Method, system and storage medium for determining an entry diagnostic index

Abstract

The present disclosure relates to a method, system and storage medium for determining an inflow infiltration diagnosis index, the method comprising: obtaining time series data of a plurality of to-be-screened water environment monitoring indicators of a monitoring point in a target drainage pipe network; determining a first fluctuation degree of each to-be-screened water environment monitoring indicator, taking the first fluctuation degree as an effectiveness feature; determining a plurality of effectiveness water environment monitoring indicators based on the effectiveness feature; performing hypothesis testing on the time series data of each effectiveness water environment monitoring indicator to obtain a sensitivity feature of the effectiveness water environment monitoring indicator; determining a plurality of sensitivity water environment monitoring indicators from the plurality of effectiveness water environment monitoring indicators based on the sensitivity feature; determining a second fluctuation degree of the time series data of each sensitivity water environment monitoring indicator; and determining a target water environment monitoring indicator based on the second fluctuation degree. Using the method can improve the accuracy and reliability of rain day inflow infiltration problem diagnosis.

Description

Technical Field

[0001] This disclosure relates to the field of water environment management technology, and in particular to a method, system and storage medium for determining inflow and infiltration diagnostic indicators. Background Technology

[0002] Urban drainage pipe networks are prone to severe infiltration problems during rainy weather. This issue not only dilutes the concentration of sewage in the drainage network and reduces the treatment efficiency of sewage treatment plants, but can also lead to pipe overflows and environmental pollution. Therefore, accurately diagnosing and identifying infiltration problems is crucial for pipe network operation and maintenance and urban water environment management.

[0003] However, traditional inflow and infiltration diagnosis often assumes that all water environment monitoring indicators have the same characterizing ability, or selects conventional indicators such as Chemical Oxygen Demand (COD) and ammonia nitrogen based on subjective experience. The physicochemical properties of different indicators, their background concentrations in groundwater / rainwater, regional drainage issues, and the degree of influence from sediment disturbance in pipe networks vary greatly. Therefore, relying on a single indicator or a subjective combination is insufficient to objectively reflect the actual inflow and infiltration situation under rainy conditions. Summary of the Invention

[0004] In view of this, this disclosure proposes a scheme for determining infiltration diagnostic indicators.

[0005] According to one aspect of this disclosure, a method for determining inflow and infiltration diagnostic indicators is provided, comprising: acquiring time series data of multiple water environment monitoring indicators to be screened at any monitoring point in a target drainage network, wherein the multiple water environment monitoring indicators to be screened include: flow rate, liquid level, and multiple water quality indicators; determining a first degree of fluctuation of the time series data of each of the water environment monitoring indicators to be screened, using the first degree of fluctuation as a validity feature, wherein the validity feature is used to exclude water environment monitoring indicators to be screened that indicate an abnormal health status of the target drainage network under dry weather conditions; and determining multiple valid water environment monitoring indicators from the multiple water environment monitoring indicators to be screened based on the validity feature. The method involves: performing hypothesis testing on the time series data of each of the aforementioned effective water environment monitoring indicators to obtain the test results for each effective water environment monitoring indicator; using the test results as sensitivity features to screen effective water environment monitoring indicators sensitive to rainfall; determining multiple sensitive water environment monitoring indicators from the multiple effective water environment monitoring indicators based on the sensitivity features; determining the second degree of fluctuation of the time series data of each of the aforementioned sensitive water environment monitoring indicators; and determining, from the multiple sensitive water environment monitoring indicators, a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any monitoring point in a rainy weather scenario based on the second degree of fluctuation.

[0006] In one possible implementation, determining a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any monitoring point in a rainy weather scenario, based on the second degree of fluctuation, from the plurality of sensitive water environment monitoring indicators, includes: determining a plurality of robust water environment monitoring indicators from the plurality of sensitive water environment monitoring indicators based on the second degree of fluctuation of the time series data of each of the sensitive water environment monitoring indicators; fitting the third degree of fluctuation of the time series data of each of the robust water environment monitoring indicators to the rainfall to obtain the slope of the numerical change characterizing the sensitivity of the numerical change of each robust water environment monitoring indicator to the rainfall response, and the determination coefficient characterizing the goodness of fit of each robust water environment monitoring indicator, and using the slope of the numerical change and the determination coefficient as robustness features; and determining a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any monitoring point in a rainy weather scenario from the plurality of robust water environment monitoring indicators based on the robustness features.

[0007] In one possible implementation, fitting the third fluctuation degree of the time series data of each robust water environment monitoring indicator to rainfall to obtain the slope of numerical change characterizing the sensitivity of the numerical change of each robust water environment monitoring indicator to rainfall, and the coefficient of determination characterizing the goodness of fit of each robust water environment monitoring indicator, includes: determining a target model for each robust water environment monitoring indicator based on the response relationship between the third fluctuation degree of the time series data of each robust water environment monitoring indicator and rainfall; and fitting the third fluctuation degree of the time series data of the robust water environment monitoring indicator corresponding to the target model to rainfall based on the target model to obtain the slope of numerical change and the coefficient of determination.

[0008] In one possible implementation, the hypothesis testing for the time series data of each of the effective water environment monitoring indicators includes: for any one of the effective water environment monitoring indicators, dividing the time series data of the effective water environment monitoring indicator into dry day time series data and rainy day time series data according to weather scenarios to obtain a data group for the effective water environment monitoring indicator; and performing hypothesis testing on the data group.

[0009] In one possible implementation, determining the first degree of fluctuation of the time series data of each of the water environment monitoring indicators to be screened includes: dividing the time series data of each of the water environment monitoring indicators to be screened into multiple time series segments; determining a first coefficient of variation corresponding to each time series segment of the multiple time series segments of the water environment monitoring indicators to be screened, wherein the first coefficient of variation characterizes the first degree of fluctuation; or, determining multiple preset quantiles of the historical time series data of each of the water environment monitoring indicators to be screened; determining a relative range based on the multiple quantiles; selecting any water environment monitoring indicator to be screened whose relative range satisfies a first range condition as a first water environment monitoring indicator; selecting any water environment monitoring indicator to be screened whose relative range satisfies a second range condition as a second water environment monitoring indicator; dividing the time series data of each of the first water environment monitoring indicator and the second water environment monitoring indicator into multiple time series segments; determining the standard deviation of each time series segment of the first water environment monitoring indicator; and determining a second coefficient of variation of each time series segment of the second water environment monitoring indicator, wherein the standard deviation and the second coefficient of variation characterize the first degree of fluctuation.

[0010] In one possible implementation, determining multiple effective water environment monitoring indicators from the multiple water environment monitoring indicators to be screened based on the effectiveness characteristics includes: determining the distribution characteristics of the first coefficient of variation corresponding to each of the multiple water environment monitoring indicators to be screened, the distribution characteristics including: the statistical value of the first coefficient of variation, or a first number of first coefficients of variation not less than a coefficient of variation threshold; determining the water environment monitoring indicators to be screened whose distribution characteristics satisfy the first effectiveness condition as the effective water environment monitoring indicators; or, for each second water environment monitoring indicator, calculating a second number of second coefficients of variation that satisfy the second effectiveness condition; determining the second water environment monitoring indicators whose second number is not less than a second number threshold as the effective water environment monitoring indicators; for each first water environment monitoring indicator, calculating a third number of standard deviations that satisfy the third effectiveness condition; determining the first water environment monitoring indicators whose third number is not less than a third number threshold as the effective water environment monitoring indicators.

[0011] In one possible implementation, acquiring time-series data of multiple water environment monitoring indicators to be screened at any monitoring point in the target drainage network includes: acquiring the original time-series data of any water environment monitoring indicator to be screened; identifying outliers in the original time-series data of the water environment monitoring indicator to be screened; and obtaining the time-series data of the water environment monitoring indicator to be screened by using a corresponding processing strategy based on the duration of the outliers; the processing strategy includes: setting an invalidation flag for the outliers when the duration is greater than a duration threshold; and removing the outliers when the duration is not greater than the duration threshold.

[0012] According to another aspect of this disclosure, an apparatus for determining inflow and infiltration diagnostic indicators is provided. The apparatus includes:

[0013] The time series data determination unit is used to acquire time series data of multiple water environment monitoring indicators to be screened at any monitoring point in the target drainage network. The multiple water environment monitoring indicators to be screened include: flow rate, liquid level and multiple water quality indicators.

[0014] The effective feature determination unit is used to determine the first fluctuation degree of the time series data of each of the water environment monitoring indicators to be screened, and to use the first fluctuation degree as the effectiveness feature. The effectiveness feature is used to exclude water environment monitoring indicators to be screened that indicate an abnormality in the health status of the target drainage network under arid weather scenarios.

[0015] An effective water environment indicator determination unit is used to determine multiple effective water environment monitoring indicators from the multiple water environment monitoring indicators to be screened based on the effectiveness characteristics.

[0016] The sensitivity feature determination unit is used to perform hypothesis testing on the time series data of each of the effective water environment monitoring indicators, obtain the test results of each of the effective water environment monitoring indicators, and use the test results as sensitivity features. The sensitivity features are used to screen effective water environment monitoring indicators that are sensitive to rainfall.

[0017] A sensitive water environment monitoring index determination unit is used to determine multiple sensitive water environment monitoring indicators from the multiple effective water environment monitoring indicators based on the sensitivity characteristics.

[0018] The second fluctuation degree determination unit is used to determine the second fluctuation degree of the time series data of each of the aforementioned sensitive water environment monitoring indicators;

[0019] The target water environment monitoring indicator determination unit, based on the second fluctuation level, determines the target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any one of the monitoring points in a rainy weather scenario from among the multiple sensitive water environment monitoring indicators.

[0020] In one possible implementation, the target water environment monitoring index determination unit is further configured to:

[0021] Based on the second degree of fluctuation of the time series data of each of the aforementioned sensitive water environment monitoring indicators, a plurality of robust water environment monitoring indicators are determined from the plurality of sensitive water environment monitoring indicators.

[0022] The third fluctuation degree of the time series data of each robust water environment monitoring index is fitted with the rainfall to obtain the slope of the numerical change that characterizes the sensitivity of the numerical change of each robust water environment monitoring index to the response of rainfall, and the coefficient of determination that characterizes the goodness of fit of each robust water environment monitoring index. The slope of the numerical change and the coefficient of determination are used as robustness features.

[0023] Based on the robustness characteristics, a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any of the monitoring points in rainy weather scenarios is determined from the plurality of robust water environment monitoring indicators.

[0024] In one possible implementation, the target water environment monitoring index determination unit is further configured to:

[0025] Based on the third degree of fluctuation of the time series data of each of the robust water environment monitoring indicators and the response relationship with rainfall, target models are determined for each of the robust water environment monitoring indicators.

[0026] Based on the target model, the third fluctuation degree of the time series data of the robust water environment monitoring index corresponding to the target model is fitted with the rainfall to obtain the slope of the numerical change and the coefficient of determination.

[0027] In one possible implementation, the sensitivity feature determination unit is further configured to:

[0028] For any of the time series data of the effective water environment monitoring indicator, the time series data of the effective water environment monitoring indicator is divided into dry day time series data and rainy day time series data according to the weather scenario to obtain the data group of the effective water environment monitoring indicator;

[0029] Hypothesis testing was performed on the data set.

[0030] In one possible implementation, the effective feature determination unit is further configured to:

[0031] The time series data of each of the water environment monitoring indicators to be screened are divided into multiple time series segments;

[0032] Determine the first coefficient of variation for each time segment among multiple time segments of the water environment monitoring indicators to be screened, wherein the first coefficient of variation characterizes the first degree of fluctuation;

[0033] or,

[0034] Determine multiple preset quantiles for the historical time series data of each of the water environment monitoring indicators to be screened;

[0035] Based on multiple quantiles, a relative range is determined. Any water environment monitoring indicator to be screened that satisfies the first range condition is selected as the first water environment monitoring indicator, and any water environment monitoring indicator to be screened that satisfies the second range condition is selected as the second water environment monitoring indicator.

[0036] The time series data of the first water environment monitoring indicator and the second water environment monitoring indicator are each divided into multiple time series segments;

[0037] Determine the standard deviation of each time segment of the first water environment monitoring indicator;

[0038] The second coefficient of variation for each time segment of the second water environment monitoring indicator is determined, wherein the standard deviation and the second coefficient of variation characterize the first degree of fluctuation.

[0039] In one possible implementation, the effective water environment index determination unit is further configured to:

[0040] Determine the distribution characteristics of the first coefficient of variation corresponding to each of the plurality of water environment monitoring indicators to be screened, wherein the distribution characteristics include: the statistical value of the first coefficient of variation, or the first number of the first coefficient of variation that is not less than the coefficient of variation threshold;

[0041] The water environment monitoring indicators whose distribution characteristics meet the first effective condition are identified as the effective water environment monitoring indicators.

[0042] or,

[0043] For each of the second water environment monitoring indicators, the second number of the second coefficients of variation that meet the second effective conditions is calculated.

[0044] The second water environment monitoring indicator whose second quantity is not less than the second quantity threshold is determined as the effective water environment monitoring indicator;

[0045] For each of the first water environment monitoring indicators, the number of standard deviations that meet the third validity condition is calculated.

[0046] The first water environment monitoring indicator whose third quantity is not less than the third quantity threshold is determined as the effective water environment monitoring indicator.

[0047] In one possible implementation, the time series data determination unit is further configured to:

[0048] For any water environment monitoring indicator to be screened, obtain the original time series data of that water environment monitoring indicator;

[0049] Identify outliers in the original time series data of the water environment monitoring indicator to be screened;

[0050] Based on the duration of the outlier, the time series data of the water environment monitoring indicator to be screened are obtained using the corresponding processing strategy.

[0051] The processing strategy includes: setting a failure flag for outliers when the duration exceeds a duration threshold;

[0052] Furthermore, if the duration is not greater than the duration threshold, the outlier is removed.

[0053] In one possible implementation, the outlier includes one or more of a first outlier, a second outlier, and a third outlier, and the time series data determination unit is further configured to:

[0054] Identify the first outlier in the original time series data of the water environment monitoring indicator to be screened that does not meet the first preset condition;

[0055] Statistical analysis is performed on the data in the original time series of the water environment monitoring indicator to be screened, excluding the first outlier, to identify the second outlier that exceeds the statistical range.

[0056] The data in the original time series of the water environment monitoring indicator to be screened, excluding the first outlier and the second outlier, and the data of at least one first water environment monitoring indicator that is related to the water environment monitoring indicator to be screened, are subjected to logical relationship verification to determine the third outlier that does not conform to the logical relationship. The data of the first water environment monitoring indicator to be screened are the data in the original time series of the first water environment monitoring indicator to be screened, excluding the first outlier and the second outlier.

[0057] According to another aspect of this disclosure, a system for determining inflow and infiltration diagnostic indicators is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0058] According to another aspect of this disclosure, a non-volatile computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the above-described method.

[0059] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0060] In this embodiment, for the water environment monitoring indicators to be screened at each monitoring point, indicators indicating drought background problems can be eliminated based on the first fluctuation degree (effectiveness characteristic) of each indicator, thus obtaining effective water environment monitoring indicators. Hypothesis testing is performed on each effective indicator to obtain the test results. Based on the test results (sensitivity characteristics) of each effective indicator, sensitive water environment monitoring indicators sensitive to rainfall are selected. Then, based on the second fluctuation degree of each sensitive water environment monitoring indicator, sensitive water environment monitoring indicators that have a certain degree of response to rainfall intensity are selected as target water environment monitoring indicators. Therefore, using the method of this embodiment, water environment monitoring indicators that conform to the characteristics of the water environment (i.e., do not indicate drought background problems), whose values ​​change significantly compared to drought days, and which have a degree of response to rainfall intensity can be scientifically and accurately selected for each monitoring point. These indicators can then be used for diagnosing rainwater inflow and infiltration problems at each monitoring point, improving the accuracy and reliability of rainwater inflow and infiltration problem diagnosis.

[0061] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0062] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this disclosure together with the specification and serve to explain the principles of this disclosure.

[0063] Figure 1 This is a flowchart illustrating the method for determining inflow and infiltration diagnostic indicators provided in an embodiment of this disclosure.

[0064] Figure 2 This is a flowchart illustrating the method for determining inflow and infiltration diagnostic indicators provided in an embodiment of this disclosure.

[0065] Figure 3 This is a schematic diagram of the structure of the inflow and infiltration diagnostic index determination system provided in an embodiment of this disclosure. Detailed Implementation

[0066] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0067] As used herein, the terms “comprising,” “including,” “having,” or variations thereof are open-ended and include one or more of the stated features, integrals, elements, steps, components, or functions, but do not exclude the presence or addition of one or more other features, integrals, elements, steps, components, functions, or groups thereof.

[0068] When an element is referred to as “connected,” “coupled,” “responding,” or a variation thereof relative to another element, it may be directly connected, coupled, or responding to another element, or there may be an intermediate element present.

[0069] Although the terms first, second, third, etc., may be used herein to describe various elements / operations, these elements / operations should not be limited by these terms. These terms are only used to distinguish one element / operation from another. Therefore, without departing from the teachings of the inventive concept, a first element / operation in some embodiments may be referred to as a second element / operation in other embodiments.

[0070] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0071] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0072] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, data stored, data displayed, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant regions.

[0073] Figure 1 A flowchart illustrating the method for determining inflow and infiltration diagnostic indicators provided in this embodiment of the disclosure. The method includes:

[0074] S11, acquire time series data of multiple water environment monitoring indicators to be screened at any monitoring point in the target drainage network, the multiple water environment monitoring indicators to be screened include: flow rate, liquid level and multiple water quality indicators.

[0075] The target drainage network is the wastewater drainage network for which inflow and infiltration issues are monitored. The target drainage network may contain multiple monitoring points. The water environment monitoring indicators to be screened can be indicators reflecting the state of the water environment. These multiple water environment monitoring indicators include hydrological parameters (e.g., flow rate and liquid level) and multiple water quality indicators. Among these, the multiple water quality indicators may include some or all of the following: Chemical Oxygen Demand (COD), conductivity, ammonia nitrogen, five-day biochemical oxygen demand, turbidity, total organic carbon, total hardness, total phosphorus, total nitrogen, suspended solids, total dissolved solids, petroleum hydrocarbons, pH (acidity / alkalinity), anionic surfactants, cyanide, sulfide, fluoride, chloride, organophosphorus compounds, sulfate, mercury, chromium, cadmium, arsenic, lead, nickel, beryllium, silver, selenium, copper, zinc, manganese, iron, volatile phenols, benzene series compounds, aniline compounds, nitrobenzene, and water temperature.

[0076] For each water environment monitoring indicator to be screened, values ​​corresponding to multiple time points can be obtained to form time series data.

[0077] S12, determine the first degree of fluctuation of the time series data of each of the water environment monitoring indicators to be screened, and use the first degree of fluctuation as a validity feature. The validity feature is used to exclude water environment monitoring indicators to be screened that indicate abnormal health status of the target drainage network under drought conditions.

[0078] The first degree of volatility can reflect the numerical dispersion of the time series data of the water environment monitoring indicators to be screened. For example, the first degree of volatility can be the coefficient of variation of the time series data of the water environment monitoring indicators to be screened. Another example is the standard deviation of the time series data of the water environment monitoring indicators to be screened.

[0079] In dry weather scenarios, the data values ​​of the water environment monitoring indicators to be screened already show anomalies. That is, given the abnormal background levels during dry weather, the data values ​​of these indicators will not change significantly during rainy weather. For example, if the main sewage pipes are laid along a river, and river water continues to seep in during dry weather due to pipe damage, the sewage in the network will be diluted by a large amount of river water, thus reducing the COD concentration. In this case, even if it rains, the infiltration of rainwater will further dilute the COD concentration, and the COD concentration will not decrease significantly. If the water environment monitoring indicator to be screened is used to identify infiltration problems at monitoring points, it will mislead the conclusion that no infiltration occurs during rainy weather, reducing the accuracy of infiltration risk assessment. Therefore, this disclosure, by determining validity characteristics, can screen out water environment monitoring indicators whose data values ​​already show anomalies during dry weather scenarios, thereby accurately determining the target water environment monitoring indicator.

[0080] S13, Based on the effectiveness characteristics, determine a number of effective water environment monitoring indicators from the multiple water environment monitoring indicators to be screened.

[0081] For example, an effectiveness threshold can be preset, and water environment monitoring indicators whose feature values ​​are greater than the effectiveness threshold can be selected as effective water environment monitoring indicators.

[0082] For example, the water environment monitoring indicators to be screened can be sorted from high to low according to the characteristic value of the effectiveness feature. Starting from the top of the sort, a preset first selection number of water environment monitoring indicators to be screened are obtained and used as the effective water environment monitoring indicators.

[0083] The above are merely examples, and the embodiments disclosed herein are not intended to limit the scope of the invention. Other methods will be described below.

[0084] S14, perform hypothesis testing on the time series data of each of the effective water environment monitoring indicators to obtain the test results of each of the effective water environment monitoring indicators, and use the test results as sensitivity features to screen effective water environment monitoring indicators that are sensitive to rainfall.

[0085] Among them, effective water environment monitoring indicators that are sensitive to rainfall refer to effective water environment monitoring indicators whose data values ​​change significantly and rapidly in rainy weather scenarios compared to dry weather scenarios.

[0086] To scientifically verify whether there are fundamental differences between various effective water environment monitoring indicators in dry and rainy weather scenarios, and thus select the most sensitive effective water environment monitoring indicators to rainfall events, we can verify them from the perspective of statistical significance, thereby providing an objective and quantifiable decision-making basis for the selection of target indicators.

[0087] Hypothesis testing yields the p-values, which characterize the significance level, for each effective water environment monitoring indicator. Hypothesis testing can be either a t-test or a z-test.

[0088] The test results reflect the probability that, under the null hypothesis that "there is no difference in the statistical values ​​of effective water environment monitoring indicators between dry and rainy weather scenarios," the statistical value of this effective water environment monitoring indicator actually exceeds the first difference threshold. In this case, the smaller the test result, the less likely it is that accidental factors will affect the first difference of this effective water environment monitoring indicator. Because this difference is extremely difficult to occur under accidental circumstances, yet it actually occurs, we have reason to reject the null hypothesis, that is, to infer that this first difference is not dominated by accidental factors, thus demonstrating that rainfall has a significant impact on this effective water environment monitoring indicator. The statistical value can be the mean or the median.

[0089] In this embodiment of the disclosure, the test results of the effective water environment monitoring indicators can be used as the sensitivity characteristics of the effective water environment monitoring indicators. Using these sensitivity characteristics, sensitive water environment monitoring indicators can be selected from the effective water environment monitoring indicators.

[0090] In one possible implementation, the hypothesis testing for the time series data of each of the effective water environment monitoring indicators includes: for any one of the effective water environment monitoring indicators, dividing the time series data of the effective water environment monitoring indicator into dry day time series data and rainy day time series data according to weather scenarios to obtain a data group for the effective water environment monitoring indicator; and performing hypothesis testing on the data group.

[0091] Weather scenarios can include rainy and dry weather scenarios. A dry weather scenario is defined as a period of at least 48 consecutive hours without rain following a rainfall event. A rainy weather scenario encompasses the entire period from the rainfall to the return of water flow to dry weather levels. Therefore, dry weather time series data and rainy weather time series data can be obtained from the time series data of effective water environment monitoring indicators, forming a data set. That is, this data set contains two independent samples. Hypothesis testing is performed on the data set for each effective water environment monitoring indicator. The null hypothesis is that there is no difference in the statistical values ​​of the effective water environment monitoring indicators between the dry and rainy weather scenarios. The alternative hypothesis is that there are significant differences in the statistical values ​​of the effective water environment monitoring indicators between the dry and rainy weather scenarios. In this embodiment of the disclosure, determining the two independent samples in the data set according to the weather scenario facilitates the setting of hypotheses and improves the reliability and rationality of the hypothesis testing.

[0092] The core objective of hypothesis testing is to examine whether there is a significant difference between the population means of two independent sample groups (referred to in this disclosure as "dry day time series data" and "rainy day time series data"). Its null hypothesis (…) The alternative hypothesis is that the mean of this indicator is no different between dry and rainy days. The hypothesis is that the means of the two samples are significantly different. Before conducting a hypothesis test, the data can be preprocessed to ensure that the basic preconditions for the test are met, including the data's approximate normal distribution (which can be verified by the Shapiro-Wilk test or by observing the QQ plot) and homogeneity of variance (verified by Levene's test). If the aforementioned preconditions are not met, non-parametric tests (such as the Mann-Whitney U test) will be used as alternatives to ensure the rigor of the analytical method. Here, not meeting the preconditions means that the data at least does not conform to a normal distribution, or at least does not conform to homogeneity of variance.

[0093] S15, Based on the aforementioned sensitivity characteristics, determine a plurality of sensitive water environment monitoring indicators from the plurality of effective water environment monitoring indicators.

[0094] For example, a sensitivity threshold can be preset, and effective water environment monitoring indicators whose feature values ​​are less than the sensitivity threshold can be used as sensitive water environment monitoring indicators.

[0095] For example, the effective water environment monitoring indicators can be sorted from high to low according to the characteristic values ​​of the sensitivity features. Starting from the top of the sort, a preset second selection number of effective water environment indicators are obtained and used as sensitive water environment monitoring indicators.

[0096] The above are merely examples, and the embodiments disclosed herein are not intended to limit the scope of the invention.

[0097] S16, determine the second degree of fluctuation of the time series data of each of the aforementioned sensitive water environment monitoring indicators; based on the second degree of fluctuation, determine the target water environment monitoring indicator suitable for diagnosing the inflow and infiltration problem at any one of the monitoring points in a rainy weather scenario from the plurality of sensitive water environment monitoring indicators.

[0098] The second degree of fluctuation can reflect the response of the time series data of sensitive water environment monitoring indicators to rainfall intensity. The second degree of fluctuation can be the peak-to-trough ratio, and / or the magnitude of the decline, and / or the rate of increase.

[0099] For example, a first fluctuation threshold can be preset, and a second fluctuation threshold greater than the first fluctuation threshold can be used as the target water environment monitoring indicator.

[0100] For example, the sensitive water environment monitoring indicators can be sorted from high to low according to the value of the second fluctuation level. Starting from the top of the sort, a preset third selection number of sensitive water environment indicators are obtained and used as the target water environment monitoring indicators.

[0101] The above are merely examples, and the embodiments disclosed herein are not intended to limit the scope of the invention. Other methods will be described below.

[0102] In this embodiment of the disclosure, steps S11-S16 can be performed simultaneously or at different times at one or more monitoring points to determine the target water environment monitoring indicators for the one or more monitoring points. In other words, the method of this disclosure can determine target water environment monitoring indicators suitable for diagnosing inflow and infiltration problems in rainy weather scenarios for the monitoring points.

[0103] If a selected water environment indicator indicates a dry-weather background problem in the drainage network, it is highly likely that the indicator's value will not change significantly during rainy days. Therefore, it is unsuitable as a target water environment monitoring indicator for assessing inflow and infiltration problems in rainy weather and should be eliminated. Furthermore, under normal dry-weather background conditions, rainfall can significantly amplify the signal of external water intrusion in the drainage network, revealing previously hidden inflow and infiltration problems. Thus, water environment indicators that are more sensitive to rainfall—that is, whose values ​​change significantly compared to dry days and who exhibit a certain degree of responsiveness to rainfall intensity—are suitable for selection as target water environment monitoring indicators for assessing inflow and infiltration problems in rainy weather.

[0104] In this embodiment, for the water environment monitoring indicators to be screened at each monitoring point, indicators indicating drought background problems can be eliminated based on the first fluctuation degree (effectiveness characteristic) of each indicator, thus obtaining effective water environment monitoring indicators. Hypothesis testing is performed on each effective indicator to obtain the test results. Based on the test results (sensitivity characteristics) of each effective indicator, sensitive water environment monitoring indicators sensitive to rainfall are selected. Then, based on the second fluctuation degree of each sensitive water environment monitoring indicator, sensitive water environment monitoring indicators that have a certain degree of response to rainfall intensity are selected as target water environment monitoring indicators. Therefore, using the method of this embodiment, water environment monitoring indicators that conform to the characteristics of the water environment (i.e., do not indicate drought background problems), whose values ​​change significantly compared to drought days, and which have a degree of response to rainfall intensity can be scientifically and accurately selected for each monitoring point. These indicators can then be used for diagnosing rainwater inflow and infiltration problems at each monitoring point, improving the accuracy and reliability of rainwater inflow and infiltration problem diagnosis.

[0105] In one possible implementation, determining a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any monitoring point in a rainy weather scenario, based on the second degree of fluctuation, from the plurality of sensitive water environment monitoring indicators, includes: determining a plurality of robust water environment monitoring indicators from the plurality of sensitive water environment monitoring indicators based on the second degree of fluctuation of the time series data of each of the sensitive water environment monitoring indicators; fitting the third degree of fluctuation of the time series data of each of the robust water environment monitoring indicators to the rainfall to obtain the slope of the numerical change characterizing the sensitivity of the numerical change of each robust water environment monitoring indicator to the rainfall response, and the determination coefficient characterizing the goodness of fit of each robust water environment monitoring indicator, and using the slope of the numerical change and the determination coefficient as robustness features; and determining a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any monitoring point in a rainy weather scenario from the plurality of robust water environment monitoring indicators based on the robustness features.

[0106] In this embodiment of the disclosure, robust water environment indicators can be determined from sensitive water environment monitoring indicators based on a second fluctuation level.

[0107] For example, a second fluctuation level threshold can be preset, and sensitive water environment monitoring indicators with a second fluctuation level greater than the second fluctuation level threshold can be used as robust water environment monitoring indicators.

[0108] For example, the second fluctuation level may include three sub-indicators: a first sub-indicator, a second sub-indicator, and a third sub-indicator. For instance, the first sub-indicator could be the peak-to-trough ratio, the second sub-indicator the magnitude of the decrease in value, and the third sub-indicator the rate of increase in value. Each sensitive water environment monitoring indicator can be sorted from highest to lowest value according to the three sub-indicators. For each sub-indicator, starting from the top of the ranking, a preset fourth selection number of sensitive water environment monitoring indicators (e.g., selecting the top 50% of each ranked indicator) is obtained and used as robust water environment monitoring indicators. The fourth selection number can be greater than the second and third selection numbers.

[0109] The peak-to-trough ratio is the ratio of the peak or trough of a sensitive water environment monitoring indicator in a rainy weather scenario to the mean of the same indicator in a dry weather scenario. The decrease rate is the ratio of the indicator value at the start of rainfall in the rainy weather scenario to the first trough value of the same indicator in the same rainy weather scenario. The rise rate is the ratio of the indicator value at the start of rainfall in the rainy weather scenario to the first peak value of the same indicator in the same rainy weather scenario.

[0110] The third degree of fluctuation can be of the same type as the second degree of fluctuation. That is, the third degree of fluctuation can be: peak-to-trough ratio, and / or decrease magnitude, and / or increase rate. The time series data of robust water environment monitoring indicators can be divided into multiple time series segments. For ease of distinction, these time series segments are named robust indicator time series segments. For example, multiple robust indicator time series segments can be divided from the time series data of robust water environment monitoring indicators by hour, day, or quarter. The third degree of fluctuation can be the peak-to-trough ratio, and / or decrease magnitude, and / or increase rate of each robust indicator time series segment of the corresponding robust water environment monitoring indicator. The third degree of fluctuation corresponding to each robust indicator time series segment in the time series data of the robust water environment monitoring indicator can be used as the dependent variable, and the rainfall within the corresponding time period (i.e., the total rainfall within the corresponding time period) can be used as the independent variable. A preset mathematical model is then fitted. The slope and coefficient of determination of the fitted numerical changes are used as the robustness characteristics of the robust water environment monitoring indicator.

[0111] Using numerical change slope and coefficient of determination, water environment monitoring indicators that have a stable and reliable relationship with rainfall as a driving factor can be screened out and used as the final target water environment monitoring indicators for diagnosing inflow and infiltration problems in rainy weather scenarios.

[0112] Furthermore, based on robustness characteristics, target water environment monitoring indicators can be determined from multiple robust water environment monitoring indicators.

[0113] The slope of the numerical change is positively correlated with the sensitivity of robust water environment monitoring indicators to rainfall; that is, the larger the slope of the numerical change, the greater the degree of third fluctuation caused by rainfall, and the more sensitive the robust water environment monitoring indicator is. The coefficient of determination is positively correlated with the degree to which rainfall dominates the numerical change of robust water environment monitoring indicators; that is, the higher the coefficient of determination, the lower the probability that the numerical change of robust water environment monitoring indicators is caused by non-rainfall factors.

[0114] In one possible implementation, a slope threshold and a coefficient of determination threshold can be preset. Robust water environment monitoring indicators whose numerical change slope is greater than the slope threshold and whose coefficient of determination is greater than the coefficient of determination threshold are selected as the target water environment monitoring indicators. Robust water environment monitoring indicators whose numerical change slope is greater than the slope threshold and whose coefficient of determination is not greater than the coefficient of determination threshold are selected as alternative indicators.

[0115] When the number of target water environment monitoring indicators is insufficient, or when additional water environment monitoring indicators are needed as supplementary measures, alternative indicators and target water environment monitoring indicators can be used together to diagnose rainwater infiltration problems. In this way, even when the number or effectiveness of target water environment monitoring indicators is insufficient, the accuracy and reliability of infiltration problem diagnosis can still be maintained.

[0116] In another approach, the robust water environment monitoring indicators are ranked from highest to lowest according to their numerical change slope and coefficient of determination. The robust water environment monitoring indicators that rank at the top (e.g., 30%) in each ranking are then used as the target water environment monitoring indicators.

[0117] The target water environment monitoring indicators determined in this way are not only sensitive to rainfall, but also have a stable and reliable relationship with rainfall as a driving factor, which can improve the reliability of diagnosing inflow and infiltration problems.

[0118] In this embodiment, a progressive screening process is conducted on the water environment monitoring indicators to be screened, based on effectiveness, sensitivity, and robustness characteristics. Specifically, target water environment monitoring indicators are progressively selected from three dimensions: the dispersion of the time series of the indicators, the responsiveness of the indicator values ​​to rainfall intensity (i.e., sensitivity to rainfall), and the stability of this responsiveness (i.e., not only sensitivity to rainfall but also a reliable and stable relationship with rainfall). This progressive screening aims to meet the diagnostic criteria for rainwater infiltration problems and match the water environment characteristics of the corresponding monitoring points, thereby improving the quality of rainwater infiltration problem assessment and making the diagnostic results more convincing. Through the method of this embodiment, the optimal, few key water environment monitoring indicators—the target water environment monitoring indicators—can be selected, guiding the targeted deployment of online monitoring equipment, reducing redundant monitoring, and lowering monitoring costs.

[0119] In one possible implementation, fitting the third fluctuation degree of the time series data of each robust water environment monitoring indicator to rainfall to obtain the slope of numerical change characterizing the sensitivity of the numerical change of each robust water environment monitoring indicator to rainfall, and the coefficient of determination characterizing the goodness of fit of each robust water environment monitoring indicator, includes: determining a target model for each robust water environment monitoring indicator based on the response relationship between the third fluctuation degree of the time series data of each robust water environment monitoring indicator and rainfall; and fitting the third fluctuation degree of the time series data of the robust water environment monitoring indicator corresponding to the target model to rainfall based on the target model to obtain the slope of numerical change and the coefficient of determination.

[0120] In this embodiment of the disclosure, the response relationship can characterize the changing trend between the third fluctuation level of the robust water environment monitoring index and rainfall. This changing trend can be a positive correlation, a negative correlation, an exponential change, a logarithmic change, or a power function change, etc.

[0121] For example, the time series data of robust water environment monitoring indicators can be divided into multiple time series segments, each named a robust time series segment. The degree of third fluctuation corresponding to each robust time series segment in the time series data of the robust water environment monitoring indicators is used as the dependent variable, and the rainfall corresponding to each robust time series segment is used as the independent variable. A curve is plotted with the horizontal axis representing the degree of third fluctuation corresponding to each robust time series segment and the vertical axis representing the rainfall corresponding to each robust time series segment (the total rainfall within that robust time series segment). The response relationship is determined based on this curve.

[0122] For example, statistical analysis software can be used to analyze the changing trends of the third fluctuation degree and rainfall corresponding to each robust time series segment in the time series data of robust water environment monitoring indicators, and determine the response relationship.

[0123] In this embodiment of the disclosure, the target model can be determined based on the response relationship. The target model is a mathematical model. For example, the target model can be a linear model or a nonlinear model.

[0124] In this way, a matching target model can be determined based on the response relationships of each robust water environment monitoring index at each monitoring point. This ensures that the fitting results accurately reflect the response relationship between the third fluctuation degree of the robust water environment index at the fitted monitoring point and rainfall, improving the reliability and accuracy of the determined robustness characteristics. Furthermore, the robustness characteristics can accurately characterize the degree to which rainfall dominates the third fluctuation degree of the robust water environment monitoring index, thereby improving the accuracy of the determined target water environment monitoring index.

[0125] In one possible implementation, determining the first degree of fluctuation of the time series data of each of the water environment monitoring indicators to be screened includes: dividing the time series data of each of the water environment monitoring indicators to be screened into multiple time series segments; determining a first coefficient of variation corresponding to each time series segment of the multiple time series segments of the water environment monitoring indicators to be screened, wherein the first coefficient of variation characterizes the first degree of fluctuation; or, determining multiple preset quantiles of the historical time series data of each of the water environment monitoring indicators to be screened; determining a relative range based on the multiple quantiles; selecting any water environment monitoring indicator to be screened whose relative range satisfies a first range condition as a first water environment monitoring indicator; selecting any water environment monitoring indicator to be screened whose relative range satisfies a second range condition as a second water environment monitoring indicator; dividing the time series data of each of the first water environment monitoring indicator and the second water environment monitoring indicator into multiple time series segments; determining the standard deviation of each time series segment of the first water environment monitoring indicator; and determining a second coefficient of variation of each time series segment of the second water environment monitoring indicator, wherein the standard deviation and the second coefficient of variation characterize the first degree of fluctuation.

[0126] In this embodiment, the time series data of the water environment monitoring indicators to be screened can be divided into multiple time segments according to a preset time interval. The duration of each time segment can be equal. The time interval can be set according to the time point density of the time series and the actual scenario requirements. The time interval can be: a quarter, a day, or an hour, etc.

[0127] For example, for each water environment monitoring indicator to be screened, a first coefficient of variation can be determined for each of its corresponding multiple time series segments. The first coefficient of variation can be used to represent the first degree of fluctuation of the water environment monitoring indicator to be screened.

[0128] For example, for each water environment monitoring indicator to be screened, the range of historical time series data for that indicator can be determined. For instance, the relative range of the historical time series data for the water environment monitoring indicator to be screened can be determined. The first range condition is that the relative range is less than a range threshold. The second range condition is that the relative range is not less than a range threshold. If the relative range meets the first range condition, it indicates that the data variation range of the water environment monitoring indicator to be screened is small; the corresponding water environment monitoring indicators to be screened include, for example, water temperature and liquid level. If the relative range meets the second range condition, it indicates that the data variation range of the water environment monitoring indicator to be screened is large; the corresponding water environment monitoring indicators to be screened include, for example, chemical oxygen demand, conductivity, and flow rate. Therefore, the water environment monitoring indicators to be screened can be divided into first water environment monitoring indicators and second water environment monitoring indicators.

[0129] The time series data of both the first and second water environment monitoring indicators can be divided into multiple time segments according to time intervals. For the first water environment monitoring indicator, the standard deviation of each of its corresponding time segments can be determined. The standard deviation can be used to represent the first degree of fluctuation of the first water environment monitoring indicator. For the second water environment monitoring indicator, the second coefficient of variation of each of its corresponding time segments can be determined. The second coefficient of variation can be used to represent the first degree of fluctuation of the second water environment monitoring indicator.

[0130] Furthermore, traditional methods use the ratio of the difference between the maximum and minimum values ​​of multiple data points to their arithmetic mean as the relative range. This determined relative range is easily affected by extreme values, such as sensor malfunctions or sudden pollution spikes, leading to overestimation of the calculated range and failing to accurately reflect normal fluctuations.

[0131] This disclosure proposes a novel method for determining the relative range. Multiple preset quantiles can be obtained from the historical time-series data of the water environment monitoring indicators to be screened. Here, historical time-series data refers to data from periods other than the time-series data of the water environment monitoring indicators to be screened. For example, if the time-series data for the water environment monitoring indicators to be screened is from July 2026, then the historical time-series data could be from June 2026.

[0132] For example, the arithmetic mean of multiple quantiles can be determined, and this arithmetic mean can be used as the relative range of the water environment monitoring indicator to be screened.

[0133] For example, multiple quantiles include the median. Each quantile other than the median can be a symmetrically selected quantile. For example, multiple quantiles could be the 10th quantile, the 50th quantile, and the 90th quantile. For ease of understanding, formula (1) is used to represent the process of determining the relative range.

[0134] (1)

[0135] in, Indicates relative range. The quantile, representing the typical central level of the data (i.e., the reference level), is used to convert the absolute fluctuation width of the numerator into a relative proportion. The subscript m can be 50. The median, The core width represents the data fluctuation. and Represents symmetric quantiles. This indicates the higher quantile, such as the upper quantile. The subscript n is a percentage value, such as 90, 95, or 75. This indicates the lower quantile, such as the lower quantile. The subscript n1 is a percentage value that is symmetrical to n, i.e., n + n1 = 100.

[0136] Using the quantile-based relative range in this embodiment of the disclosure can more robustly characterize the magnitude of data variation, free from the interference of extreme values. This allows for a more accurate differentiation between the first and second water environment monitoring indicators. Furthermore, it improves the fit between the first fluctuation level determined for each water environment monitoring indicator to be screened and the data characteristics of the indicator itself, and more accurately reflects the numerical variation characteristics of the water environment monitoring indicators to be screened, thereby accurately determining the effective water environment monitoring indicators.

[0137] In this embodiment of the disclosure, a first fluctuation level suitable for a water environment monitoring indicator to be screened can be determined under the condition that its relative range meets the range conditions. This improves the adaptability of the first fluctuation level to the water environment indicator to be screened and enhances the accuracy of the effective water environment monitoring indicators.

[0138] In one possible implementation, determining multiple effective water environment monitoring indicators from the multiple water environment monitoring indicators to be screened based on the effectiveness characteristics includes: determining the distribution characteristics of the first coefficient of variation corresponding to each of the multiple water environment monitoring indicators to be screened, the distribution characteristics including: the statistical value of the first coefficient of variation, or a first number of first coefficients of variation not less than a coefficient of variation threshold; determining the water environment monitoring indicators to be screened whose distribution characteristics satisfy the first effectiveness condition as the effective water environment monitoring indicators; or, for each second water environment monitoring indicator, calculating a second number of second coefficients of variation that satisfy the second effectiveness condition; determining the second water environment monitoring indicators whose second number is not less than a second number threshold as the effective water environment monitoring indicators; for each first water environment monitoring indicator, calculating a third number of standard deviations that satisfy the third effectiveness condition; determining the first water environment monitoring indicators whose third number is not less than a third number threshold as the effective water environment monitoring indicators.

[0139] The statistical value of the first coefficient of variation can be the median or mean of multiple first coefficients of variation corresponding to the same water environment monitoring indicator to be screened.

[0140] When the distribution characteristic is a statistical value, the first valid condition is that the statistical value is greater than a preset first statistical threshold. When the distribution characteristic is a first quantity of a first coefficient of variation that is not less than a threshold, the first valid condition is that the first quantity is greater than a first quantity threshold.

[0141] The first quantity can be the sum of the first sub-quantity of the first coefficient of variation falling within the first coefficient of variation threshold range and the second sub-quantity of the first coefficient of variation greater than the maximum value of the first coefficient of variation threshold range. For example, the first coefficient of variation threshold range is [0.2, 0.3]. The aforementioned coefficient of variation threshold can be the lower limit of the first coefficient of variation threshold range.

[0142] In this embodiment of the disclosure, water environment monitoring indicators whose distribution characteristics meet the first effective condition can be identified as effective water environment monitoring indicators.

[0143] The second validity condition can be falling within the first coefficient of variation threshold range, or exceeding the maximum value within the first coefficient of variation threshold range. For ease of description, the number of second coefficients of variation for the same second water environment monitoring indicator that satisfy the second validity condition is named the second quantity. In the embodiments of this disclosure, a second water environment monitoring indicator whose second quantity is not less than the second quantity threshold can be determined as a valid water environment monitoring indicator.

[0144] The third validity condition can be greater than a standard deviation threshold. For ease of description, the number of standard deviations of the same first water environment monitoring indicator that meet the third validity condition is named the third quantity. In the embodiments of this disclosure, the first water environment monitoring indicator whose third quantity is not less than the third quantity threshold can be determined as the valid water environment monitoring indicator.

[0145] In this embodiment of the disclosure, different methods (a first validity condition (using statistical values ​​or a first quantity), a second validity condition, or a third validity condition) can be used to screen the same water environment monitoring indicator to obtain effective water environment monitoring indicators. Therefore, the method of this embodiment can improve the flexibility of determining effective water environment monitoring indicators. Additionally, a method matching each water environment monitoring indicator to be screened can be used to obtain effective water environment monitoring indicators. For example, a second validity condition can be used for the second water environment monitoring indicator, and a third validity condition can be used for the first water environment monitoring indicator. Therefore, the method of this embodiment can improve the reliability of determining effective water environment monitoring indicators.

[0146] In one possible implementation, when the number of effective water environment monitoring indicators corresponding to a monitoring point is lower than the lower limit, the maximum value of the first coefficient of variation threshold range can be reduced, or both the maximum and minimum values ​​of the first coefficient of variation threshold range can be reduced. Additionally, the standard deviation threshold can be reduced.

[0147] In one possible implementation, indicators other than the effective water environment monitoring indicators among the plurality of water environment monitoring indicators to be screened can be regarded as ineffective indicators; time series segments of the ineffective indicators within a preset time period are obtained; ineffective indicators whose values ​​are less than the indicator threshold within the time series segments are determined as explicit indicators; the method further includes: when an explicit indicator is determined for any one of the monitoring points, the any one of the monitoring points is determined as the point where drought inflow infiltration occurs.

[0148] When the infiltration volume of external water far exceeds the sewage volume, the characteristics of the external water are relatively constant. For example, tap water might be mistakenly connected to the drainage network. When the infiltration volume of external water significantly exceeds the sewage volume, the coefficients of variation for all the water environment monitoring indicators to be screened may be very small. Therefore, it is difficult to select a sufficient number of effective water environment monitoring indicators. Furthermore, the coefficient of variation may indicate a stable water environment, but in reality, there may be serious problems. This could lead to the omission of monitoring points with severe infiltration issues.

[0149] In this embodiment of the disclosure, ineffective indicators can be processed to determine explicit indicators. Once explicit indicators are determined for a monitoring point, that monitoring point can be directly identified as a point where drought-induced infiltration occurs.

[0150] Specifically, a preset time period can be defined as the time period during which at least some water environment monitoring indicators in the drainage network show significant changes. For example, the time period with the highest sewage volume during the day can be used as the preset time period. Alternatively, since some water environment monitoring indicators in sewage tend to rise significantly in summer, summer can be used as the preset time period. Based on the preset time period, time series segments collected during the preset time period are extracted from the time series data of ineffective indicators. For ineffective indicators, an indicator threshold can be set. This threshold can be the lowest value of the ineffective indicator under normal conditions without inflow or infiltration. If all values ​​in the time series segment are less than the corresponding indicator threshold, or if a first proportion of values ​​in the time series segment are less than the corresponding indicator threshold, the ineffective indicator is determined as a visible indicator.

[0151] In this embodiment, after determining the effective water environment monitoring indicators, ineffective indicators can be processed. If a visible indicator can be identified after processing, the presence of inflow / infiltration problems at the monitoring point can be directly determined, eliminating subsequent steps and improving the accuracy and efficiency of identifying inflow / infiltration problems. Alternatively, ineffective indicators can be processed when sufficient effective water environment monitoring indicators cannot be determined. If a visible indicator can be identified, the number of monitoring points that have already experienced inflow / infiltration problems but are difficult to identify can be reduced, thereby improving the ability to identify monitoring points experiencing large-scale external water inflow / infiltration.

[0152] In one possible implementation, acquiring time-series data of multiple water environment monitoring indicators to be screened at any monitoring point in the target drainage network includes: acquiring the original time-series data of any water environment monitoring indicator to be screened; identifying outliers in the original time-series data of the water environment monitoring indicator to be screened; and obtaining the time-series data of the water environment monitoring indicator to be screened by using a corresponding processing strategy based on the duration of the outliers; the processing strategy includes: setting an invalidation flag for the outliers when the duration is greater than a duration threshold; and removing the outliers when the duration is not greater than the duration threshold.

[0153] In this embodiment of the disclosure, different strategies are used to process outliers in the original time series data according to the different durations of outliers, so as to effectively reduce the negative impact of outliers on the determination of target water environment monitoring indicators, thereby improving the quality of the determination of target water environment monitoring indicators.

[0154] If the duration exceeds the duration threshold, the outlier will be directly marked as invalid. This way, outliers will not participate in the steps of determining the target water environment monitoring indicators, effectively eliminating the negative impact of outliers.

[0155] If the duration is no greater than a duration threshold, it indicates that the data anomaly is short-lived, and the outlier can be removed, effectively eliminating its negative impact. Furthermore, since the data anomaly is short-lived, there is still a correlation between the normal values ​​before and after the outlier. Further, data analysis methods or time series prediction models can be used to predict the missing values ​​after outlier removal, and these predicted values ​​can be used to fill the gaps. The duration threshold can be set according to the density of time points in the time series and the needs of the actual scenario. For example, the duration threshold could be 3 hours.

[0156] In one possible implementation, the outlier includes one or more of a first outlier, a second outlier, and a third outlier. Determining the outlier in the original time series data of the water environment monitoring indicator to be screened includes: identifying a first outlier in the original time series data of the water environment monitoring indicator to be screened that does not meet a first preset condition; performing statistical analysis on the data in the original time series of the water environment monitoring indicator to be screened, excluding the first outlier, to determine a second outlier that exceeds the statistical range; and performing logical relationship verification on the data in the original time series of the water environment monitoring indicator to be screened, excluding the first and second outliers, and the data of at least one first water environment monitoring indicator to be screened that has a correlation with the water environment monitoring indicator to be screened, to determine a third outlier that does not conform to the logical relationship, wherein the data of the first water environment monitoring indicator to be screened is the data in the original time series of the first water environment monitoring indicator to be screened, excluding the first and second outliers.

[0157] The first outlier can be an unreasonable value. For example, a flow rate value less than 0, a liquid level value exceeding the top of the pipe, or a water quality value exceeding the range of the measuring instrument. The first preset condition can correspond to the water environment monitoring indicator to be screened. Failure to meet the first preset condition indicates that the value does not conform to the physical characteristics of the pipeline.

[0158] The second outlier can be a value that does not conform to the statistical distribution. For example, the second outlier can be an extreme value outside the preset statistical range. Box plots or the Laida criterion can be used to identify second outlier extreme values.

[0159] The third outlier is a value where the value of the water environment monitoring indicator to be screened logically contradicts at least one of the first water environment monitoring indicators to be screened that is related to it. Here, "related" means that the two water environment monitoring indicators to be screened in the drainage network are complementary, reflecting water quality and quantity characteristics from multiple perspectives. In the field of drainage network monitoring, no single water environment monitoring indicator can independently and comprehensively reflect the quality of a water body. Typically, different water environment monitoring indicators can reflect the characteristics and pollution status of water bodies from different angles and levels. Only by combining, verifying, and complementing different water environment monitoring indicators can the quality of the water environment be completely and accurately reflected. This characteristic between different water environment monitoring indicators is called complementary nature. The first water environment monitoring indicator to be screened is another water environment monitoring indicator that is different from the one to be screened.

[0160] For example, liquid level and flow rate have a positive correlation. If the first candidate value in the liquid level index or the second candidate value in the flow rate index causes the positive correlation between liquid level and flow rate to be lost, then the first candidate value and the flow rate value in the flow rate index that has the same time point as the first candidate value are all third outliers; or the second candidate value and the liquid level value in the liquid level index that has the same time point as the second candidate value are all third outliers.

[0161] For example, chemical oxygen demand (COD) and rainfall have a negative correlation. If the third candidate value in the COD index or the fourth candidate value in the rainfall index causes COD and rainfall to no longer satisfy the negative correlation, then the third candidate value, as well as the rainfall value in the rainfall index that has the same time point as the third candidate value, are all considered third outliers; or the fourth candidate value, as well as the COD value in the COD index that has the same time point as the fourth candidate value, are all considered third outliers.

[0162] In this disclosed example, a logical relationship verification can be performed on the original time series of the water environment monitoring indicators to be screened and the first water environment monitoring indicator to be screened that is related to them, after removing the first and second outliers. This logical relationship verification can be causal detection or correlation calculation, etc.

[0163] In this embodiment of the disclosure, a first outlier is first determined from the original time series data of the water environment monitoring indicators to be screened; then a second outlier is determined from the original time series data of the water environment monitoring indicators to be screened that does not contain the first outlier; and then a third outlier is extracted from the original time series data of the water environment monitoring indicators to be screened and the first water environment monitoring indicator to be screened that does not contain the first and second outliers.

[0164] This disclosure proposes a step-by-step method for determining three types of outliers, enabling precise screening from illogical errors to data distribution errors and then to logical errors. First, physical range verification uses deterministic rules to quickly eliminate illogical errors, directly purifying the data foundation. Next, statistical distribution verification avoids outliers contaminating the mean and variance, ensuring that box plots or Laida criteria reflect distributions of actual operating conditions, thus more accurately identifying measurement jumps or random disturbances. Finally, logical relationship verification, under the premise of physically reasonable and statistically compliant data, specifically captures hidden contradictions that violate the operational logic of the drainage network (such as reversed liquid level and flow rate, or inconsistent water quality patterns during rainy days), making the anomaly identification results physically interpretable. Furthermore, this sequence improves processing efficiency and ensures that the screening of the second and third types of outliers is not affected by other outliers, allowing for accurate and rapid identification of each of the three types of outliers. Overall, it improves the efficiency and accuracy of outlier identification.

[0165] In one example, raw time-series data of multiple water environment monitoring indicators to be screened from monitoring points can be obtained. The raw time-series data is unprocessed data collected by monitoring equipment. For each water environment monitoring indicator to be screened, a first outlier that does not meet a first preset condition is identified from the raw time-series data of the indicator. A second outlier that does not meet a second preset condition is identified from the raw time-series data of the indicator that does not contain the first outlier. A third outlier that does not meet a third preset condition is identified and removed from the raw time-series data of the indicator that does not contain the first and second outliers. The duration of the outlier is also determined. If the duration exceeds a duration threshold, an outlier is flagged as invalid. If the duration is not greater than the duration threshold, the outlier is removed. Thus, the time-series data of each water environment monitoring indicator to be screened can be obtained.

[0166] In this example, the time series data of each water environment monitoring indicator to be screened are evenly divided into multiple time series segments. The first coefficient of variation for each time series segment of each water environment monitoring indicator to be screened is determined. The first coefficient of variation is used to characterize the first degree of fluctuation of the water environment monitoring indicator to be screened. The statistical values ​​of the multiple first coefficients of variation for the water environment monitoring indicators to be screened are determined. Water environment monitoring indicators to be screened whose statistical values ​​are greater than a first statistical threshold are considered effective water environment monitoring indicators. Alternatively, the first number of water environment monitoring indicators to be screened whose multiple first coefficients of variation fall within the first coefficient of variation threshold range and whose values ​​exceed the maximum value of the first coefficient of variation threshold range are determined, and water environment monitoring indicators to be screened whose first number exceeds a first number threshold are considered effective water environment monitoring indicators.

[0167] Alternatively, multiple preset quantiles are determined for the historical time series data of each of the water environment monitoring indicators to be screened. Based on the multiple quantiles of the water environment monitoring indicators to be screened, the relative range of the water environment monitoring indicators to be screened is determined. The water environment monitoring indicators to be screened with a relative range less than a range threshold are determined as the first water environment monitoring indicators. The water environment monitoring indicators to be screened with a relative range not less than a range threshold are determined as the second water environment monitoring indicators. The historical time series data of each of the first and second water environment monitoring indicators are evenly divided into multiple time series segments. The standard deviation of each time series segment of the first water environment monitoring indicator is determined. For the water environment monitoring indicators to be screened, a third number of standard deviations greater than a standard deviation threshold is determined. The first water environment monitoring indicators with a third number not less than a third number threshold are determined as effective water environment monitoring indicators. Furthermore, a second coefficient of variation is determined for each time series segment of the second water environment monitoring indicator. For the water environment monitoring indicators to be screened, a second number of second coefficients of variation falling within the first coefficient of variation threshold range and greater than the maximum value of the second coefficient of variation within the first coefficient of variation threshold range is determined. The second water environment monitoring indicator with a second quantity not less than the second quantity threshold is determined as the effective water environment monitoring indicator.

[0168] When the number of effective water environment monitoring indicators is below the lower limit, the maximum value of the first coefficient of variation threshold range can be reduced, or both the maximum and minimum values ​​of the first coefficient of variation threshold range can be reduced. Additionally, the standard deviation threshold can be reduced.

[0169] In this example, after determining the effective water environment monitoring indicators, indicators other than the effective water environment monitoring indicators among multiple water environment monitoring indicators to be screened can be regarded as ineffective indicators; time series segments of ineffective indicators within a preset time period can be obtained; ineffective indicators with values ​​less than the indicator threshold within the time series segment can be identified as explicit indicators; and when explicit indicators are identified, the monitoring points corresponding to the explicit indicators can be identified as the points where inflow and infiltration occur.

[0170] In this example, the time series data of effective water environment monitoring indicators can be divided into dry day time series data and rainy day time series data according to weather scenarios, resulting in data groups for effective water environment monitoring indicators. Hypothesis testing is performed on each data group of effective water environment monitoring indicators to obtain the test results for each effective water environment monitoring indicator. The test results are used as sensitivity features. Based on the sensitivity features, multiple sensitive water environment monitoring indicators are determined from the multiple effective water environment monitoring indicators. Furthermore, the second degree of fluctuation of each sensitive water environment monitoring indicator can be determined. The second degree of fluctuation includes: peak-to-valley ratio, magnitude of numerical decline, and rate of numerical rise. The sensitive water environment monitoring indicators are sorted according to peak-to-valley ratio, magnitude of numerical decline, and rate of numerical rise, respectively. Starting from the top of each sort, a preset fourth selection number of sensitive water environment monitoring indicators are obtained and used as robust water environment monitoring indicators.

[0171] In this example, the third degree of fluctuation for each robust water environment monitoring indicator can be determined. The third degree of fluctuation can be represented by the peak-to-trough ratio, the magnitude of numerical decline, and the rate of numerical rise for each robust time series segment in the time series data. For each robust water environment monitoring indicator, a curve can be plotted with the third degree of fluctuation on the horizontal axis and rainfall (the total rainfall within the corresponding robust time series segment) on the vertical axis. Based on the curve, the response relationship between the third degree of fluctuation of the time series data of the corresponding robust water environment monitoring indicator and rainfall can be determined. Based on the response relationship, the target model corresponding to the robust water environment monitoring indicator is determined, and the target model is used to fit the third degree of fluctuation of the time series data of the robust water environment monitoring indicator to rainfall, obtaining the robustness characteristics of the robust water environment monitoring indicator, namely the slope of numerical change and the coefficient of determination.

[0172] In this example, robust water environment monitoring indicators whose numerical change slope is greater than a slope threshold and whose coefficient of determination is greater than a coefficient of determination threshold are identified as target water environment monitoring indicators. These target water environment monitoring indicators can be used to determine inflow and infiltration issues at the corresponding monitoring points during rainy weather scenarios.

[0173] In addition, robust water environment monitoring indicators whose numerical change slope is greater than a slope threshold and whose coefficient of determination is not greater than a coefficient of determination threshold can be used as alternative indicators. When the number of target water environment monitoring indicators is insufficient, or when additional water environment monitoring indicators are needed as supplementary measures, alternative indicators and target water environment monitoring indicators can be used together to diagnose inflow and infiltration problems.

[0174] Figure 2 A schematic flowchart illustrating the method for determining inflow and infiltration diagnostic indicators provided in this embodiment of the disclosure. The device 20 includes:

[0175] The time series data determination unit 21 is used to acquire time series data of multiple water environment monitoring indicators to be screened at any monitoring point in the target drainage network. The multiple water environment monitoring indicators to be screened include: flow rate, liquid level and multiple water quality indicators.

[0176] The effective feature determination unit 22 is used to determine the first fluctuation degree of the time series data of each of the water environment monitoring indicators to be screened, and use the first fluctuation degree as the effectiveness feature. The effectiveness feature is used to exclude water environment monitoring indicators to be screened that indicate an abnormality in the health status of the target drainage network under arid weather scenarios.

[0177] Effective water environment indicator determination unit 23 is used to determine multiple effective water environment monitoring indicators from the multiple water environment monitoring indicators to be screened based on the effectiveness characteristics;

[0178] Sensitivity feature determination unit 24 is used to perform hypothesis testing on the time series data of each of the effective water environment monitoring indicators, obtain the test results of each of the effective water environment monitoring indicators, and use the test results as sensitivity features. The sensitivity features are used to screen effective water environment monitoring indicators that are sensitive to rainfall.

[0179] Sensitive water environment monitoring index determination unit 25 is used to determine multiple sensitive water environment monitoring indicators from the multiple effective water environment monitoring indicators based on the sensitivity characteristics.

[0180] The second fluctuation degree determination unit 26 is used to determine the second fluctuation degree of the time series data of each of the sensitive water environment monitoring indicators;

[0181] The target water environment monitoring indicator determination unit 27, based on the second fluctuation level, determines the target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any one of the monitoring points in a rainy weather scenario from the plurality of sensitive water environment monitoring indicators.

[0182] In one possible implementation, the target water environment monitoring index determination unit 27 is further configured to:

[0183] Based on the second degree of fluctuation of the time series data of each of the aforementioned sensitive water environment monitoring indicators, a plurality of robust water environment monitoring indicators are determined from the plurality of sensitive water environment monitoring indicators.

[0184] The third fluctuation degree of the time series data of each robust water environment monitoring index is fitted with the rainfall to obtain the slope of the numerical change that characterizes the sensitivity of the numerical change of each robust water environment monitoring index to the response of rainfall, and the coefficient of determination that characterizes the goodness of fit of each robust water environment monitoring index. The slope of the numerical change and the coefficient of determination are used as robustness features.

[0185] Based on the robustness characteristics, a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any of the monitoring points in rainy weather scenarios is determined from the plurality of robust water environment monitoring indicators.

[0186] In one possible implementation, the target water environment monitoring index determination unit 27 is further configured to:

[0187] Based on the third degree of fluctuation of the time series data of each of the robust water environment monitoring indicators and the response relationship with rainfall, target models are determined for each of the robust water environment monitoring indicators.

[0188] Based on the target model, the third fluctuation degree of the time series data of the robust water environment monitoring index corresponding to the target model is fitted with the rainfall to obtain the slope of the numerical change and the coefficient of determination.

[0189] In one possible implementation, the sensitivity feature determination unit 24 is further configured to:

[0190] For any of the time series data of the effective water environment monitoring indicator, the time series data of the effective water environment monitoring indicator is divided into dry day time series data and rainy day time series data according to the weather scenario to obtain the data group of the effective water environment monitoring indicator;

[0191] Hypothesis testing was performed on the data set.

[0192] In one possible implementation, the effective feature determination unit 22 is further configured to:

[0193] The time series data of each of the water environment monitoring indicators to be screened are divided into multiple time series segments;

[0194] Determine the first coefficient of variation for each time segment among multiple time segments of the water environment monitoring indicators to be screened, wherein the first coefficient of variation characterizes the first degree of fluctuation;

[0195] or,

[0196] Determine multiple preset quantiles for the historical time series data of each of the water environment monitoring indicators to be screened;

[0197] Based on multiple quantiles, a relative range is determined. Any water environment monitoring indicator to be screened that satisfies the first range condition is taken as the first water environment monitoring indicator, and any water environment monitoring indicator to be screened that satisfies the second range condition is taken as the second water environment monitoring indicator.

[0198] The time series data of the first water environment monitoring indicator and the second water environment monitoring indicator are each divided into multiple time series segments;

[0199] Determine the standard deviation of each time segment of the first water environment monitoring indicator;

[0200] The second coefficient of variation for each time segment of the second water environment monitoring indicator is determined, wherein the standard deviation and the second coefficient of variation characterize the first degree of fluctuation.

[0201] In one possible implementation, the effective water environment index determination unit 23 is further configured to:

[0202] Determine the distribution characteristics of the first coefficient of variation corresponding to each of the plurality of water environment monitoring indicators to be screened, wherein the distribution characteristics include: the statistical value of the first coefficient of variation, or the first number of the first coefficient of variation that is not less than the coefficient of variation threshold;

[0203] The water environment monitoring indicators whose distribution characteristics meet the first effective condition are identified as the effective water environment monitoring indicators.

[0204] or,

[0205] For each of the second water environment monitoring indicators, the second number of the second coefficients of variation that meet the second effective conditions is calculated.

[0206] The second water environment monitoring indicator whose second quantity is not less than the second quantity threshold is determined as the effective water environment monitoring indicator;

[0207] For each of the first water environment monitoring indicators, the number of standard deviations that meet the third validity condition is calculated.

[0208] The first water environment monitoring indicator whose third quantity is not less than the third quantity threshold is determined as the effective water environment monitoring indicator.

[0209] In one possible implementation, the time series data determination unit 21 is further configured to:

[0210] For any water environment monitoring indicator to be screened, obtain the original time series data of that water environment monitoring indicator;

[0211] Identify outliers in the original time series data of the water environment monitoring indicator to be screened;

[0212] Based on the duration of the outlier, the time series data of the water environment monitoring indicator to be screened are obtained using the corresponding processing strategy.

[0213] The processing strategy includes: setting a failure flag for outliers when the duration exceeds a duration threshold;

[0214] Furthermore, if the duration is not greater than the duration threshold, the outlier is removed.

[0215] In one possible implementation, the outlier includes one or more of a first outlier, a second outlier, and a third outlier, and the time series data determination unit 21 is further configured to:

[0216] Identify the first outlier in the original time series data of the water environment monitoring indicator to be screened that does not meet the first preset condition;

[0217] Statistical analysis is performed on the data in the original time series of the water environment monitoring indicator to be screened, excluding the first outlier, to identify the second outlier that exceeds the statistical range.

[0218] The data in the original time series of the water environment monitoring indicator to be screened, excluding the first outlier and the second outlier, and the data of at least one first water environment monitoring indicator that is related to the water environment monitoring indicator to be screened, are subjected to logical relationship verification to determine the third outlier that does not conform to the logical relationship. The data of the first water environment monitoring indicator to be screened are the data in the original time series of the first water environment monitoring indicator to be screened, excluding the first outlier and the second outlier.

[0219] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0220] This disclosure also provides a system for determining inflow and infiltration diagnostic indicators, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0221] This disclosure also provides a non-volatile computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described method.

[0222] This disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described method.

[0223] Figure 3 This is a schematic diagram of the structure of an inflow / infiltration diagnostic index determination system provided in an embodiment of this disclosure. For example, system 1900 can be provided as a server or terminal device. (Refer to...) Figure 3 System 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0224] System 1900 may also include a power supply component 1926 configured to perform power management of system 1900, a wired or wireless network interface 1950 configured to connect system 1900 to a network, and an input / output interface 1958 (I / O interface). System 1900 can operate on an operating system, such as Windows Server, stored in memory 1932. TM macOS X TM Unix TM Linux TM FreeBSD TM Or similar.

[0225] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions that can be executed by a processing component 1922 of system 1900 to perform the above-described method.

[0226] Computer-readable storage media can be tangible devices capable of holding and storing programs / instructions used by instruction execution devices. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0227] The computer program (or computer-readable program instructions) described herein can be downloaded from a computer-readable storage medium to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage medium in the respective computing / processing device.

[0228] The computer program (or computer program instructions) used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions to implement various aspects of this disclosure.

[0229] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0230] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0231] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0232] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0233] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for determining infiltration diagnostic indicators, characterized in that, include: Obtain time series data of multiple water environment monitoring indicators to be screened at any monitoring point in the target drainage network. The multiple water environment monitoring indicators to be screened include: flow rate, liquid level and multiple water quality indicators. Determine the first degree of fluctuation of the time series data of each of the water environment monitoring indicators to be screened, and use the first degree of fluctuation as a validity feature. The validity feature is used to exclude water environment monitoring indicators to be screened that indicate an abnormality in the health status of the target drainage network under arid weather conditions. Based on the aforementioned effectiveness characteristics, multiple effective water environment monitoring indicators are determined from the plurality of water environment monitoring indicators to be screened. Hypothesis tests were performed on the time series data of each of the aforementioned effective water environment monitoring indicators to obtain the test results of each of the aforementioned effective water environment monitoring indicators. The test results were used as sensitivity features, and the sensitivity features were used to screen effective water environment monitoring indicators that are sensitive to rainfall. Based on the aforementioned sensitivity characteristics, multiple sensitive water environment monitoring indicators are determined from the multiple effective water environment monitoring indicators. Determine the second degree of fluctuation in the time series data of each of the aforementioned sensitive water environment monitoring indicators; Based on the second fluctuation level, a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any of the monitoring points under rainy weather scenarios is determined from the plurality of sensitive water environment monitoring indicators.

2. The method according to claim 1, characterized in that, Based on the second fluctuation level, the target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any monitoring point under rainy weather conditions is determined from the plurality of sensitive water environment monitoring indicators, including: Based on the second degree of fluctuation of the time series data of each of the aforementioned sensitive water environment monitoring indicators, a plurality of robust water environment monitoring indicators are determined from the plurality of sensitive water environment monitoring indicators. The third fluctuation degree of the time series data of each robust water environment monitoring index is fitted with the rainfall to obtain the slope of the numerical change that characterizes the sensitivity of the numerical change of each robust water environment monitoring index to the response of rainfall, and the coefficient of determination that characterizes the goodness of fit of each robust water environment monitoring index. The slope of the numerical change and the coefficient of determination are used as robustness features. Based on the robustness characteristics, a target water environment monitoring indicator suitable for diagnosing inflow and infiltration problems at any of the monitoring points in rainy weather scenarios is determined from the plurality of robust water environment monitoring indicators.

3. The method according to claim 2, characterized in that, The fitting of the third fluctuation degree of the time series data of each robust water environment monitoring index with rainfall yields the slope of the numerical change characterizing the sensitivity of the numerical change of each robust water environment monitoring index to the response of rainfall, and the determination coefficient characterizing the goodness of fit of each robust water environment monitoring index, including: Based on the third degree of fluctuation of the time series data of each of the robust water environment monitoring indicators and the response relationship with rainfall, target models are determined for each of the robust water environment monitoring indicators. Based on the target model, the third fluctuation degree of the time series data of the robust water environment monitoring index corresponding to the target model is fitted with the rainfall to obtain the slope of the numerical change and the coefficient of determination.

4. The method according to claim 1, characterized in that, The hypothesis testing is performed on the time series data of each of the aforementioned effective water environment monitoring indicators, including: For any of the time series data of the effective water environment monitoring indicator, the time series data of the effective water environment monitoring indicator is divided into dry day time series data and rainy day time series data according to the weather scenario to obtain the data group of the effective water environment monitoring indicator; Hypothesis testing was performed on the data set.

5. The method according to claim 1, characterized in that, The determination of the first degree of fluctuation of the time series data of each of the water environment monitoring indicators to be screened includes: The time series data of each of the water environment monitoring indicators to be screened are divided into multiple time series segments; Determine the first coefficient of variation for each time segment among multiple time segments of the water environment monitoring indicators to be screened, wherein the first coefficient of variation characterizes the first degree of fluctuation; or, Determine multiple preset quantiles for the historical time series data of each of the water environment monitoring indicators to be screened; Based on multiple quantiles, a relative range is determined. Any water environment monitoring indicator to be screened that satisfies the first range condition is taken as the first water environment monitoring indicator, and any water environment monitoring indicator to be screened that satisfies the second range condition is taken as the second water environment monitoring indicator. The time series data of the first water environment monitoring indicator and the second water environment monitoring indicator are each divided into multiple time series segments; Determine the standard deviation of each time segment of the first water environment monitoring indicator; The second coefficient of variation for each time segment of the second water environment monitoring indicator is determined, wherein the standard deviation and the second coefficient of variation characterize the first degree of fluctuation.

6. The method according to claim 5, characterized in that, Based on the effectiveness characteristics, multiple effective water environment monitoring indicators are determined from the multiple water environment monitoring indicators to be screened, including: Determine the distribution characteristics of the first coefficient of variation corresponding to each of the plurality of water environment monitoring indicators to be screened, wherein the distribution characteristics include: the statistical value of the first coefficient of variation, or the first number of the first coefficient of variation that is not less than the coefficient of variation threshold; The water environment monitoring indicators whose distribution characteristics meet the first effective condition are identified as the effective water environment monitoring indicators. or, For each of the second water environment monitoring indicators, the second number of the second coefficients of variation that meet the second effective conditions is calculated. The second water environment monitoring indicator whose second quantity is not less than the second quantity threshold is determined as the effective water environment monitoring indicator; For each of the first water environment monitoring indicators, the number of standard deviations that meet the third validity condition is calculated. The first water environment monitoring indicator whose third quantity is not less than the third quantity threshold is determined as the effective water environment monitoring indicator.

7. The method according to claim 1, characterized in that, The acquisition of time series data of multiple water environment monitoring indicators to be screened at any monitoring point in the target drainage network includes: For any water environment monitoring indicator to be screened, obtain the original time series data of that water environment monitoring indicator; Identify outliers in the original time series data of the water environment monitoring indicator to be screened; Based on the duration of the outlier, the time series data of the water environment monitoring indicator to be screened are obtained using the corresponding processing strategy. The processing strategy includes: setting a failure flag for outliers when the duration exceeds a duration threshold; Furthermore, if the duration is not greater than the duration threshold, the outlier is removed.

8. The method according to claim 7, characterized in that, The outliers include one or more of the following: a first outlier, a second outlier, and a third outlier. The process of determining the outliers in the original time series data of the water environment monitoring indicator to be screened includes: Identify the first outlier in the original time series data of the water environment monitoring indicator to be screened that does not meet the first preset condition; Statistical analysis is performed on the data in the original time series of the water environment monitoring indicator to be screened, excluding the first outlier, to identify the second outlier that exceeds the statistical range. The data in the original time series of the water environment monitoring indicator to be screened, excluding the first outlier and the second outlier, and the data of at least one first water environment monitoring indicator that is related to the water environment monitoring indicator to be screened, are subjected to logical relationship verification to determine the third outlier that does not conform to the logical relationship. The data of the first water environment monitoring indicator to be screened are the data in the original time series of the first water environment monitoring indicator to be screened, excluding the first outlier and the second outlier.

9. A system for determining infiltration diagnostic indicators, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

10. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images