A self-adaptive correction and prediction method and system for surface water environmental monitoring data anomaly

Abstract

The application relates to a kind of surface water environment monitoring data exception self-adapting correction prediction method and system, comprising: according to water quality automatic monitoring data and the position parameter of corresponding automatic monitoring point, detect the exception data in water quality automatic monitoring data;According to water quality manual monitoring data and the position parameter of corresponding manual monitoring position, the water quality manual monitoring data prediction value at the current automatic monitoring point at the beginning of monitoring period is obtained by prediction;According to the meteorological monitoring data of current time and the meteorological monitoring data at the beginning of monitoring period, obtain the meteorological change amplitude;According to the meteorological change amplitude, the water quality manual monitoring data prediction value at the current automatic monitoring point at the beginning of monitoring period is adjusted, and the exception data in water quality automatic monitoring data is corrected according to the adjustment value.The application can improve the quality and availability of monitoring data, and provide more powerful data support for water environment protection and management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water environment, and in particular to a self-adaptive correction and prediction method and system for surface water environment monitoring data anomalies. BACKGROUND

[0002] With the rapid development of industrialization and urbanization, the quality of surface water environment is increasingly concerned. Surface water environment monitoring is crucial for understanding water quality, assessing water ecological health, and ensuring the sustainable use of water resources. By obtaining accurate monitoring data, water pollution problems can be detected in a timely manner, providing a scientific basis for water environment protection decisions. However, in actual monitoring processes, surface water environment monitoring data may be disturbed by various factors, resulting in abnormal data, such as monitoring equipment failure, data transmission errors, and human tampering. If these abnormal data cannot be monitored and processed in a timely and accurate manner, it will mislead environmental decisions and affect the scientificity and effectiveness of environmental protection work.

[0003] Currently, the processing method for surface water environment monitoring data anomalies has certain limitations. Traditional data processing methods rely on manual experience judgment and simple statistical analysis, such as setting fixed thresholds to determine whether data is abnormal. However, this approach is difficult to adapt to complex and changing surface water environments, and setting thresholds too high or too low may lead to misjudgment and missed judgment of abnormal data. In rivers with frequent water quality changes, fixed thresholds cannot capture subtle changes in water quality, resulting in some abnormal surface water environment monitoring data not being effectively identified. For water bodies with seasonal changes, the same threshold may not be applicable in different seasons. Moreover, traditional methods often lack in-depth analysis of the trend and correlation of surface water environment monitoring data when processing abnormal data, making it difficult to accurately distinguish between real anomalies caused by environmental changes and anomalies caused by monitoring errors, resulting in poor accuracy and reliability of data correction, and failing to provide precise data support for water environment management.

[0004] Therefore, how to improve the quality and usability of monitoring data to provide stronger data support for water environment protection and management is a technical problem that needs to be solved by technical personnel in the field. SUMMARY

[0005] The present application provides a self-adaptive correction and prediction method and system for surface water environment monitoring data anomalies to improve the quality and usability of monitoring data and provide stronger data support for water environment protection and management.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] The adaptive correction and prediction method for surface water environment monitoring data anomaly comprises the following steps: step S110, obtaining water quality automatic monitoring data of surface water environment by automatic monitoring point in a monitoring period; step S120, detecting abnormal data in the obtained water quality automatic monitoring data according to the water quality automatic monitoring data and the position parameters of the corresponding automatic monitoring point; step S130, obtaining water quality manual monitoring data prediction value at the current automatic monitoring point at the beginning of the monitoring period according to the water quality manual monitoring data obtained at the beginning of the monitoring period and the position parameters of the corresponding manual monitoring position; step S140, in response to obtaining abnormal data in the water quality automatic monitoring data, obtaining meteorological monitoring data in the monitoring range at the current time, and obtaining meteorological change amplitude according to the meteorological monitoring data at the current time and the meteorological monitoring data at the beginning of the monitoring period; step S150, adjusting the water quality manual monitoring data prediction value at the current automatic monitoring point at the beginning of the monitoring period according to the meteorological change amplitude, and correcting the abnormal data in the water quality automatic monitoring data according to the adjustment value.

[0008] The adaptive correction and prediction method for surface water environment monitoring data anomaly, wherein preferably, step S120 comprises the following sub-steps: step S121, obtaining the relative distance between the upstream automatic monitoring point and the current automatic monitoring point according to the position parameters of the current automatic monitoring point and the upstream automatic monitoring point; step S122, obtaining the influence weight of the upstream water quality automatic monitoring data on the prediction value of the current water quality automatic monitoring data according to the upstream water quality automatic monitoring data and the relative distance between the upstream automatic monitoring point and the current automatic monitoring point; step S123, obtaining the prediction value of the current water quality automatic monitoring data according to the upstream water quality automatic monitoring data and its influence weight; step S124, comparing the prediction value of the current water quality automatic monitoring data with the current water quality automatic monitoring data to detect whether the current water quality automatic monitoring data is abnormal data.

[0009] The adaptive correction and prediction method for surface water environment monitoring data anomaly as described above, preferably, the step S130 comprises the following sub-steps: step S131, obtaining the distance between the current automatic monitoring point and the upstream / downstream manual monitoring position according to the position parameters of the current automatic monitoring point and the position parameters of the upstream / downstream manual monitoring position; step S132, calculating the influence weight of the upstream / downstream water quality manual monitoring data on the predicted value of the water quality manual monitoring data at the current automatic monitoring point according to the distance between the current automatic monitoring point and the upstream / downstream manual monitoring position; and step S133, obtaining the predicted value of the water quality manual monitoring data at the current automatic monitoring point at the beginning of the monitoring period according to the influence weight of the upstream / downstream water quality manual monitoring data on the predicted value of the water quality manual monitoring data at the current automatic monitoring point and the upstream / downstream water quality manual monitoring data.

[0010] The adaptive correction and prediction method for surface water environment monitoring data anomaly as described above, preferably, the expression of the meteorological change amplitude is: wherein, is the meteorological change amplitude at t1 relative to t; P tj is the jth meteorological monitoring data at t; is the jth meteorological monitoring data at t1; is the weight value of the jth meteorological monitoring data; and J is the number of obtained meteorological monitoring data.

[0011] The adaptive correction and prediction method for surface water environment monitoring data anomaly as described above, preferably, the step S150 comprises the following sub-steps: step S151, judging whether the meteorological change amplitude exceeds the threshold value; step S152, if not, adjusting the meteorological change amplitude by using the fine tuning coefficient; and step S153, if yes, adjusting the meteorological change amplitude by using the adjustment coefficient.

[0012] The adaptive correction and prediction system for surface water environment monitoring data anomaly comprises an automatic monitoring point, a manual monitoring tool, a meteorological monitoring platform and a surface water environment monitoring platform; wherein the surface water environment monitoring platform comprises an abnormal data detection unit, a predicted value prediction unit, a change amplitude calculation unit and an abnormal data correction unit; in a monitoring period, the automatic monitoring point obtains water quality automatic monitoring data of the surface water environment; the abnormal data detection unit detects abnormal data in the obtained water quality automatic monitoring data according to the water quality automatic monitoring data and the position parameters of the corresponding automatic monitoring point; the predicted value prediction unit predicts the water quality manual monitoring data prediction value of the current automatic monitoring point at the beginning of the monitoring period according to the water quality manual monitoring data obtained by the manual monitoring tool at the beginning of the monitoring period and the position parameters of the corresponding manual monitoring position; in response to obtaining the abnormal data in the water quality automatic monitoring data, the meteorological monitoring platform obtains the meteorological monitoring data in the monitoring range at the current time, and the change amplitude calculation unit obtains the meteorological change amplitude according to the meteorological monitoring data at the current time and the meteorological monitoring data at the beginning of the monitoring period; the abnormal data correction unit adjusts the water quality manual monitoring data prediction value of the current automatic monitoring point at the beginning of the monitoring period according to the meteorological change amplitude, and corrects the abnormal data in the water quality automatic monitoring data according to the adjusted value.

[0013] The adaptive correction and prediction system for surface water environment monitoring data anomaly as described above, wherein preferably, the abnormal data detection unit obtains the relative distance between the upstream automatic monitoring point and the current automatic monitoring point according to the position parameters of the current automatic monitoring point and the upstream automatic monitoring point; the abnormal data detection unit obtains the influence weight of the upstream water quality automatic monitoring data on the prediction value of the current water quality automatic monitoring data according to the upstream water quality automatic monitoring data and the relative distance between the upstream automatic monitoring point and the current automatic monitoring point; the abnormal data detection unit obtains the prediction value of the current water quality automatic monitoring data according to the upstream water quality automatic monitoring data and the influence weight; the abnormal data detection unit compares the prediction value of the current water quality automatic monitoring data with the current water quality automatic monitoring data to detect whether the current water quality automatic monitoring data is abnormal data.

[0014] The adaptive correction and prediction system for surface water environment monitoring data anomaly as described above, preferably, the prediction value prediction unit obtains the distance between the current automatic monitoring point and the upstream / downstream manual monitoring position according to the position parameters of the current automatic monitoring point and the position parameters of the upstream / downstream manual monitoring position; the prediction value prediction unit calculates the influence weight of the upstream / downstream water quality manual monitoring data on the water quality manual monitoring data prediction value at the current automatic monitoring point according to the distance between the current automatic monitoring point and the upstream / downstream manual monitoring position; and the prediction value prediction unit obtains the water quality manual monitoring data prediction value at the current automatic monitoring point at the beginning of the monitoring period according to the influence weight of the upstream / downstream water quality manual monitoring data on the water quality manual monitoring data prediction value at the current automatic monitoring point and the upstream / downstream water quality manual monitoring data.

[0015] The adaptive correction and prediction system for surface water environment monitoring data anomaly as described above, preferably, the expression of the meteorological change amplitude is: wherein, is the meteorological change amplitude at t1 relative to t; P tj is the jth meteorological monitoring data at t; is the jth meteorological monitoring data at t1; is the weight value of the jth meteorological monitoring data; and J is the number of obtained meteorological monitoring data.

[0016] The adaptive correction and prediction system for surface water environment monitoring data anomaly as described above, preferably, the anomaly data correction unit judges whether the meteorological change amplitude exceeds a threshold value; if not, the anomaly data correction unit adjusts the meteorological change amplitude by using a fine tuning coefficient; and if so, the anomaly data correction unit adjusts the meteorological change amplitude by using an adjustment coefficient.

[0017] Compared with the prior art, in the present application, the upstream water quality automatic monitoring data is used to identify the anomaly of the downstream water quality automatic monitoring data, so that the anomaly identification depends on the change trend of the surface water environment monitoring data and the correlation between the upstream and downstream water environment monitoring data, thereby the real anomaly caused by environmental change and the anomaly data caused by monitoring error can be more accurately distinguished, which provides an accurate basis for subsequent data correction. In addition, in the present application, the water quality manual monitoring data and the meteorological monitoring data are used to correct the abnormal water quality automatic monitoring data, so that the correction of the abnormal data is more accurate and more consistent with the actual situation, thereby providing an accurate basis for the prediction of water quality change in the future period of time and providing forward-looking support for environmental decision-making. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description only show some embodiments of the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0019] Figure 1 is a flow chart of an adaptive correction prediction method for surface water environment monitoring data anomalies;

[0020] Figure 2 is a flow chart of detecting abnormal data in automatic monitoring data of water quality;

[0021] Figure 3 is a flow chart of predicting the predicted value of manual monitoring data of water quality at the current automatic monitoring point at the start time of the monitoring period;

[0022] Figure 4 is a flow chart of adjusting the predicted value of manual monitoring data of water quality at the current automatic monitoring point at the start time of the monitoring period;

[0023] Figure 5 is a schematic diagram of an adaptive correction prediction system for surface water environment monitoring data anomalies. DETAILED DESCRIPTION

[0024] The embodiments of the present application will be described in detail below, and examples of the embodiments are shown in the drawings, wherein the same or similar reference signs represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are only used to explain the present application, and cannot be interpreted as a limitation on the present application.

[0025] Embodiment one

[0026] As shown in Figure 1 The present application provides an adaptive correction prediction method for surface water environment monitoring data anomalies, comprising the following steps:

[0027] Step S110, in a monitoring period, the automatic monitoring point obtains water quality automatic monitoring data of the surface water environment;

[0028] Automatic monitoring points (for example, monitoring instruments) are arranged at different sections of a surface water system, thereby forming an automatic monitoring network of the surface water system. In a monitoring period T, all automatic monitoring points in the automatic monitoring network will perform automatic monitoring according to a predetermined automatic monitoring frequency, thereby continuously obtaining water quality automatic monitoring data, and the automatic monitoring points will send the obtained water quality automatic monitoring data to a surface water environment monitoring platform. For example, the automatic monitoring point D i At time t, the water quality automatic monitoring data A is automatically monitored and obtainedit1 A it2 ..., where A it1 For automatic monitoring point D i The first automatic water quality monitoring data obtained at time t, A it2 For automatic monitoring point D i The second automatic water quality monitoring data obtained at time t. As an example, automatic water quality monitoring data can include turbidity, conductivity, pH value, dissolved oxygen, chlorophyll concentration, flow rate, flow velocity, etc.

[0029] Step S120: Based on the automatic water quality monitoring data and the location parameters of the corresponding automatic monitoring points, detect abnormal data in the obtained automatic water quality monitoring data;

[0030] Since automatic monitoring points are usually fixed, each point generally has fixed location parameters. However, these points may occasionally be adjusted as needed. Therefore, to ensure the accuracy of initial detections and to avoid increasing data transmission volume, at the beginning of each monitoring cycle T, the automatic monitoring points send their location parameters to the surface water environment monitoring platform so that the platform is aware of the location parameters of each point. For example: Automatic monitoring point D... i The position parameter is (X i Y i If so, then the automatic monitoring point D i Acquired water quality automatic monitoring data A it1 A it2 The position parameters of ... are all (X) i Y i ), where X i For automatic monitoring point D i longitude, Y i For automatic monitoring point D i Latitude.

[0031] like Figure 2 As shown, step S120 includes the following sub-steps:

[0032] Step S121: Based on the location parameters of the current automatic monitoring point and its upstream automatic monitoring point, obtain the relative distance between the upstream automatic monitoring point and the current automatic monitoring point.

[0033] Because the distance between upstream automatic monitoring stations and the current automatic monitoring station has a significant impact on changes in water quality automatic monitoring data, the surface water environment monitoring platform in this application calculates the relative distance between each upstream automatic monitoring station and the current automatic monitoring station based on the location parameters of multiple upstream automatic monitoring stations and the location parameters of the current automatic monitoring station. This distance is then used to calculate the influence weight of the upstream water quality automatic monitoring data on the predicted value of the current water quality automatic monitoring data. Here, the upstream water quality automatic monitoring data refers to the data monitored by the upstream automatic monitoring stations, and the current water quality automatic monitoring data refers to the data monitored by the current automatic monitoring stations.

[0034] Furthermore, the expression for the relative distance between the upstream automatic monitoring point and the current automatic monitoring point is:

[0035]

[0036] Among them, X i-n For automatic monitoring point D i-n longitude, Y i-n For automatic monitoring point D i-n Latitude, automatic monitoring point D i-n Located at automatic monitoring point D i Upstream; L i-n / i For automatic monitoring point D i-n With automatic monitoring point D i The relative distance.

[0037] Step S122: Based on the upstream water quality automatic monitoring data and the relative distance between the upstream automatic monitoring point and the current automatic monitoring point, obtain the influence weight of the upstream water quality automatic monitoring data on the predicted value of the current water quality automatic monitoring data.

[0038] Because there is a close relationship between upstream water quality monitoring data and current water quality monitoring data, and current water quality monitoring data is greatly affected by upstream water quality (e.g., upstream water temperature, turbidity, pH value, algae, suspended solids, etc.) can affect downstream water temperature, turbidity, pH value, algae, suspended solids, etc., the surface water environment monitoring platform in this application uses the water quality automatic monitoring data monitored by upstream automatic monitoring points to conduct preliminary detection on the water quality automatic monitoring data monitored by current automatic monitoring points, in order to preliminarily detect whether the current water quality automatic monitoring data is abnormal.

[0039] To conduct preliminary testing of current automatic water quality monitoring data, the surface water environment monitoring platform needs to first predict the current automatic water quality monitoring data based on upstream automatic water quality monitoring data. In order to predict the current automatic water quality monitoring data, the surface water environment monitoring platform needs to know the influence weight of each upstream automatic water quality monitoring data participating in the prediction on the predicted value of the current automatic water quality monitoring data.

[0040] Specifically, the surface water environment monitoring platform obtains the basic influence weight of each upstream water quality automatic monitoring data on the predicted value of the current water quality automatic monitoring data based on the flow of all tributaries within a predetermined range (e.g., 1000 meters upstream) of the current automatic monitoring point, the relative distance between the upstream automatic monitoring point and the current automatic monitoring point, and the upstream level parameter of the upstream automatic monitoring point (e.g., the upstream level of the first upstream automatic monitoring point is greater than that of the second upstream automatic monitoring point, which can be determined based on experience or trained based on historical data).

[0041] Furthermore, the expression for the basic influence weight of upstream water quality automatic monitoring data on the current water quality automatic monitoring data prediction value is as follows:

[0042]

[0043] Among them, JQ i-n Located at automatic monitoring point D i The nth automatic monitoring point D upstream i-n The corresponding automatic water quality monitoring data for automatic monitoring point D i The corresponding basic influence weight of the predicted value of automatic water quality monitoring data; S w Located at automatic monitoring point D i The flow rate of the w-th tributary upstream is a type of data from automatic water quality monitoring; W represents the flow rate at monitoring point D. i The number of upstream tributaries; S i-n For automatic monitoring point D i-n The flow rate of the tributary; α n For automatic monitoring point D i-n Upstream level parameters.

[0044] After obtaining the basic influence weights of each upstream water quality automatic monitoring data participating in the prediction on the current water quality automatic monitoring data prediction value, the surface water environment monitoring platform also needs to multiply each basic influence weight by a comprehensive adjustment coefficient to obtain the influence weight. Furthermore, the value of the comprehensive adjustment coefficient needs to ensure that the sum of all influence weights is 1, thereby guaranteeing that the prediction value of the current water quality automatic monitoring data meets the normal water quality change pattern.

[0045] Furthermore, the expression for the influence of the weights is:

[0046] Q i-n =β N JQ i-n

[0047] Among them, Q i-n Located at automatic monitoring point D i The nth automatic monitoring point D upstream i-n The corresponding automatic water quality monitoring data for automatic monitoring point D i The influence weight of the corresponding water quality automatic monitoring data prediction value; β N β is the comprehensive adjustment factor for all upstream automatic water quality monitoring data involved in the prediction. N The value of should be made N represents the total number of all upstream automatic monitoring points involved in the prediction.

[0048] Step S123: Based on the upstream water quality automatic monitoring data and its influence weight, obtain the predicted value of the current water quality automatic monitoring data;

[0049] After obtaining the influence weight of upstream water quality automatic monitoring data on the current water quality automatic monitoring data prediction value, the surface water environment monitoring platform also needs to multiply the upstream water quality automatic monitoring data with its corresponding influence weight, and then add the products to obtain the current water quality automatic monitoring data prediction value.

[0050] Furthermore, the expression for the predicted value of the current automatic water quality monitoring data is as follows:

[0051]

[0052] in, For automatic monitoring point D i The predicted value of the j-th automatic water quality monitoring data obtained at time t; A (i-n)tj For automatic monitoring point D i-n The j-th automatic water quality monitoring data obtained at time t.

[0053] Step S124: Compare the predicted value of the current automatic water quality monitoring data with the current automatic water quality monitoring data to detect whether the current automatic water quality monitoring data is abnormal.

[0054] After obtaining the automatic monitoring point D i The predicted value of the j-th automatic water quality monitoring data obtained at time t. Afterwards, the surface water environment monitoring platform will also need to automatically monitor point D. i The predicted value of the j-th automatic water quality monitoring data obtained at time t. With automatic monitoring point Di The j-th automatic water quality monitoring data A obtained at time t itj The comparison is performed, and if the difference is not greater than the abnormal threshold, that is: If δ is the abnormal threshold, then the automatic monitoring point D is considered to be abnormal. i If the j-th automatic water quality monitoring data obtained at time t is considered normal, then the automatic monitoring point D is considered to be normal. i The j-th automatic water quality monitoring data obtained at time t is abnormal data caused by monitoring errors.

[0055] Since this application identifies anomalies in downstream water quality automatic monitoring data using upstream water quality automatic monitoring data, the anomaly identification relies on the changing trends of surface water environment monitoring data and the correlation between upstream and downstream water environment monitoring data. This allows for a more accurate distinction between real anomalies caused by environmental changes and abnormal data caused by monitoring errors, providing an accurate basis for subsequent data correction.

[0056] Step S130: Based on the water quality manual monitoring data obtained at the start of the monitoring cycle and the location parameters of the corresponding manual monitoring location, predict the predicted value of the water quality manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle.

[0057] At the start of the monitoring period T, monitoring personnel will manually monitor multiple locations within the surface water system to obtain manual water quality data, which will then be sent to the surface water environment monitoring platform. For example, at manual monitoring location d... m Manual monitoring was conducted at point B to obtain manual monitoring data. m1 B m2 ..., among which, B m1 To manually monitor location d m The first manually obtained water quality monitoring data, B m2 To manually monitor location d m The second manual water quality monitoring data obtained; at manual monitoring location d m+1 Manual monitoring was conducted at point B to obtain manual monitoring data. (m+1)1 B (m+1)2 ..., among which, B (m+1)1 To manually monitor location d m+1 The first manually obtained water quality monitoring data, B (m+1)2 To manually monitor location d m+1 The second set of manually monitored water quality data was obtained. As an example, the type of manually monitored water quality data is the same as that of automatically monitored water quality data, and can also be turbidity, conductivity, pH value, dissolved oxygen, chlorophyll concentration, flow rate, flow velocity, etc.

[0058] Since manual water quality monitoring data is obtained by operators going to manual monitoring locations in surface water systems and using manual monitoring tools, and since the manual monitoring tools used by the operators are not used for long-term monitoring, their monitoring accuracy is relatively high. Therefore, this application can use manual water quality monitoring data to correct abnormal automatic water quality monitoring data.

[0059] Furthermore, since the water quality data obtained from manual monitoring at different locations vary significantly, the manual monitoring tools also need to obtain the location parameters for each manual monitoring location and send these parameters to the surface water environment monitoring platform so that the platform is aware of the location parameters for each manual monitoring location. For example: manual monitoring location d m The position parameter is (X m Y m ), then at the manually monitored location d m B data obtained from manual water quality monitoring m1 B m2 The position parameters of ... are all (X) m Y m ), where X m For manual monitoring of position d m longitude, Y m Location d of the manually monitored point m Latitude.

[0060] like Figure 3 As shown, step S130 includes the following sub-steps:

[0061] Step S131: Based on the location parameters of the current automatic monitoring point and the location parameters of its upstream / downstream manual monitoring points, obtain the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points;

[0062] The surface water environment monitoring platform uses the current automatic monitoring point D i Position parameters (X) i Y i ) and located at the current automatic monitoring point D i Upstream manual monitoring location d m Position parameters (X) m Y m ) and located at the current automatic monitoring point D i Downstream manual monitoring location d m+1 Position parameters (X) m+1 Y m+1 The current automatic monitoring point D is calculated. i Its upstream manual monitoring location d m The distance between them l imand the current automatic monitoring point D i Its downstream manual monitoring location d m+1 The distance between them l i(m+1) .

[0063] Furthermore, the expression for the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points is:

[0064]

[0065] Among them, X m+1 For manual monitoring of position d m+1 longitude, Y m+1 Location d of the manually monitored point m+1 Latitude.

[0066] Step S132: Based on the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points, calculate the influence weight of the upstream / downstream water quality manual monitoring data on the predicted value of the water quality manual monitoring data at the current automatic monitoring point.

[0067] After obtaining the current automatic monitoring point D i Its upstream manual monitoring location d m The distance between them l im and the current automatic monitoring point D i Its downstream manual monitoring location d m+1 The distance between them l i(m+1) Afterwards, the surface water environment monitoring platform will send the current automatic monitoring point D i Its upstream manual monitoring location d m The distance between them l im With the current automatic monitoring point D i Its upstream manual monitoring location d m The distance between them l im and the current automatic monitoring point D i Its downstream manual monitoring location d m+1 The distance between them l i(m+1) The ratio of the sums is used as the weight σ for the influence of upstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring data at manual water quality monitoring points. im And it will set the current automatic monitoring point D i Its downstream manual monitoring location d m+1 The distance between them l i(m+1) With the current automatic monitoring point D i Its upstream manual monitoring location d m The distance between them l im and the current automatic monitoring point D iIts downstream manual monitoring location d m+1 The distance between them l i(m+1) The ratio of the sums is used as the weight σ for the influence of downstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring data at manual water quality monitoring points. i(m+1) .

[0068] Furthermore, the expression for the weighting of the influence of upstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring points' manual water quality monitoring data is as follows:

[0069]

[0070] Step S133: Based on the influence weight of upstream / downstream water quality manual monitoring data on the predicted value of water quality manual monitoring data at the current automatic monitoring point, and the upstream / downstream water quality manual monitoring data, obtain the predicted value of water quality manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle.

[0071] The influence weight σ of upstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring data at manual water quality monitoring points is obtained. im The influence weight σ of downstream manual water quality monitoring data on the predicted values ​​of manual water quality monitoring data at current automatic monitoring points. i(m+1) Subsequently, the surface water environment monitoring platform will also consider the influence weight σ of the upstream manual water quality monitoring data on the predicted values ​​of the current automatic monitoring points' manual water quality monitoring data. im The influence weight σ of downstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring points for manual water quality monitoring data. i(m+1) The water quality manual monitoring data from upstream and downstream are used to obtain the predicted value of the water quality manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle.

[0072] Furthermore, the expression for the predicted value of the water quality manually monitored at the current automatic monitoring point at the start of the monitoring cycle is:

[0073] B ij =ω1σ im B mj +ω2σ i(m+1) B (m+1)j

[0074] Among them, B ij The current automatic monitoring point D at the start of the monitoring cycle. i The predicted value of the j-th manual water quality monitoring data; ω1 is the upstream monitoring weight, ω2 is the downstream monitoring weight; B mj For manual monitoring location d upstream m The j-th manually monitored water quality data point, B(m+1)j For manual monitoring location d downstream m+1 The j-th manually monitored water quality data.

[0075] Step S140: In response to obtaining abnormal data in the automatic water quality monitoring data, obtain the meteorological monitoring data at the current moment within the monitoring range, and obtain the meteorological change amplitude based on the meteorological monitoring data at the current moment and the meteorological monitoring data at the beginning of the monitoring cycle.

[0076] Because weather conditions within the monitoring area significantly impact surface water quality, correcting abnormal automatic water quality monitoring data requires considering these conditions. This ensures the corrected data more accurately reflects reality. Therefore, upon detecting anomalies in the automatic water quality monitoring data, the surface water environment monitoring platform utilizes various meteorological applications (e.g., MoWeather, WeatherTong) or interfaces with various meteorological platforms (e.g., the National Meteorological Science Data Center) to obtain the current meteorological monitoring data. For example, obtaining meteorological monitoring data P at time t. t1 P t2 ..., where P t1 To obtain the first meteorological monitoring data at time t, P t2 To obtain the second meteorological monitoring data for t. As an example, meteorological monitoring data could be temperature, wind speed, cumulative rainfall (TPS), etc.

[0077] Because adjustments based on meteorological monitoring data require knowledge of the meteorological monitoring data at time t1, the surface water environment monitoring platform acquires the meteorological monitoring data at time t1 before the start of monitoring period T. For example, it acquires the meteorological monitoring data at time t1. ……,in, To obtain the first meteorological monitoring data at time t1, To obtain the second meteorological monitoring data for t1.

[0078] The surface water environment monitoring platform will use the current meteorological monitoring data P t1 P t2 Meteorological monitoring data at the start of the monitoring period, ... ...and thus the amplitude of meteorological changes is obtained.

[0079] Furthermore, the expression for the amplitude of meteorological changes is:

[0080]

[0081] in, P represents the magnitude of the meteorological change at time t1 relative to time t;tj To obtain the j-th meteorological monitoring data at time t; To obtain the j-th meteorological monitoring data of t1; denoted as the weight value of the j-th meteorological monitoring data; J represents the number of meteorological monitoring data obtained.

[0082] Step S150: Adjust the predicted value of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle based on the meteorological change amplitude, and correct the abnormal data in the automatic water quality monitoring data according to the adjusted value;

[0083] Specifically, such as Figure 4 As shown, adjusting the predicted value of the water quality manually monitored at the current automatic monitoring point at the start of the monitoring cycle includes the following sub-steps:

[0084] Step S151: Determine whether the amplitude of meteorological changes exceeds the threshold;

[0085] The magnitude of the meteorological change at time t1 relative to time t is obtained. The surface water environment monitoring platform will determine the amplitude of meteorological changes. Whether it exceeds the threshold FY (e.g., ≤10%).

[0086] Step S152: If the value is not exceeded, apply a fine-tuning coefficient to adjust the meteorological change amplitude.

[0087] If the threshold FY is not exceeded, the surface water environment monitoring platform will fine-tune the predicted value B of the manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle. ij Get the adjustment value Right now: TX1 is a fine-tuning coefficient, TX1≈1. For example, TX1 can be any constant between 0.98 and 1.02, or the manually predicted water quality data B at the current automatic monitoring point at the start of the monitoring cycle. ij That is, the fine-tuning coefficient TX1 = 1.

[0088] Step S153: If the value exceeds the limit, apply an adjustment coefficient to adjust the meteorological change amplitude.

[0089] If the threshold FY is exceeded, the surface water environment monitoring platform will adjust the monitoring according to the meteorological change amplitude. The value of the adjustment coefficient TX is selected in segments. r TX r Meteorological variation amplitude The value of is the adjustment factor for the r-th segment, and the adjustment factor TX is applied. r Adjust the predicted value B of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle. ij Get the adjustment value Right now:

[0090] The adjusted value is obtained by adjusting the predicted values ​​of the manual water quality monitoring data at the current automatic monitoring points at the start of the monitoring cycle. Then, apply the adjustment value. Automatic monitoring point D, which is considered as abnormal data i The j-th automatic water quality monitoring data A obtained at time t itj Make corrections.

[0091] Furthermore, the expression for applying the adjustment value to correct outlier data is:

[0092]

[0093] Where γ is the correction weight for the adjusted value, τ is the correction weight for outlier data, and γ = μτ, where μ is a positive integer greater than 1.

[0094] This application corrects abnormal automatic water quality monitoring data using manual water quality monitoring data and meteorological monitoring data, thereby making the correction of abnormal data more accurate and more in line with the actual situation. This provides an accurate basis for predicting water quality changes in the future and provides forward-looking support for environmental decision-making.

[0095] Example 2

[0096] like Figure 5 As shown, this application provides an adaptive correction prediction system 500 for surface water environment monitoring data anomalies, including: automatic monitoring points 510, manual monitoring tools 520, meteorological monitoring platform 530, and surface water environment monitoring platform 540; wherein, the surface water environment monitoring platform 540 includes: anomaly data detection unit 541, prediction value prediction unit 542, change amplitude calculation unit 543, and anomaly data correction unit 544.

[0097] During the monitoring period, automatic monitoring points 510 acquire automatic water quality monitoring data of the surface water environment.

[0098] Automatic monitoring points 510 (e.g., monitoring instruments) are deployed at different cross-sections of a surface water system to form an automatic monitoring network for that surface water system. Within a monitoring period T, all automatic monitoring points 510 in this network will automatically monitor according to a predetermined frequency, continuously acquiring water quality monitoring data. The automatic monitoring points 510 will then transmit the acquired water quality monitoring data to the surface water environment monitoring platform 540. For example, automatic monitoring point D... i Water quality automatic monitoring data A is obtained at time t. it1 Ait2 ..., where A it1 For automatic monitoring point D i The first automatic water quality monitoring data obtained at time t, A it2 For automatic monitoring point D i The second automatic water quality monitoring data obtained at time t. As an example, automatic water quality monitoring data can include turbidity, conductivity, pH value, dissolved oxygen, chlorophyll concentration, flow rate, flow velocity, etc.

[0099] The abnormal data detection unit 541 detects abnormal data in the obtained automatic water quality monitoring data based on the automatic water quality monitoring data and the location parameters of the corresponding automatic monitoring points.

[0100] Since the automatic monitoring points 510 are usually fixed, each automatic monitoring point 510 generally has fixed location parameters. However, the automatic monitoring points 510 may be adjusted occasionally as needed. Therefore, to ensure the accuracy of the initial detection and to avoid increasing the amount of data transmission, at the beginning of each monitoring cycle T, the automatic monitoring points 510 will send their location parameters to the surface water environment monitoring platform 540 so that the surface water environment monitoring platform 540 knows the location parameters of each automatic monitoring point 510. For example: Automatic monitoring point D i The position parameter is (X i Y i If so, then the automatic monitoring point D i Acquired water quality automatic monitoring data A it1 A it2 The position parameters of ... are all (X) i Y i ), where X i For automatic monitoring point D i longitude, Y i For automatic monitoring point D i Latitude.

[0101] The abnormal data detection unit 541 obtains the relative distance between the upstream automatic monitoring point and the current automatic monitoring point based on the location parameters of the current automatic monitoring point and its upstream automatic monitoring point.

[0102] Because the distance between upstream automatic monitoring points and the current automatic monitoring point has a significant impact on changes in water quality automatic monitoring data, the anomaly detection unit 541 of the surface water environment monitoring platform 540 in this application calculates the relative distance between each upstream automatic monitoring point and the current automatic monitoring point based on the location parameters of multiple upstream automatic monitoring points and the location parameters of the current automatic monitoring point. This distance is then used to subsequently calculate the influence weight of the upstream water quality automatic monitoring data on the predicted value of the current water quality automatic monitoring data. Here, the upstream water quality automatic monitoring data refers to the data monitored by the upstream automatic monitoring points, and the current water quality automatic monitoring data refers to the data monitored by the current automatic monitoring points.

[0103] Furthermore, the expression for the relative distance between the upstream automatic monitoring point and the current automatic monitoring point is:

[0104]

[0105] Among them, X i-n For automatic monitoring point D i-n longitude, Y i-n For automatic monitoring point D i-n Latitude, automatic monitoring point D i-n Located at automatic monitoring point D i Upstream; L i-n / i For automatic monitoring point D i-n With automatic monitoring point D i The relative distance.

[0106] The abnormal data detection unit 541 obtains the influence weight of the upstream water quality automatic monitoring data on the predicted value of the current water quality automatic monitoring data based on the upstream water quality automatic monitoring data and the relative distance between the upstream automatic monitoring point and the current automatic monitoring point.

[0107] Because there is a close connection between upstream water quality monitoring data and current water quality monitoring data, and current water quality monitoring data is largely affected by upstream water quality (e.g., upstream water temperature, turbidity, pH value, algae, suspended solids, etc.), downstream water temperature, turbidity, pH value, algae, suspended solids, etc., the abnormal data detection unit 541 of the surface water environment monitoring platform 540 in this application uses the automatic water quality monitoring data monitored by the upstream automatic monitoring points to detect the current automatic water quality monitoring data monitored by the automatic monitoring points, so as to detect whether the current automatic water quality monitoring data is abnormal.

[0108] To perform preliminary testing on the current automatic water quality monitoring data, the abnormal data detection unit 541 of the surface water environment monitoring platform 540 needs to predict the current automatic water quality monitoring data based on the upstream automatic water quality monitoring data. In order to predict the current automatic water quality monitoring data, the abnormal data detection unit 541 of the surface water environment monitoring platform 540 needs to know the influence weight of each upstream automatic water quality monitoring data participating in the prediction on the predicted value of the current automatic water quality monitoring data.

[0109] Specifically, the abnormal data detection unit 541 of the surface water environment monitoring platform 540 obtains the basic influence weight of each upstream water quality automatic monitoring data on the predicted value of the current water quality automatic monitoring data based on the flow rate of all tributaries within a predetermined range (e.g., 1000 meters upstream) of the current automatic monitoring point, the relative distance between the upstream automatic monitoring point and the current automatic monitoring point, and the upstream level parameter of the upstream automatic monitoring point (e.g., the upstream level of the first upstream automatic monitoring point is greater than the upstream level of the second upstream automatic monitoring point, which can be determined based on experience or trained based on historical data).

[0110] Furthermore, the expression for the basic influence weight of upstream water quality automatic monitoring data on the current water quality automatic monitoring data prediction value is as follows:

[0111]

[0112] Among them, JQ i-n Located at automatic monitoring point D i The nth automatic monitoring point D upstream i-n The corresponding automatic water quality monitoring data for automatic monitoring point D i The corresponding basic influence weight of the predicted value of automatic water quality monitoring data; S w Located at automatic monitoring point D i The flow rate of the w-th tributary upstream is a type of data from automatic water quality monitoring; W represents the flow rate at monitoring point D. i The number of upstream tributaries; S i-n For automatic monitoring point D i-n The flow rate of the tributary; α n For automatic monitoring point D i-n Upstream level parameters.

[0113] After obtaining the basic influence weights of each upstream water quality automatic monitoring data participating in the prediction on the current water quality automatic monitoring data prediction value, the abnormal data detection unit 541 of the surface water environment monitoring platform 540 also needs to multiply each basic influence weight by a comprehensive adjustment coefficient to obtain the influence weight. Furthermore, the value of the comprehensive adjustment coefficient needs to ensure that the sum of all influence weights is 1, thereby ensuring that the prediction value of the current water quality automatic monitoring data meets the normal water quality change pattern.

[0114] Furthermore, the expression for the influence of the weights is:

[0115] Q i-n =β N JQ i-n

[0116] Among them, Q i-n Located at automatic monitoring point D i The nth automatic monitoring point D upstream i-n The corresponding automatic water quality monitoring data for automatic monitoring point D i The influence weight of the corresponding water quality automatic monitoring data prediction value; β N β is the comprehensive adjustment factor for all upstream automatic water quality monitoring data involved in the prediction. N The value of should be made N represents the total number of all upstream automatic monitoring points involved in the prediction.

[0117] The abnormal data detection unit 541 obtains the predicted value of the current automatic water quality monitoring data based on the upstream water quality automatic monitoring data and its influence weight.

[0118] After obtaining the influence weight of upstream water quality automatic monitoring data on the current water quality automatic monitoring data prediction value, the abnormal data detection unit 541 of the surface water environment monitoring platform 540 also multiplies the upstream water quality automatic monitoring data with its corresponding influence weight, and then adds the products to obtain the current water quality automatic monitoring data prediction value.

[0119] Furthermore, the expression for the predicted value of the current automatic water quality monitoring data is as follows:

[0120]

[0121] in, For automatic monitoring point D i The predicted value of the j-th automatic water quality monitoring data obtained at time t; A (i-n)tj For automatic monitoring point D i-n The j-th automatic water quality monitoring data obtained at time t.

[0122] The abnormal data detection unit 541 compares the predicted value of the current automatic water quality monitoring data with the current automatic water quality monitoring data to detect whether the current automatic water quality monitoring data is abnormal.

[0123] After obtaining the automatic monitoring point D i The predicted value of the j-th automatic water quality monitoring data obtained at time t. Afterwards, the abnormal data detection unit 541 of the surface water environment monitoring platform 540 will also need to detect the abnormal data at the automatic monitoring point D. i The predicted value of the j-th automatic water quality monitoring data obtained at time t. With automatic monitoring point D i The j-th automatic water quality monitoring data A obtained at time t itj The comparison is performed, and if the difference is not greater than the abnormal threshold, that is: If δ is the abnormal threshold, then the automatic monitoring point D is considered to be abnormal. i If the j-th automatic water quality monitoring data obtained at time t is considered normal, then the automatic monitoring point D is considered to be normal. i The j-th automatic water quality monitoring data obtained at time t is abnormal data caused by monitoring errors.

[0124] Since this application identifies anomalies in downstream water quality automatic monitoring data using upstream water quality automatic monitoring data, the anomaly identification relies on the changing trends of surface water environment monitoring data and the correlation between upstream and downstream water environment monitoring data. This allows for a more accurate distinction between real anomalies caused by environmental changes and abnormal data caused by monitoring errors, providing an accurate basis for subsequent data correction.

[0125] The prediction unit 542 predicts the predicted value of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle based on the manual water quality monitoring data obtained by the manual monitoring tool 520 at the start of the monitoring cycle and the location parameters of the corresponding manual monitoring location.

[0126] At the start of the monitoring period T, monitoring personnel will travel to multiple locations within the surface water system to conduct manual monitoring using manual monitoring tools 520, thereby obtaining manual water quality monitoring data. This data will then be sent to the surface water environment monitoring platform 540. For example, at manual monitoring location d... m Manual monitoring was conducted at point B to obtain manual monitoring data. m1 B m2 ..., among which, B m1 To manually monitor location d m The first manually obtained water quality monitoring data, B m2 To manually monitor location d mThe second manual water quality monitoring data obtained; at manual monitoring location d m+1 Manual monitoring was conducted at point B to obtain manual monitoring data. (m+1)1 B (m+1)2 ..., among which, B (m+1)1 To manually monitor location d m+1 The first manually obtained water quality monitoring data, B (m+1)2 To manually monitor location d m+1 The second set of manually monitored water quality data was obtained. As an example, the type of manually monitored water quality data is the same as that of automatically monitored water quality data, and can also be turbidity, conductivity, pH value, dissolved oxygen, chlorophyll concentration, flow rate, flow velocity, etc.

[0127] Since the water quality manual monitoring data is obtained by operators going to the manual monitoring location of the surface water system and operating the manual monitoring tool 520, and the manual monitoring tool used by the operators is not used for long-term monitoring, its monitoring accuracy is relatively high. Therefore, this application can use the water quality manual monitoring data to correct abnormal water quality automatic monitoring data.

[0128] Furthermore, since the water quality data obtained from manual monitoring at different locations vary significantly, the manual monitoring tool 520 also needs to obtain the location parameters for each manual monitoring location and send these parameters to the surface water environment monitoring platform 540 so that the platform is aware of the location parameters for each location. For example: manual monitoring location d... m The position parameter is (X m Y m (, then at the manually monitored location d) m B data obtained from manual water quality monitoring m1 B m2 The position parameters of ... are all (X) m Y m ), where X m For manual monitoring of position d m longitude, Y m Location d of the manually monitored point m Latitude.

[0129] The prediction unit 542 obtains the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points based on the location parameters of the current automatic monitoring point and the location parameters of its upstream / downstream manual monitoring points.

[0130] The prediction unit 542 of the surface water environment monitoring platform 540 predicts the value based on the current automatic monitoring point D. i Position parameters (X) i Yi ) and located at the current automatic monitoring point D i Upstream manual monitoring location d m Position parameters (X) m Y m ) and located at the current automatic monitoring point D i Downstream manual monitoring location d m+1 Position parameters (X) m+1 Y m+1 The current automatic monitoring point D is calculated. i Its upstream manual monitoring location d m The distance between them l im and the current automatic monitoring point D i Its downstream manual monitoring location d m+1 The distance between them l i(m+1) .

[0131] Furthermore, the expression for the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points is:

[0132]

[0133] Among them, X m+1 For manual monitoring of position d m+1 longitude, Y m+1 Location d of the manually monitored point m+1 Latitude.

[0134] The prediction unit 542 calculates the influence weight of the upstream / downstream water quality manual monitoring data on the predicted value of the water quality manual monitoring data at the current automatic monitoring point based on the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points.

[0135] After obtaining the current automatic monitoring point D i Its upstream manual monitoring location d m The distance between them l im and the current automatic monitoring point D i Its downstream manual monitoring location d m+1 The distance between them l i(m+1) Afterwards, the prediction unit 542 of the surface water environment monitoring platform 540 will predict the current automatic monitoring point D. i Its upstream manual monitoring location d m The distance between them l im With the current automatic monitoring point D i Its upstream manual monitoring location d m The distance between them l im and the current automatic monitoring point D iIts downstream manual monitoring location d m+1 The distance between them l i(m+1) The ratio of the sums is used as the weight σ for the influence of upstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring data at manual water quality monitoring points. im And it will set the current automatic monitoring point D i Its downstream manual monitoring location d m+1 The distance between them l i(m+1) With the current automatic monitoring point D i Its upstream manual monitoring location d m The distance between them l im and the current automatic monitoring point D i Its downstream manual monitoring location d m+1 The distance between them l i(m+1) The ratio of the sums is used as the weight σ for the influence of downstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring data at manual water quality monitoring points. i(m+1) .

[0136] Furthermore, the expression for the weighting of the influence of upstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring points' manual water quality monitoring data is as follows:

[0137]

[0138] The prediction unit 542 obtains the predicted value of the manual water quality data at the current automatic monitoring point at the start of the monitoring cycle based on the influence weight of the manual water quality monitoring data of the upstream / downstream water quality on the predicted value of the manual water quality data at the current automatic monitoring point, as well as the manual water quality monitoring data of the upstream / downstream.

[0139] The influence weight σ of upstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring data at manual water quality monitoring points is obtained. im The influence weight σ of downstream manual water quality monitoring data on the predicted values ​​of manual water quality monitoring data at current automatic monitoring points. i(m+1) Subsequently, the prediction unit 542 of the surface water environment monitoring platform 540 will also consider the influence weight σ of the upstream manual water quality monitoring data on the predicted value of the current automatic monitoring point's manual water quality data. im The influence weight σ of downstream manual water quality monitoring data on the predicted values ​​of current automatic monitoring points for manual water quality monitoring data. i(m+1) The water quality manual monitoring data from upstream and downstream are used to obtain the predicted value of the water quality manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle.

[0140] Furthermore, the expression for the predicted value of the water quality manually monitored at the current automatic monitoring point at the start of the monitoring cycle is:

[0141] B ij =ω1σ im B mj +ω2σ i(m+1) B (m+1)j

[0142] Among them, B ij The current automatic monitoring point D at the start of the monitoring cycle. i The predicted value of the j-th manual water quality monitoring data; ω1 is the upstream monitoring weight, ω2 is the downstream monitoring weight; B mj For manual monitoring location d upstream m The j-th manually monitored water quality data point, B (m+1)j For manual monitoring location d downstream m+1 The j-th manually monitored water quality data.

[0143] In response to the abnormal data obtained in the automatic water quality monitoring data, the meteorological monitoring platform 530 acquires the meteorological monitoring data of the current moment within the monitoring range, and the change amplitude calculation unit 543 obtains the meteorological change amplitude based on the meteorological monitoring data of the current moment and the meteorological monitoring data at the beginning of the monitoring cycle.

[0144] Because weather conditions within the monitoring area significantly impact the water quality of surface water systems, it's crucial to consider weather conditions within the monitoring area when correcting abnormal automatic water quality monitoring data. This ensures the corrected data more accurately reflects reality. Therefore, upon detecting anomalies in the automatic water quality monitoring data, the surface water environment monitoring platform 540 uses the meteorological monitoring platform 530 (various meteorological applications (e.g., MoWeather, WeatherTong) or interfaces with various meteorological platforms (e.g., the National Meteorological Science Data Center)) to obtain the current meteorological monitoring data. For example, obtaining meteorological monitoring data P at time t. t1 P t2 ..., where P t1 To obtain the first meteorological monitoring data at time t, P t2 To obtain the second meteorological monitoring data for t. As an example, meteorological monitoring data could be temperature, wind speed, cumulative rainfall (TPS), etc.

[0145] Since adjustments based on meteorological monitoring data require knowledge of the meteorological monitoring data at the start time t1 of the monitoring period T, the surface water environment monitoring platform 540 uses the meteorological monitoring platform 530 to pre-acquire the meteorological monitoring data at the start time t1 of the monitoring period T. For example, acquiring the meteorological monitoring data at time t1. ……,in, To obtain the first meteorological monitoring data at time t1, To obtain the second meteorological monitoring data for t1.

[0146] The surface water environment monitoring platform 540's change amplitude calculation unit 543 will calculate the change amplitude based on the current meteorological monitoring data P. t1 P t2 Meteorological monitoring data at the start of the monitoring period, ... ...and thus the amplitude of meteorological changes is obtained.

[0147] Furthermore, the expression for the amplitude of meteorological changes is:

[0148]

[0149] in, P represents the magnitude of the meteorological change at time t1 relative to time t; tj To obtain the j-th meteorological monitoring data at time t; To obtain the j-th meteorological monitoring data of t1; denoted as the weight value of the j-th meteorological monitoring data; J represents the number of meteorological monitoring data obtained.

[0150] The abnormal data correction unit 544 adjusts the predicted value of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle based on the amplitude of meteorological changes, and corrects the abnormal data in the automatic water quality monitoring data according to the adjusted value.

[0151] Specifically, the abnormal data correction unit 544 determines whether the amplitude of meteorological changes exceeds the threshold.

[0152] The magnitude of the meteorological change at time t1 relative to time t is obtained. The surface water environment monitoring platform will determine the amplitude of meteorological changes. Whether it exceeds the threshold FY (e.g., ≤10%).

[0153] If the value is not exceeded, the abnormal data correction unit 544 applies a fine-tuning coefficient to adjust the meteorological change amplitude.

[0154] If the threshold FY is not exceeded, the surface water environment monitoring platform will fine-tune the predicted value B of the manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle. ij Get the adjustment value Right now: TX1 is a fine-tuning coefficient, TX1≈1. For example, TX1 can be any constant between 0.98 and 1.02, or the manually predicted water quality data B at the current automatic monitoring point at the start of the monitoring cycle. ij That is, the fine-tuning coefficient TX1 = 1.

[0155] If the value exceeds the limit, the abnormal data correction unit 544 applies an adjustment coefficient to adjust the meteorological change amplitude.

[0156] If the threshold FY is exceeded, the surface water environment monitoring platform will adjust the monitoring according to the meteorological change amplitude. The value of the adjustment coefficient TX is selected in segments. r TX r Meteorological variation amplitude The value of is the adjustment factor for the r-th segment, and the adjustment factor TX is applied. r Adjust the predicted value B of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle. ij Get the adjustment value Right now:

[0157] The adjusted value is obtained by adjusting the predicted values ​​of the manual water quality monitoring data at the current automatic monitoring points at the start of the monitoring cycle. Then, the abnormal data correction unit 544 applies the adjustment value. Automatic monitoring point D, which is considered as abnormal data i The j-th automatic water quality monitoring data A obtained at time t itj Make corrections.

[0158] Furthermore, the expression for applying the adjustment value to correct outlier data is:

[0159]

[0160] Where γ is the correction weight for the adjusted value, τ is the correction weight for outlier data, and γ = μτ, where μ is a positive integer greater than 1.

[0161] This application corrects abnormal automatic water quality monitoring data using manual water quality monitoring data and meteorological monitoring data, thereby making the correction of abnormal data more accurate and more in line with the actual situation. This provides an accurate basis for predicting water quality changes in the future and provides forward-looking support for environmental decision-making.

[0162] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0163] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An adaptive correction and prediction method for anomalies in surface water environmental monitoring data, characterized in that, Includes the following steps: Step S110: During the monitoring period, automatic monitoring points acquire automatic water quality monitoring data of the surface water environment; Step S120: Based on the automatic water quality monitoring data and the location parameters of the corresponding automatic monitoring points, detect abnormal data in the obtained automatic water quality monitoring data; Step S130: Based on the water quality manual monitoring data obtained at the start of the monitoring cycle and the location parameters of the corresponding manual monitoring location, predict the predicted value of the water quality manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle. Step S140: In response to obtaining abnormal data in the automatic water quality monitoring data, obtain the meteorological monitoring data at the current moment within the monitoring range, and obtain the meteorological change amplitude based on the meteorological monitoring data at the current moment and the meteorological monitoring data at the beginning of the monitoring cycle. The expression for the amplitude of meteorological changes is: ； in, for Time relative to The magnitude of meteorological changes at any given time; To obtain The first moment Meteorological monitoring data; To obtain The Meteorological monitoring data; For the first Weight values ​​for each meteorological monitoring data point; The quantity of meteorological monitoring data obtained; Step S150: Adjust the predicted value of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle based on the meteorological change amplitude, and correct the abnormal data in the automatic water quality monitoring data according to the adjusted value; Step S150 includes the following sub-steps: Step S151: Determine whether the amplitude of meteorological changes exceeds the threshold; Step S152: If the time limit is not exceeded, fine-tune the predicted value of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle. Get the adjustment value ,Right now: , For fine-tuning coefficients, ; Step S153: If it exceeds, then it is based on the meteorological change amplitude. The value of the adjustment coefficient is selected in segments. , Meteorological variation amplitude The value of belongs to the th The adjustment factor for the segment, the application of the adjustment factor Adjust the predicted value of manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle. Get the adjustment value ,Right now: ; The adjusted value is obtained by adjusting the predicted values ​​of the manual water quality monitoring data at the current automatic monitoring points at the start of the monitoring cycle. Then, apply the adjustment value. Automatic monitoring points for abnormal data exist The first time obtained Water quality automatic monitoring data Make corrections; The expression for applying adjustment values ​​to correct outlier data is: ； in, Adjustment weights for the adjusted values The corrected weights for outlier data, and , It is a positive integer greater than 1.

2. The adaptive correction and prediction method for anomalies in surface water environmental monitoring data according to claim 1, characterized in that, Step S120 includes the following sub-steps: Step S121: Based on the location parameters of the current automatic monitoring point and its upstream automatic monitoring point, obtain the relative distance between the upstream automatic monitoring point and the current automatic monitoring point. Step S122: Based on the upstream water quality automatic monitoring data and the relative distance between the upstream automatic monitoring point and the current automatic monitoring point, obtain the influence weight of the upstream water quality automatic monitoring data on the predicted value of the current water quality automatic monitoring data. Step S123: Based on the upstream water quality automatic monitoring data and its influence weight, obtain the predicted value of the current water quality automatic monitoring data; Step S124: Compare the predicted value of the current automatic water quality monitoring data with the current automatic water quality monitoring data to detect whether the current automatic water quality monitoring data is abnormal.

3. The adaptive correction and prediction method for anomalies in surface water environmental monitoring data according to claim 1 or 2, characterized in that, Step S130 includes the following sub-steps: Step S131: Based on the location parameters of the current automatic monitoring point and the location parameters of its upstream / downstream manual monitoring points, obtain the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points; Step S132: Based on the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points, calculate the influence weight of the upstream / downstream water quality manual monitoring data on the predicted value of the water quality manual monitoring data at the current automatic monitoring point. Step S133: Based on the influence weight of upstream / downstream water quality manual monitoring data on the predicted value of water quality manual monitoring data at the current automatic monitoring point, and the upstream / downstream water quality manual monitoring data, obtain the predicted value of water quality manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle.

4. An adaptive correction and prediction system for anomalies in surface water environmental monitoring data, characterized in that, include: Automatic monitoring points, manual monitoring tools, meteorological monitoring platform, and surface water environment monitoring platform; among which, the surface water environment monitoring platform includes: anomaly data detection unit, predicted value prediction unit, change amplitude calculation unit, and anomaly data correction unit; During the monitoring period, automatic monitoring points acquire automatic water quality monitoring data of the surface water environment; The abnormal data detection unit detects abnormal data in the obtained automatic water quality monitoring data based on the automatic water quality monitoring data and the location parameters of the corresponding automatic monitoring points. The prediction unit predicts the water quality manual monitoring data at the current automatic monitoring point at the start of the monitoring cycle based on the water quality manual monitoring data obtained by the manual monitoring tool at the start of the monitoring cycle and the location parameters of the corresponding manual monitoring location. In response to the abnormal data obtained in the automatic water quality monitoring data, the meteorological monitoring platform obtains the meteorological monitoring data of the current moment within the monitoring range, and the change amplitude calculation unit obtains the meteorological change amplitude based on the meteorological monitoring data of the current moment and the meteorological monitoring data at the beginning of the monitoring cycle; The expression for the amplitude of meteorological changes is: ； in, for Time relative to The magnitude of meteorological changes at any given time; To obtain The first moment Meteorological monitoring data; To obtain The Meteorological monitoring data; For the first Weight values ​​for each meteorological monitoring data point; The quantity of meteorological monitoring data obtained; The abnormal data correction unit adjusts the predicted value of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle based on the amplitude of meteorological changes, and corrects the abnormal data in the automatic water quality monitoring data according to the adjusted value; The abnormal data correction unit determines whether the magnitude of meteorological changes exceeds a threshold. If the time limit is not exceeded, the abnormal data correction unit will fine-tune the predicted value of the manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle. Get the adjustment value ,Right now: , For fine-tuning coefficients, ; If it exceeds the limit, the abnormal data correction unit will adjust it according to the meteorological change amplitude. The value of the adjustment coefficient is selected in segments. , Meteorological variation amplitude The value of belongs to the th The adjustment factor for the segment, the application of the adjustment factor Adjust the predicted value of manual water quality monitoring data at the current automatic monitoring point at the start of the monitoring cycle. Get the adjustment value ,Right now: ; The adjusted value is obtained by adjusting the predicted values ​​of the manual water quality monitoring data at the current automatic monitoring points at the start of the monitoring cycle. Then, the abnormal data correction unit applies the adjustment value. Automatic monitoring points for abnormal data exist The first time obtained Water quality automatic monitoring data Make corrections; The expression for applying adjustment values ​​to correct outlier data is: ； in, Adjustment weights for the adjusted values The corrected weights for outlier data, and , It is a positive integer greater than 1.

5. The adaptive correction and prediction system for anomalies in surface water environmental monitoring data according to claim 4, characterized in that, The abnormal data detection unit obtains the relative distance between the upstream automatic monitoring point and the current automatic monitoring point based on the location parameters of the current automatic monitoring point and its upstream automatic monitoring point. The abnormal data detection unit obtains the influence weight of the upstream water quality automatic monitoring data on the predicted value of the current water quality automatic monitoring data based on the upstream water quality automatic monitoring data and the relative distance between the upstream automatic monitoring point and the current automatic monitoring point. The abnormal data detection unit obtains the predicted value of the current automatic water quality monitoring data based on the upstream water quality automatic monitoring data and its influence weight; The abnormal data detection unit compares the predicted value of the current automatic water quality monitoring data with the current automatic water quality monitoring data to detect whether the current automatic water quality monitoring data is abnormal.

6. The adaptive correction and prediction system for anomalies in surface water environmental monitoring data according to claim 4 or 5, characterized in that, The prediction unit obtains the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points based on the location parameters of the current automatic monitoring point and the location parameters of its upstream / downstream manual monitoring points. The prediction unit calculates the influence weight of the upstream / downstream water quality manual monitoring data on the predicted value of the water quality manual monitoring data at the current automatic monitoring point based on the distance between the current automatic monitoring point and its upstream / downstream manual monitoring points. The prediction unit obtains the predicted value of the manual water quality data at the current automatic monitoring point at the start of the monitoring cycle based on the influence weight of the upstream / downstream manual water quality monitoring data on the predicted value of the manual water quality data at the current automatic monitoring point, as well as the upstream / downstream manual water quality monitoring data.

Images