Water pollutant flux prediction method and system, electronic equipment and product

Abstract

The invention belongs to the technical field of water environment monitoring and simulation analysis, and aims to provide a water pollutant flux prediction method and system, electronic equipment and a product. According to the invention, the final water pollutant flux prediction model is obtained based on historical data, and the real-time water monitoring data obtained by water quality monitoring is used as the input data of the final water pollutant flux prediction model, so that the real-time water pollutant flux prediction value is obtained. And finally, performing correction processing on the real-time water body pollutant flux prediction value to obtain a corrected real-time water body pollutant flux prediction value of the target water body area, thereby realizing pollutant flux prediction of the target water body area. According to the method, the flux of the water pollutants can be predicted, and the prediction precision is high.

Description

Technical Field

[0001] The present invention belongs to the technical field of water environment monitoring and simulation analysis, and specifically relates to a method, system, electronic equipment and product for predicting the flux of water pollutants. Background Art

[0002] Water pollution is one of the major environmental problems facing my country. Among them, nitrogen and phosphorus are the main pollutants in non-point source pollution in farmland irrigation areas in the south and sewage discharged from residents' production and life. Therefore, scientifically and rationally predicting the flux of pollutants in water bodies, especially nitrogen and phosphorus fluxes, is of great significance for understanding the pollution status of river sections, characterizing the transport characteristics of regional nitrogen and phosphorus pollutants, and accurately analyzing the carrying capacity of river water environment. In this process, high-frequency monitoring of water quantity and quality data is usually required as support.

[0003] At present, flow measurement equipment such as pressure water level meters, radar flow meters and Doppler flow meters are widely used, the river flow monitoring system is relatively complete, and flow data is easy to obtain and continuous. However, due to the technical level of current water quality monitoring methods and the limitations of manpower and material resources, the current routine water quality monitoring frequency is usually once a month, which makes the water quantity and quality monitoring data have the problems of discrete distribution of water quality data, large time span and low overlap with water quantity data. Based on this, it can only reflect the overall change trend of river pollutant concentration, and it is difficult to quantitatively reflect the daily change process of pollutants. Therefore, how to truly simulate the change process of water pollutant flux is the core and difficulty in the existing technology.

[0004] In the prior art, global averaging and linear interpolation methods are usually used to predict water pollutant fluxes. Among them, the global averaging method assumes that the water quality monitoring data once a month represents the overall average water quality status of the month, while the linear interpolation method assumes that the water quality status between single measured data points obeys the law of linear change. Although the global averaging method and the linear interpolation method are simple and easy to use, they are only suitable for rivers with little impact from human activities and stable water volume and quality. As the intensity of water resource development continues to increase, the changes in river water volume and quality are becoming more and more obvious. The use conditions of the above two methods deviate seriously from the actual situation, and the predicted water body nitrogen and phosphorus pollutant fluxes have large errors compared with the actual situation. Summary of the invention

[0005] The present invention aims to solve the above technical problems at least to a certain extent. The present invention provides a method, system, electronic equipment and product for predicting the flux of water pollutants.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for predicting water pollutant flux, comprising:

[0008] Acquire a water body monitoring data set of a target water body area in a specified monitoring period, and preprocess the water body monitoring data set to obtain a preprocessed water body monitoring data set;

[0009] Obtaining the measured value of the water pollutant flux in the target water area during the specified monitoring period based on the pre-processed water monitoring data set;

[0010] Constructing an initial water pollutant flux prediction model, and performing parameter estimation processing on the initial water pollutant flux prediction model based on the pre-processed water monitoring data set and the measured value of the water pollutant flux, and performing parameter verification processing on the model parameters of the initial water pollutant flux prediction model, so as to obtain a final water pollutant flux prediction model that passes the parameter verification;

[0011] Acquire real-time water body monitoring data of the target water body area, and use the real-time water body monitoring data as input data of the final water body pollutant flux prediction model, and obtain the real-time water body pollutant flux prediction value of the target water body area based on the final water body pollutant flux prediction model;

[0012] The real-time water pollutant flux prediction value is corrected to obtain a corrected real-time water pollutant flux prediction value for the target water body area.

[0013] In a possible design, the water body monitoring data set is preprocessed to obtain a preprocessed water body monitoring data set, including:

[0014] Obtaining abnormal values ​​in the water body monitoring data set;

[0015] The abnormal values ​​in the water body monitoring data set are respectively subjected to mean value correction processing, elimination processing, and correction processing as missing values ​​to obtain three initial pre-processed water body monitoring data sets that match the water body monitoring data set;

[0016] The fitting effect parameters of the three initial pre-processed water body monitoring data sets are obtained respectively, and the initial pre-processed water body monitoring data set with the best fitting effect parameters among the three initial pre-processed water body monitoring data sets is used as the final pre-processed water body monitoring data set.

[0017] In one possible design, a box plot method is used to obtain outliers in the water body monitoring dataset.

[0018] In a possible design, the water body monitoring data set includes a flow monitoring data set and a water body pollutant concentration monitoring data set of a specified type; correspondingly, the measured value of the water body pollutant flux in the target water body area during the specified monitoring period is:

[0019]

[0020] In the formula, c i is the water pollutant concentration monitoring data at the i-th discrete moment in the water pollutant concentration monitoring data set of the specified type; Q i is the average flow rate corresponding to the i-th discrete moment in the flow monitoring data set; N is the total number of discrete moments in the specified monitoring period; Δt is the time period length of adjacent discrete moments in the specified monitoring period.

[0021] In one possible design, when performing parameter estimation processing on the initial water pollutant flux prediction model based on the pre-processed water monitoring data set and the actual measured value of the water pollutant flux, the asymptotic maximum likelihood estimation method, the maximum likelihood estimation method or the minimum absolute deviation method is adopted.

[0022] In a possible design, when the model parameters of the initial water pollutant flux prediction model are subjected to parameter testing, a determination coefficient test method, a Nash efficiency coefficient test method, a residual sequence correlation coefficient test method and a T ratio test method are used.

[0023] In a possible design, the Kalman filter algorithm is used to correct the real-time water pollutant flux prediction value.

[0024] In a second aspect, the present invention provides a water pollutant flux prediction system, which is used to implement the water pollutant flux prediction method as described in any one of the above; the water pollutant flux prediction system comprises:

[0025] A model acquisition module, used to acquire a water body monitoring data set of a target water body area in a specified monitoring period, and preprocess the water body monitoring data set to obtain a preprocessed water body monitoring data set; used to obtain the measured value of the water body pollutant flux of the target water body area in the specified monitoring period based on the preprocessed water body monitoring data set; and also used to construct an initial water body pollutant flux prediction model, and perform parameter estimation processing on the initial water body pollutant flux prediction model based on the preprocessed water body monitoring data set and the measured value of the water body pollutant flux, and perform parameter verification processing on the model parameters of the initial water body pollutant flux prediction model, so as to obtain a final water body pollutant flux prediction model that passes the parameter verification;

[0026] A flux prediction module, which is in communication with the model acquisition module, is used to obtain real-time water body monitoring data of the target water body area, and use the real-time water body monitoring data as input data of the final water body pollutant flux prediction model, and obtain the real-time water body pollutant flux prediction value of the target water body area based on the final water body pollutant flux prediction model;

[0027] The result correction module is in communication with the flux prediction module and is used to perform correction processing on the real-time water pollutant flux prediction value to obtain the corrected real-time water pollutant flux prediction value of the target water area.

[0028] In a third aspect, the present invention provides an electronic device, comprising:

[0029] a memory for storing computer program instructions; and,

[0030] A processor is used to execute the computer program instructions to complete the operation of the water pollutant flux prediction method as described in any one of the above.

[0031] In a fourth aspect, the present invention provides a computer program product, comprising a computer program or instructions, which, when executed by a computer, implements a method for predicting the flux of pollutants in a water body as described in any one of the above.

[0032] The beneficial effects of the present invention are:

[0033] The present invention discloses a method, system, electronic device and product for predicting water pollutant flux, which can realize the prediction of water pollutant flux with high prediction accuracy. Specifically, in the implementation process of the present invention, first, a water body monitoring data set of a target water body area in a specified monitoring period is obtained, and the water body monitoring data set is preprocessed to obtain a preprocessed water body monitoring data set, so as to obtain the measured value of the water body pollutant flux of the target water body area in the specified monitoring period based on the preprocessed water body monitoring data set; then, an initial water body pollutant flux prediction model is constructed, and parameter estimation processing is performed on the initial water body pollutant flux prediction model based on the preprocessed water body monitoring data set and the measured value of the water body pollutant flux, and the initial water body The model parameters of the pollutant flux prediction model are subjected to parameter verification processing to obtain a final water pollutant flux prediction model that passes the parameter verification; then, the real-time water body monitoring data of the target water body area is obtained, and the real-time water body monitoring data is used as input data of the final water body pollutant flux prediction model, and the real-time water body pollutant flux prediction value of the target water body area is obtained based on the final water body pollutant flux prediction model; finally, the real-time water body pollutant flux prediction value is corrected to obtain the corrected real-time water body pollutant flux prediction value of the target water body area. In this process, the present invention obtains a final water pollutant flux prediction model based on historical data, and uses the real-time water monitoring data obtained from water quality monitoring as input data of the final water pollutant flux prediction model, thereby obtaining a real-time water pollutant flux prediction value, and finally corrects the real-time water pollutant flux prediction value to obtain the corrected real-time water pollutant flux prediction value of the target water body area, thereby realizing the pollutant flux prediction of the target water body area, which can not only characterize the fluctuation and change characteristics of the water quality process based on the real-time water monitoring data, but also realize the rapid prediction of the water pollutant flux of the target water body, the prediction result has high accuracy, and the application difficulty is moderate, which is easy to promote.

[0034] Other beneficial effects of the present invention will be further described in the specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of the method for predicting water pollutant flux in Example 1;

[0036] Figure 2 is a schematic diagram of the correlation between the estimated fluxes of ammonia nitrogen and total phosphorus and the corrected fluxes;

[0037] Figure 3 It is a module block diagram of the water pollutant flux prediction system in Example 2;

[0038] Figure 4 This is a module block diagram of an electronic device in Example 3. DETAILED DESCRIPTION

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structure of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0040] Embodiment 1:

[0041] This embodiment discloses a method for predicting water pollutant flux, which can be executed by, but is not limited to, a computer device or a virtual machine with certain computing resources, such as a personal computer, a smart phone, a personal digital assistant, or a wearable device, or by a virtual machine.

[0042] like Figure 1 As shown, a method for predicting water pollutant flux may include but is not limited to the following steps:

[0043] S1. Obtain a water body monitoring data set of the target water body area in a specified monitoring period, and preprocess the water body monitoring data set to obtain a preprocessed water body monitoring data set; it should be noted that due to the human errors caused by sampling, transportation and storage factors in the process of monitoring water pollutants such as nitrogen and phosphorus in rivers, there may be abnormal values ​​in the water body monitoring data set, which may easily affect the accuracy of subsequent water pollutant flux prediction results. Based on this, after obtaining the water body monitoring data set of the target water body area in the specified monitoring period, this embodiment preprocesses the water body monitoring data set, thereby ensuring the accuracy of subsequent water pollutant flux prediction results.

[0044] In this embodiment, the box plot method is used to preprocess the water body monitoring data set. Specifically, in step S1, the water body monitoring data set is preprocessed to obtain a preprocessed water body monitoring data set, including:

[0045] S101. Obtaining outliers in the water body monitoring data set;

[0046] In step S101, the box plot method is used to obtain the outliers in the water body monitoring data set. It should be noted that the box plot is a statistical chart that shows the dispersion of data information and is often used for data outlier processing. The box plot method can be used to determine the bias and dispersion of the water body monitoring data set, and can also intuitively identify the outliers in the water body monitoring data set, that is, outlier data. In the implementation process, the upper quartile Q3, lower quartile Q1 and median X in the water body monitoring data set are obtained in advance. m , and obtain the difference IQR between the upper quartile Q3 and the lower quartile Q1, and regard the data outside the 1.5IQR range in the water body monitoring data set as outliers.

[0047] S102. performing mean value correction processing, elimination processing, and correction processing as missing values ​​on the abnormal values ​​in the water body monitoring data set, respectively, to obtain three initial pre-processed water body monitoring data sets that match the water body monitoring data set;

[0048] S103. Obtain the fitting effect parameters of the three initial pre-processed water body monitoring data sets respectively, and use the initial pre-processed water body monitoring data set with the best fitting effect parameters among the three initial pre-processed water body monitoring data sets as the final pre-processed water body monitoring data set. Specifically, in this embodiment, the determination coefficient R 2 As a fitting effect parameter, it is possible to judge the fitting effect of the water body monitoring data set after initial preprocessing.

[0049] In step S1, the water body monitoring data set includes a flow monitoring data set and a water body pollutant concentration monitoring data set of a specified type; correspondingly, in step S2, the measured value of the water body pollutant flux in the target water body area during the specified monitoring period is:

[0050]

[0051] In the formula, c i is the water pollutant concentration monitoring data at the i-th discrete moment in the water pollutant concentration monitoring data set of the specified type; Q i is the average flow rate corresponding to the i-th discrete moment in the flow monitoring data set; N is the total number of discrete moments in the specified monitoring period, corresponding to the number of monitoring moments of the water pollutant concentration monitoring data; Δt is the time period length of adjacent discrete moments in the specified monitoring period.

[0052] It should be noted that, in this embodiment, the specified types include nitrogen and / or phosphorus. During implementation, under the same section conditions, the monitoring frequency of nitrogen and phosphorus concentration data is usually much lower than the monitoring frequency of flow rate. That is, the amount of data in the water pollutant concentration monitoring data set of the specified type is much lower than the amount of data in the flow monitoring data set. Based on this, this embodiment adopts a multivariate linear regression method to realize the calculation of the measured values ​​of nitrogen and phosphorus pollutant fluxes within the specified monitoring period.

[0053] S2. Based on the pre-processed water body monitoring data set, the actual measured value of the water pollutant flux in the target water body area during the specified monitoring period is obtained.

[0054] S3. Construct an initial water pollutant flux prediction model, and perform parameter estimation processing on the initial water pollutant flux prediction model based on the pre-treated water monitoring data set and the actual measured value of the water pollutant flux, and perform parameter verification processing on the model parameters of the initial water pollutant flux prediction model, so as to obtain a final water pollutant flux prediction model that passes the parameter verification; It should be noted that, in this embodiment, the water pollutant flux prediction model is a multivariate statistical model of the correlation regression relationship between water pollutant flux and water flow established based on discrete data in the pre-treated water monitoring data set, which can be used to estimate the flux or pollution load of water pollutants such as nitrogen and phosphorus in rivers at different time scales and spatial scales, and can simulate the continuous changes in the load of water pollutants such as nitrogen and phosphorus in the target water area.

[0055] In step S3, when the parameter estimation processing of the initial water pollutant flux prediction model is performed based on the pre-treated water monitoring data set and the measured value of the water pollutant flux, the asymptotic maximum likelihood estimation method, the maximum likelihood estimation method or the least absolute deviation method is adopted. Specifically, in this embodiment, when the pre-treated water monitoring data set includes censored data, the asymptotic maximum likelihood estimation method is adopted; when the data in the pre-treated water monitoring data set is non-censored data, the maximum likelihood estimation method is adopted; when the residual of the initial water pollutant flux prediction model (that is, the difference between the predicted value obtained by inputting the pre-treated water monitoring data set into the initial water pollutant flux prediction model and the measured value of the water pollutant flux) does not meet the normal distribution assumption, the least absolute deviation method is adopted.

[0056] In step S3, when the model parameters of the initial water pollutant flux prediction model are subjected to parameter verification, the determination coefficient (R 2 ) test method, Nash efficiency coefficient test method, residual sequence correlation coefficient test method and T ratio test method.

[0057] Specifically, the determination coefficient (R 2) test method for parameter testing, using the determination coefficient R 2 The fitting of the regression equation in the initial water pollutant flux prediction model was tested, and the regression determination coefficient R 2 When it is >80%, it indicates that the initial water pollutant flux prediction model has a good fitting degree, R 2 >90% indicates that the initial water pollutant flux prediction model has a very good fit and the parameter test has passed.

[0058] When the Nash-Sutcliffeefficiency (NSE) coefficient test method is used for parameter test processing, the NSE coefficient represents the relationship between the calculated value and the average measured value of the initial water pollutant flux prediction model, and its value range is -∞ to 1, and the larger the NSE value, the higher the consistency between the simulation value and the measured value. Specifically, the NSE value calculation formula is as follows:

[0059]

[0060] In the formula, Q mea,i represents the water pollutant concentration monitoring data at the i-th discrete moment in the water pollutant concentration monitoring data set, Q sim,i Indicates that the water pollutant concentration monitoring data Q mea,i The simulated value obtained after inputting the initial water pollutant flux prediction model, Represents the average value of the water pollutant concentration monitoring data in the water pollutant concentration monitoring data set.

[0061] When the residual sequence correlation coefficient test method is used for parameter test processing, the residual sequence correlation coefficient is used to test whether the residuals of the initial water pollutant flux prediction model have serial correlation. The smaller the value, the more independent the residuals are. When the data in the pre-treated water body monitoring data set is non-censored data, the probability curve correlation coefficient is used to test whether the residuals of the regression equation of the optimized initial water body pollutant flux prediction model conform to the normal distribution. If the value is greater than 0.9, it indicates that the residuals obey the normal distribution. When the data in the pre-treated water body monitoring data set includes censored data, the Turnbull-Weiss test method is used to test whether the residuals of the optimized nitrogen and phosphorus flux regression equation conform to the normal distribution. If the value is less than 0.05, it indicates that the residuals obey the normal distribution.

[0062] When the T ratio test method is used for parameter inspection, since multicollinearity affects the results of regression analysis, the initial water pollutant flux prediction model uses the correlation coefficient to determine whether there is a correlation between the independent variables. In the case of multicollinear correlation, it can be eliminated by concentrating the independent variables to obtain the central value of its water pollutant concentration monitoring data set, and when the central value is less than the preset central threshold, it is determined that the model parameter test of the initial water pollutant flux prediction model has passed.

[0063] It should be noted that, in the present embodiment, a plurality of initial water pollutant flux prediction models may be pre-constructed, and parameter estimation and parameter verification processing may be performed on the plurality of initial water pollutant flux prediction models respectively. If a final water pollutant flux prediction model that passes multiple parameter tests is obtained, the initial water pollutant flux prediction models that pass the multiple parameter tests may be respectively subjected to the Akaike information criterion and the SPPC criterion based on the maximum entropy principle to judge the quality of the model fitting to the measured data, and finally the initial water pollutant flux prediction model with the best quality may be selected as the final water pollutant flux prediction model.

[0064] S4. Obtain real-time water body monitoring data of the target water body area, and use the real-time water body monitoring data as input data of the final water body pollutant flux prediction model, and obtain the real-time water body pollutant flux prediction value of the target water body area based on the final water body pollutant flux prediction model; it should be understood that, in this embodiment, after obtaining the real-time water body monitoring data of the target water body area, the real-time water body monitoring data is pre-processed in advance, and then the pre-processed data is input into the final water body pollutant flux prediction model.

[0065] S5. The real-time water pollutant flux prediction value is corrected to obtain the corrected real-time water pollutant flux prediction value of the target water area. It should be noted that due to the incomplete data in the water monitoring data set and the non-optimal estimated model parameters, there is still an error between the real-time water pollutant flux prediction value (such as the real-time nitrogen and phosphorus flux prediction value) estimated by the final water pollutant flux prediction model and the actual value. For this reason, after obtaining the real-time water pollutant flux prediction value based on the final water pollutant flux prediction model, this embodiment further corrects the real-time water pollutant flux prediction value, which can effectively improve the estimation accuracy.

[0066] In step S5, the Kalman filter algorithm is used to calibrate the real-time water pollutant flux prediction value.

[0067] It should be noted that the Kalman filter algorithm has the advantages of small storage volume and short calculation time, and has been widely used in real-time correction of statistical data in recent years.

[0068] Specifically, in this embodiment, when the real-time water pollutant flux prediction value is corrected, a Kalman filter state equation is first established according to the Kalman filter theory, as shown below:

[0069] X0=AX L +ω;

[0070]

[0071] Where X0 is the initial value of the Kalman filter correction iteration; X L represents the real-time water pollutant flux prediction value estimated by the final water pollutant flux prediction model; A is a preset state transition parameter, which can be the linear correlation coefficient between the measured value of the water pollutant flux and the estimated value obtained by the final water pollutant flux prediction model based on the pre-processed water monitoring data set; w is the preset model noise; represents the predicted value of water pollutant flux in the i-th iteration; X i-1 is the predicted value of water pollutant flux in the i-1th iteration.

[0072] The Kalman gain calculation formula is as follows:

[0073]

[0074]

[0075] In the formula, K i is the Kalman gain of the i-th iteration; H is the preset measurement conversion parameter matrix; B is the preset measurement noise variance, which can be given according to the nitrogen and phosphorus test analysis manual; is the state estimation error covariance of the i-th iteration; P i-1 is the correction covariance of the i-1th step iteration, and its initial value of the iteration can be given by referring to the water quality estimation error variance of the final water pollutant flux prediction model corresponding to nitrogen and phosphorus pollutants; D is the noise variance.

[0076] The calculation formula for Kalman filter correction and state filter error covariance update is as follows:

[0077]

[0078] In the formula, X i is the predicted value of the water pollutant flux of the i-th iteration, that is, the real-time predicted value of the water pollutant flux after correction; Y is the measured flux value calculated based on the real-time water monitoring data; I is the unit matrix.

[0079] It should be noted that the Kalman filter algorithm is used to perform autoregressive iterative calculation on the real-time water pollutant flux prediction value, and the real-time water pollutant flux prediction value after real-time correction can be obtained after the state filter error covariance converges.

[0080] The present embodiment is described in detail below through specific examples.

[0081] First, the water body monitoring data set uses the nitrogen and phosphorus water quality abnormal data of the Guilin Station of the Lijiang River in Guangxi from 2010 to 2019, and uses the box plot method to judge the effective range of the water body monitoring data set, and the abnormal values ​​of total phosphorus and ammonia nitrogen concentrations are 0 and 4 respectively, of which the abnormal values ​​with large data account for 70%. The abnormal values ​​are respectively corrected by (1) mean value; (2) shaved; (3) as missing values, and regression simulation is performed with the unprocessed data to obtain the correlation coefficient R of ammonia nitrogen and total phosphorus at the Guilin Station (see Table 1). By comparison, it is found that the fitting effects of ammonia nitrogen and total phosphorus are improved by 2.37% and 0, respectively.

[0082] Table 1 Effect of nitrogen and phosphorus data preprocessing

[0083]

[0084]

[0085] Note: (1) mean value corrected; (2) shaved; (3) censored.

[0086] Then, using the daily average flow, monitoring time and nitrogen and phosphorus concentration data of the Lijiang River Guilin Station, the regression equations of ammonia nitrogen flux and total phosphorus flux were obtained. The parameters were calibrated by the Akaike information criterion of the maximum entropy principle and the SPPC criterion. The results showed that the optimal regression model for ammonia nitrogen and total phosphorus fluxes was a seven-parameter equation. Among them, the regression equation for ammonia nitrogen flux is:

[0087]

[0088] The total phosphorus flux regression equation is:

[0089]

[0090] Table 2 Regression model analysis of ammonia nitrogen and total phosphorus

[0091]

[0092] As shown in Table 2 above, the pollutant flux regression equations for ammonia nitrogen and total phosphorus were well fitted, with R 2 The values ​​are 0.70 and 0.86 respectively, and the larger R 2The values ​​indicate that pollutant flux, daily discharge and time have a good correlation. When comparing the simulated pollutant flux with the measured pollutant flux, the NSE values ​​are 0.83 (ammonia nitrogen) and 0.88 (total phosphorus), indicating that the simulation results of nitrogen and phosphorus in the Lijiang River are similar to the monitoring results, with small errors, and the results of the pollutant flux regression model are reliable. The P values ​​of ammonia nitrogen and total phosphorus are lower than 0.05, indicating that the equation coefficients are statistically significant. In addition, the PPCC values ​​are 0.97 (ammonia nitrogen) and 0.99 (total phosphorus), respectively, indicating that the residuals meet the requirements of normal distribution. Based on the above results, the regression equation optimized based on the pollutant flux regression model can be used to predict the nitrogen and phosphorus pollution fluxes in the Lijiang River.

[0093] Subsequently, the Kalman filter algorithm was used to correct the predicted nitrogen and phosphorus pollution fluxes based on the pollutant flux regression model. In order to verify the rationality of the pollutant flux estimation results based on the improved prediction and correction method, in this embodiment, the corrected fluxes of ammonia nitrogen and total phosphorus were compared with the measured fluxes and the estimated fluxes based on the regression model. The results showed that the corrected nitrogen and phosphorus daily fluxes were basically consistent with the measured pollutant fluxes. Linear regression analysis was performed on 120 ammonia nitrogen and total phosphorus estimated values ​​and positive values. Figure 2 As shown, the horizontal axis is the estimated NH3-H / TP flux, and the vertical axis is the corrected NH3-H / TP flux. The results show that the correlation coefficients between the estimated flux and the corrected flux of ammonia nitrogen and total phosphorus are 0.80 and 0.90, respectively. The cumulative fluxes of ammonia nitrogen for a total of 120 measured values ​​and the corresponding estimated values ​​are 283.75t and 292.83t, respectively. After real-time correction, the relative deviation between the corrected value of ammonia nitrogen flux and the measured value dropped from 7.18% to 3.20%. The cumulative fluxes of the measured value and the corresponding estimated value of total phosphorus are 63.52t and 64.40t, respectively. After Kalman filter correction, the relative deviation between the corrected value of total phosphorus flux and the measured value dropped from 9.76% to 1.39%.

[0094] The above examples show that the ammonia nitrogen and total phosphorus flux results of the Lijiang River from 2010 to 2019 obtained by using the water pollutant flux prediction method in this embodiment are reasonable and reliable.

[0095] This embodiment can realize the prediction of water pollutant flux with high prediction accuracy. Specifically, in the implementation process of this embodiment, first, a water body monitoring data set of the target water body area in a specified monitoring period is obtained, and the water body monitoring data set is preprocessed to obtain a preprocessed water body monitoring data set, so as to obtain the measured value of the water body pollutant flux of the target water body area in the specified monitoring period based on the preprocessed water body monitoring data set; then, an initial water body pollutant flux prediction model is constructed, and parameter estimation processing is performed on the initial water body pollutant flux prediction model based on the preprocessed water body monitoring data set and the measured value of the water body pollutant flux, and the initial water body pollutant flux prediction model is used to estimate the parameters of the initial water body pollutant flux prediction model. The model parameters of the water pollutant flux prediction model are subjected to parameter verification processing so as to obtain a final water pollutant flux prediction model that passes the parameter verification; then, the real-time water body monitoring data of the target water body area is acquired, and the real-time water body monitoring data is used as input data of the final water body pollutant flux prediction model, and the real-time water body pollutant flux prediction value of the target water body area is obtained based on the final water body pollutant flux prediction model; finally, the real-time water body pollutant flux prediction value is corrected to obtain the corrected real-time water body pollutant flux prediction value of the target water body area. In this process, this embodiment obtains the final water pollutant flux prediction model based on historical data, and uses the real-time water monitoring data obtained from water quality monitoring as the input data of the final water pollutant flux prediction model, thereby obtaining the real-time water pollutant flux prediction value, and finally correcting the real-time water pollutant flux prediction value to obtain the corrected real-time water pollutant flux prediction value of the target water body area, thereby realizing the pollutant flux prediction of the target water body area, which can not only characterize the fluctuation and change characteristics of the water quality process based on the real-time water monitoring data, but also realize the rapid prediction of the water pollutant flux of the target water body, the prediction result has high accuracy, and the application difficulty is moderate, which is easy to promote.

[0096] Embodiment 2:

[0097] This embodiment discloses a water pollutant flux prediction system, which is used to implement the water pollutant flux prediction method in embodiment 1; Figure 3 As shown, the water pollutant flux prediction system includes:

[0098] A model acquisition module, used to acquire a water body monitoring data set of a target water body area in a specified monitoring period, and preprocess the water body monitoring data set to obtain a preprocessed water body monitoring data set; used to obtain the measured value of the water body pollutant flux of the target water body area in the specified monitoring period based on the preprocessed water body monitoring data set; and also used to construct an initial water body pollutant flux prediction model, and perform parameter estimation processing on the initial water body pollutant flux prediction model based on the preprocessed water body monitoring data set and the measured value of the water body pollutant flux, and perform parameter verification processing on the model parameters of the initial water body pollutant flux prediction model, so as to obtain a final water body pollutant flux prediction model that passes the parameter verification;

[0099] A flux prediction module, which is in communication with the model acquisition module, is used to obtain real-time water body monitoring data of the target water body area, and use the real-time water body monitoring data as input data of the final water body pollutant flux prediction model, and obtain the real-time water body pollutant flux prediction value of the target water body area based on the final water body pollutant flux prediction model;

[0100] The result correction module is in communication with the flux prediction module and is used to perform correction processing on the real-time water pollutant flux prediction value to obtain the corrected real-time water pollutant flux prediction value of the target water area.

[0101] It should be noted that the working process, working details and technical effects of the water pollutant flux prediction system provided in this embodiment 2 can be found in embodiment 1 and will not be repeated here.

[0102] Embodiment 3:

[0103] Based on Embodiment 1 or 2, this embodiment discloses an electronic device, which may be a smart phone, a tablet computer, a laptop computer, or a desktop computer. The electronic device may be referred to as a user terminal, a portable terminal, a desktop terminal, etc. Figure 4 As shown, the electronic equipment includes:

[0104] a memory for storing computer program instructions; and,

[0105] A processor is used to execute the computer program instructions to complete the operation of the water pollutant flux prediction method as described in any one of Example 1.

[0106] Specifically, the processor 301 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 301 may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 301 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 301 may be integrated with a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the display screen.

[0107] The memory 302 may include one or more computer-readable storage media, which may be non-transitory. The memory 302 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 302 is used to store at least one instruction, which is used to be executed by the processor 301 to implement the water pollutant flux prediction method provided in Example 1 of the present application.

[0108] In some embodiments, the terminal may further optionally include: a communication interface 303 and at least one peripheral device. The processor 301, the memory 302 and the communication interface 303 may be connected via a bus or a signal line. Each peripheral device may be connected to the communication interface 303 via a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 304, a display screen 305 and a power supply 306.

[0109] The communication interface 303 may be used to connect at least one peripheral device related to I / O (Input / Output) to the processor 301 and the memory 302. In some embodiments, the processor 301, the memory 302, and the communication interface 303 are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor 301, the memory 302, and the communication interface 303 may be implemented on a separate chip or circuit board, which is not limited in this embodiment.

[0110] The radio frequency circuit 304 is used to receive and transmit RF (Radio Frequency) signals, also known as electromagnetic signals. The radio frequency circuit 304 communicates with a communication network and other communication devices through electromagnetic signals.

[0111] The display screen 305 is used to display a UI (User Interface). The UI may include graphics, text, icons, videos, and any combination thereof.

[0112] The power supply 306 is used to supply power to various components in the electronic device.

[0113] Embodiment 4:

[0114] Based on any one of Embodiments 1 to 3, this embodiment discloses a computer program product, including a computer program or an instruction, wherein when the computer program or the instruction is executed by a computer, the water pollutant flux prediction method as described in any one of Embodiment 1 is implemented. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0115] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the above embodiments, a person skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting water pollutant flux, characterized in that: include: Acquire a water body monitoring data set of a target water body area in a specified monitoring period, and preprocess the water body monitoring data set to obtain a preprocessed water body monitoring data set; Obtaining the measured value of the water pollutant flux in the target water area during the specified monitoring period based on the pre-processed water monitoring data set; Constructing an initial water pollutant flux prediction model, and performing parameter estimation processing on the initial water pollutant flux prediction model based on the pre-processed water monitoring data set and the measured value of the water pollutant flux, and performing parameter verification processing on the model parameters of the initial water pollutant flux prediction model, so as to obtain a final water pollutant flux prediction model that passes the parameter verification; Acquire real-time water body monitoring data of the target water body area, and use the real-time water body monitoring data as input data of the final water body pollutant flux prediction model, and obtain the real-time water body pollutant flux prediction value of the target water body area based on the final water body pollutant flux prediction model; The real-time water pollutant flux prediction value is corrected to obtain a corrected real-time water pollutant flux prediction value for the target water body area.

2. A method for predicting water pollutant flux according to claim 1, characterized in that: Preprocessing the water body monitoring data set to obtain a preprocessed water body monitoring data set includes: Obtaining abnormal values ​​in the water body monitoring data set; The abnormal values ​​in the water body monitoring data set are respectively subjected to mean value correction processing, elimination processing, and correction processing as missing values ​​to obtain three initial pre-processed water body monitoring data sets that match the water body monitoring data set; The fitting effect parameters of the three initial pre-processed water body monitoring data sets are obtained respectively, and the initial pre-processed water body monitoring data set with the best fitting effect parameters among the three initial pre-processed water body monitoring data sets is used as the final pre-processed water body monitoring data set.

3. A method for predicting water pollutant flux according to claim 2, characterized in that: The box plot method was used to obtain outliers in the water body monitoring dataset.

4. A method for predicting water pollutant flux according to claim 1, characterized in that: The water body monitoring data set includes a flow monitoring data set and a water body pollutant concentration monitoring data set of a specified type; correspondingly, the measured value of the water body pollutant flux in the target water body area during the specified monitoring period is: In the formula, c i is the water pollutant concentration monitoring data at the i-th discrete moment in the water pollutant concentration monitoring data set of the specified type; Q i is the average flow rate corresponding to the i-th discrete moment in the flow monitoring data set; N is the total number of discrete moments in the specified monitoring period; Δt is the time period length of adjacent discrete moments in the specified monitoring period.

5. A method for predicting water pollutant flux according to claim 1, characterized in that: When performing parameter estimation processing on the initial water pollutant flux prediction model based on the pre-processed water monitoring data set and the actual measured value of the water pollutant flux, the asymptotic maximum likelihood estimation method, the maximum likelihood estimation method or the minimum absolute deviation method is adopted.

6. A method for predicting water pollutant flux according to claim 1, characterized in that: When the model parameters of the initial water pollutant flux prediction model are subjected to parameter test processing, the determination coefficient test method, the Nash efficiency coefficient test method, the residual sequence correlation coefficient test method and the T ratio test method are adopted.

7. A method for predicting water pollutant flux according to claim 1, characterized in that: When the real-time water pollutant flux prediction value is corrected, the Kalman filter algorithm is used.

8. A water pollutant flux prediction system, characterized in that: Used to implement the water pollutant flux prediction method as described in any one of claims 1 to 7; The water pollutant flux prediction system comprises: A model acquisition module, used to acquire a water body monitoring data set of a target water body area in a specified monitoring period, and preprocess the water body monitoring data set to obtain a preprocessed water body monitoring data set; used to obtain the measured value of the water body pollutant flux of the target water body area in the specified monitoring period based on the preprocessed water body monitoring data set; and also used to construct an initial water body pollutant flux prediction model, and perform parameter estimation processing on the initial water body pollutant flux prediction model based on the preprocessed water body monitoring data set and the measured value of the water body pollutant flux, and perform parameter verification processing on the model parameters of the initial water body pollutant flux prediction model, so as to obtain a final water body pollutant flux prediction model that passes the parameter verification; A flux prediction module, which is in communication with the model acquisition module, is used to obtain real-time water body monitoring data of the target water body area, and use the real-time water body monitoring data as input data of the final water body pollutant flux prediction model, and obtain the real-time water body pollutant flux prediction value of the target water body area based on the final water body pollutant flux prediction model; The result correction module is in communication with the flux prediction module and is used to perform correction processing on the real-time water pollutant flux prediction value to obtain the corrected real-time water pollutant flux prediction value of the target water area.

9. An electronic device, characterized in that: include: a memory for storing computer program instructions; as well as, A processor is used to execute the computer program instructions to complete the operation of the water pollutant flux prediction method as described in any one of claims 1 to 7.

10. A computer program product comprising a computer program or instructions, characterized in that The computer program or the instructions, when executed by a computer, implements the method for predicting water pollutant flux as described in any one of claims 1 to 7.

Images