify the correlation for different lag orders according to the maximum lag order to draw a curve of the F statistic changing with the lag order. According to the change trend in the curve, find the target F statistic from the curve; Statistically analyze the parameters that have a causal relationship with the charge imbalance index to generate a parameter set, and find the target F statistic related to each parameter in the parameter set.

6. An intelligent water quality monitoring method according to claim 5, characterized in that: The steps for obtaining the target F statistic include: Take three different lag orders with an adjacent relationship in the curve as a comparison group. After statistics, obtain several comparison groups; Calculate the mean and standard deviation of the F statistics corresponding to the corresponding lag orders in each comparison group. Select the comparison group with the largest mean and the smallest standard deviation as the target group; take the F statistic with the smallest difference from its corresponding mean in the target group as the target F statistic, and take the lag order corresponding to the target F statistic as the best time delay for the charge imbalance index to affect the corresponding parameter; According to the best time delay for the charge imbalance index to affect the corresponding parameter, obtain the time stamp for triggering the warning signal in advance when the corresponding parameter is abnormal, and formulate corresponding treatment measures within the time period of this best time delay.

7. An intelligent water quality monitoring system for implementing the intelligent water quality monitoring method according to any one of claims 1 to 6 above, characterized in that: Including: The information acquisition module is used to acquire the water quality information and potential time series information at each position in the target water area. After preprocessing, obtain an information set; The area identification module is used to analyze the charge imbalance situation at each position in the target water area according to the information set to determine the imbalance hot spot area and the non-imbalance hot spot area, respectively analyze the correlation between the charge imbalance situation and each parameter in the water quality information in the imbalance hot spot area and the non-imbalance hot spot area, and determine the coupling area through comparison; The setting module is used to verify the correlation between the charge imbalance index and the corresponding parameters in the coupling area, obtain a parameter set, and set the time stamp for triggering the warning signal for each parameter in the parameter set respectively to formulate corresponding treatment measures.

Images