A rural water pollution monitoring method based on multi-source heterogeneous data

Abstract

The application discloses a rural water pollution monitoring method based on multi-source heterogeneous data, which reveals the coupling relationship between decomposition and metal release by comparing the concentration of sediment organic matter and iron-manganese ions, further screens the positive change fragments, and constructs a feature set by combining factors such as pH value, water temperature and flow rate, so as to realize quantitative characterization of pollution release intensity. The comparison between the intensity value and the dissolved oxygen gradient difference refines and analyzes the oxygen difference between water layers, and forms a coefficient in combination with the nitrogen and phosphorus concentration difference, so as to reflect the correlation between nutrient salt diffusion and oxygen change. Finally, through threshold determination and duration calculation, the dynamic quantification of longitudinal diffusion intensity and duration is realized, and the explainability of the pollution migration rule and the fineness of the monitoring result are improved.

Description

Technical Field

[0001] This invention belongs to the field of rural water pollution monitoring technology, and in particular relates to a rural water pollution monitoring method based on multi-source heterogeneous data. Background Technology

[0002] The field of environmental monitoring systems involves the real-time monitoring and early warning of natural environmental elements through sensors, data acquisition, and communication technologies. This technology is particularly important in water quality monitoring, where it is typically used to accurately analyze and warn of pollution processes. Rural water environments are complex, and pollution monitoring in these areas usually requires the integration of sensor data from multiple sources (i.e., multi-source heterogeneous data). However, existing technologies are insufficient because monitoring methods tend to focus on single-parameter acquisition, resulting in weak correlations between parameters and difficulty in identifying the progressive processes of pollutant release and diffusion. Most monitoring only presents data change trends, lacking a joint characterization of temporal segments and spatial differences, making it difficult to accurately analyze pollutant migration patterns. In rural water scenarios, conventional monitoring ignores the coupling effect of sediment decomposition, metal ion release, and nutrient diffusion, easily leading to delayed risk identification and an inability to quantify the persistence of diffusion, thus affecting the targeting of pollution early warnings and the timeliness of control. Summary of the Invention

[0003] To address the problems existing in the prior art, this invention proposes a method for monitoring rural water pollution based on multi-source heterogeneous data.

[0004] The technical solution of the present invention is as follows:

[0005] A method for monitoring rural water pollution based on multi-source heterogeneous data includes:

[0006] The organic matter content of the sediment surface and the iron and manganese ion concentration in the water were obtained. The degradation rate of organic matter at adjacent time points was calculated. The fluctuation range of degradation rate and iron and manganese ion concentration were compared to generate the synchronization rate of organic matter and metal fluctuation.

[0007] Based on the synchronization rate of organic matter and metal fluctuations, segments with positive changes in degradation rate and iron and manganese ion concentration are screened. The corresponding segments are recombined as feature sets, including organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature, and flow rate. The average change amplitude of the feature set over a continuous time period is calculated to obtain the coupling release intensity value.

[0008] The coupling release intensity value is used to collect dissolved oxygen concentrations in different water layers, calculate the dissolved oxygen gradient difference between adjacent water layers, and compare the gradient difference with the coupling release intensity value to obtain the longitudinal dissolved oxygen difference value.

[0009] Based on the longitudinal dissolved oxygen difference, nitrogen and phosphorus concentrations of the same water layer are collected, the nitrogen and phosphorus concentration difference is calculated, and the nitrogen and phosphorus concentration difference is calculated together with the longitudinal dissolved oxygen difference to generate a nitrogen-phosphorus-oxygen difference joint coefficient.

[0010] The nitrogen-phosphorus-oxygen difference joint coefficient is called, compared with the diffusion determination threshold, and the diffusion state is marked when the threshold is exceeded. The duration of this state in the monitoring interval is calculated to obtain the longitudinal diffusion duration.

[0011] Furthermore, the organic matter and metal fluctuation synchronization rate includes the degree of synchronization between the organic matter degradation rate and the iron and manganese ion concentration fluctuation amplitude; the coupling release intensity value includes the average change amplitude of degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature, and flow rate within a specific time period; the longitudinal dissolved oxygen difference value includes the dissolved oxygen concentration gradient difference between adjacent water layers; the nitrogen-phosphorus and oxygen difference joint coefficient includes the joint calculation result of nitrogen-phosphorus concentration difference and longitudinal dissolved oxygen gradient difference; and the longitudinal diffusion duration includes the duration of diffusion state within the monitoring interval.

[0012] Furthermore, the specific method for obtaining the organic matter content of the sediment surface layer and the iron and manganese ion concentration in the water, calculating the degradation rate of organic matter at adjacent time points, comparing the fluctuation amplitude of the degradation rate and the iron and manganese ion concentration, and generating the synchronization rate of organic matter and metal fluctuations includes:

[0013] Data on organic matter content in the sediment surface and data on iron and manganese ion concentrations in the water were obtained. The organic matter content difference between adjacent time points was subtracted, and the difference was divided by the corresponding time interval to obtain the degradation rate. The rate was then matched with the time interval to generate an organic matter degradation rate value.

[0014] The organic matter degradation rate value is called, and the difference between the maximum and minimum values ​​of the iron ion concentration and manganese ion concentration in the same time period is calculated. The difference is used as the fluctuation amplitude in that time period, and the amplitude value is recorded in correspondence with the time period to obtain the iron and manganese ion fluctuation amplitude value.

[0015] Based on the organic matter degradation rate value and the iron and manganese ion fluctuation amplitude value, the numerical change sequence of the two within the same time period is compared, the proportion of times the change direction is consistent is calculated, and the corresponding coefficient between the two is established based on the proportion result to obtain the synchronization rate of organic matter and metal fluctuation.

[0016] Furthermore, the specific method for screening degradation rate and iron-manganese ion concentration positively changing segments based on the synchronization rate of organic matter and metal fluctuations, recombining the corresponding segments' organic matter degradation rate, iron-manganese ion concentration fluctuations, pH value, water temperature, and flow rate as a feature set, and calculating the average change amplitude of the feature set over a continuous time period to obtain the coupling release intensity value includes:

[0017] Based on the synchronization rate of organic matter and metal fluctuations, time segments in which the degradation rate and the iron and manganese ion concentration values ​​change upward or downward simultaneously are screened. The segments that meet the conditions are numbered and matched with the original time axis to obtain the positive change time segments.

[0018] The positive change time segments are called, and five types of data are extracted from these time segments: organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature, and flow rate. The five types of data are recombined into a feature matrix in chronological order, and the matrix is ​​matched with the segment identifier to obtain the time segment feature set.

[0019] Based on the time segment feature set, the numerical differences of five types of data between adjacent segments within a continuous time period are calculated. The absolute value of each difference is taken and the average value is calculated. This average value is used as the overall amplitude and corresponds to the time period to obtain the coupling release strength value.

[0020] Furthermore, the specific formula for calculating the numerical difference between adjacent segments of the five types of data within the continuous time period is as follows:

[0021] ,

[0022] in, Represents a time segment The combined magnitude of the differences between adjacent segments of the following five data categories. Representing the Class data in time segments The value, Representing the Class data in the previous time segment The value, +1 represents the amplitude adjustment coefficient at the current moment. The category index has a value range of 1 to 5. For time segment indexing, The symbols represent the cumulative addition operation on five types of data. To represent the square root operation, This indicates the absolute value operation, and the denominator 5 represents the quantity of the five types of data.

[0023] Furthermore, the specific method for calling the coupling release intensity value, collecting dissolved oxygen concentrations in different water layers, calculating the dissolved oxygen gradient difference between adjacent water layers, and comparing the gradient difference with the coupling release intensity value to obtain the longitudinal dissolved oxygen difference includes:

[0024] Based on the coupling release intensity value, dissolved oxygen concentration data are collected in different water layers to obtain the dissolved oxygen concentration difference between adjacent water layers. The dissolved oxygen gradient difference is calculated and compared with the coupling release intensity value to obtain the dissolved oxygen gradient difference.

[0025] The dissolved oxygen gradient difference is called up to further calculate the average value of the gradient difference between each water layer, and these average values ​​are compared and analyzed to identify the relationship between the gradient difference and the coupling release intensity value, so as to obtain the average value of the dissolved oxygen gradient.

[0026] Based on the comparison between the dissolved oxygen gradient difference and the average dissolved oxygen gradient difference, the variation range of the longitudinal dissolved oxygen difference is calculated, and this range is mapped to a time period to obtain the longitudinal dissolved oxygen difference.

[0027] Furthermore, the specific method for collecting nitrogen and phosphorus concentrations of the same water layer based on the longitudinal dissolved oxygen difference, calculating the nitrogen and phosphorus concentration difference, and jointly calculating the nitrogen-phosphorus-oxygen difference joint coefficient with the longitudinal dissolved oxygen difference includes:

[0028] Based on the longitudinal dissolved oxygen difference, nitrogen and phosphorus concentration data are collected at the same water layer location. The difference between the nitrogen and phosphorus concentration values ​​corresponding to each water layer is calculated to obtain the nitrogen and phosphorus difference sequence of each water layer. All sequences are then sorted to generate nitrogen and phosphorus concentration difference values.

[0029] The nitrogen and phosphorus concentration difference value and the longitudinal dissolved oxygen difference value are called to establish paired data at the corresponding positions of the same water layer. The ratio of nitrogen and phosphorus difference to dissolved oxygen difference value of each water layer is calculated to obtain the nitrogen and phosphorus to oxygen ratio of each water layer. The ratio of all water layers is weighted and averaged to obtain the nitrogen, phosphorus and oxygen ratio coefficient.

[0030] Based on the nitrogen-phosphorus-oxygen ratio coefficient and the longitudinal dissolved oxygen difference, a joint calculation is performed on each water layer to calculate the comprehensive influence of the nitrogen-phosphorus concentration difference and the longitudinal dissolved oxygen difference. The comprehensive value is then statistically analyzed to generate a nitrogen-phosphorus-oxygen difference joint coefficient.

[0031] Furthermore, the specific method for calling the nitrogen-phosphorus-oxygen difference joint coefficient, comparing it with the diffusion determination threshold, marking the diffusion state when it exceeds the threshold, and calculating the duration of this state within the monitoring interval to obtain the longitudinal diffusion duration includes:

[0032] The nitrogen-phosphorus-oxygen difference joint coefficient is called, and the value is compared with the set diffusion judgment threshold. The status of each monitoring time point is marked according to the comparison result. When the nitrogen-phosphorus-oxygen difference joint coefficient is greater than the diffusion judgment threshold, it is marked as diffusion state. The time interval of continuous marking is statistically analyzed to obtain the diffusion state duration value.

[0033] Based on the diffusion state duration value, the cumulative duration within the monitoring interval is calculated, the duration of all continuous diffusion states is added together to obtain the total overall duration, and this is used as a characterization of the longitudinal diffusion intensity to generate the longitudinal diffusion duration.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] This invention proposes a rural water pollution monitoring method based on multi-source heterogeneous data. This method reveals the coupling relationship between decomposition and metal release by comparing the concentrations of organic matter and iron / manganese ions in sediments. It further filters positive change segments and constructs a feature set by combining pH, water temperature, flow velocity, and other factors to quantitatively characterize the intensity of pollution release. Comparing this intensity value with the dissolved oxygen gradient difference refines the analysis of oxygen differences between water layers and forms a coefficient with the nitrogen and phosphorus concentration difference, thus reflecting the correlation between nutrient diffusion and oxygen changes. Finally, through threshold determination and duration calculation, it achieves dynamic quantification of longitudinal diffusion intensity and duration, improving the interpretability of pollutant migration patterns and the precision of monitoring results.

[0036] This invention focuses on integrating information from different monitoring devices and data sources, particularly sensor data, satellite remote sensing data, and meteorological data related to water quality monitoring, to form a unified data platform for analysis and processing. The method mainly involves collecting various types of monitoring data through Internet of Things (IoT) technology, integrating data from different sources using data fusion technology, and achieving continuous monitoring and early warning of water pollution through data analysis. By employing efficient multi-source data integration and processing methods, the comprehensiveness and accuracy of water quality monitoring are improved. Attached Figure Description

[0037] Figure 1 This invention relates to a rural water pollution monitoring method based on multi-source heterogeneous data. Detailed Implementation

[0038] The present invention will be further illustrated below with reference to the accompanying drawings and specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art will fall within the scope defined by the appended claims.

[0039] Example 1:

[0040] A method for monitoring rural water pollution based on multi-source heterogeneous data, such as Figure 1 As shown, it includes the following steps:

[0041] S1. Obtain data on the organic matter content of the sediment surface and the iron and manganese ion concentration in the water body, calculate the degradation rate of organic matter between adjacent time points, compare the degradation rate with the fluctuation range of iron and manganese ion concentration in the same time period, and generate the organic matter and metal fluctuation synchronization rate.

[0042] S2. Based on the synchronization rate of organic matter and metal fluctuations, screen time segments in which the degradation rate and ion concentration change in a positive direction. Recombine the corresponding data of organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature and flow rate in these time segments into a feature set, and calculate the average change of the feature set in the continuous time period to obtain the coupling release intensity value.

[0043] S3. Call the coupling release intensity value, collect dissolved oxygen concentration data at different depths in the water body, calculate the dissolved oxygen gradient difference between adjacent water layers, and compare the dissolved oxygen gradient difference with the coupling release intensity value to obtain the longitudinal dissolved oxygen difference value.

[0044] S4. Based on the longitudinal dissolved oxygen difference, collect nitrogen and phosphorus concentration data between the same water layers, calculate the nitrogen and phosphorus concentration difference, and combine the nitrogen and phosphorus concentration difference with the longitudinal dissolved oxygen difference to generate the nitrogen, phosphorus and oxygen difference joint coefficient.

[0045] S5. Call the nitrogen-phosphorus-oxygen difference joint coefficient and compare it with the diffusion judgment threshold. When its value exceeds the threshold, mark the diffusion state and calculate the longitudinal diffusion duration based on the duration of the state within the monitoring interval.

[0046] The synchronization rate of organic matter and metal fluctuations includes the degree of synchronization between the organic matter degradation rate and the fluctuation range of iron and manganese ion concentrations; the coupling release intensity value includes the average variation range of degradation rate, iron and manganese ion concentration fluctuations, pH value, water temperature, and flow rate within a specific time period; the longitudinal dissolved oxygen difference value includes the difference in dissolved oxygen concentration gradient between adjacent water layers; the nitrogen-phosphorus and oxygen difference joint coefficient includes the joint calculation result of the nitrogen-phosphorus concentration difference and the longitudinal dissolved oxygen gradient difference; the longitudinal diffusion duration includes the duration of the diffusion state within the monitoring interval;

[0047] S101. Obtain the organic matter content data of the sediment surface layer and the iron ion concentration and manganese ion concentration data in the water body. Perform subtraction operation based on the difference in organic matter content values ​​between adjacent time points, then divide the value difference by the corresponding time interval length to obtain the degradation rate, and match the rate with the time period to generate the organic matter degradation rate value.

[0048] First, the change in organic matter content between adjacent time points is explained. For example, if the organic matter content is observed to be 4.5 mg / g at time point 1 and 3.9 mg / g at time point 2, the difference needs to be obtained by subtracting the adjacent values. The difference is 0.6 mg / g. Then, this difference needs to be compared with the interval between adjacent time points and divided. Assuming the interval between time point 1 and time point 2 is 12 hours, the degradation rate is calculated as 0.6 ÷ 12 = 0.05 mg / (g·h). In this process, the starting and ending values ​​of each time period need to be called sequentially and subtracted, and then the interval length needs to be called and the division needs to be performed. All the calculated results need to be correlated with the time period one by one. This action needs to be repeated throughout the process and continuously recorded. In addition, boundary cases in the calculation need to be judged. For example, if the difference is negative, the difference value needs to be directly taken and divided by the interval without absolute value processing, so as to obtain the rate directionality that can be positive or negative. Finally, all rate results are matched with the time periods one by one to generate the organic matter degradation rate value.

[0049] S102. Call the organic matter degradation rate value, calculate the difference between the maximum and minimum values ​​of the iron ion concentration and manganese ion concentration in the same time period, take the obtained difference as the fluctuation amplitude of the time period, and record the amplitude value in correspondence with the time period to obtain the iron and manganese ion fluctuation amplitude value.

[0050] First, in time period 1, the iron ion concentration sequence data is retrieved, for example, values ​​of 0.32, 0.41, 0.37, and 0.29 mg / L. A comparison operation is then performed to find the maximum value of 0.41 and the minimum value of 0.29, and the difference is subtracted to obtain a difference of 0.12 mg / L. Subsequently, in the same time period, the manganese ion concentration sequence data is retrieved, for example, values ​​of 0.11, 0.15, 0.13, and 0.14 mg / L. A comparison operation is then performed to find the maximum value of 0.15 and the minimum value of 0.11, and the difference is subtracted to obtain a difference of 0.04 mg / L. Finally, the iron ion difference and the manganese ion difference are added together to obtain 0.16 mg / L as the fluctuation range for time period 1. This process needs to be repeated in each time period. The maximum and minimum values ​​are called sequentially, subtraction is performed, and addition is performed. If the difference between the sequence values ​​is small within a certain time period, for example, the difference between the maximum and minimum iron ion concentration values ​​is only 0.05 mg / L, then the fluctuation range is determined to be in the range of 0–0.1 mg / L. If the difference exceeds 0.3 mg / L, it is classified as a high amplitude range. In the example, if the iron ion concentration sequence in time period 2 is 0.35, 0.62, 0.58, 0.51 mg / L, then the difference is 0.27 mg / L. If the manganese ion concentration sequence is 0.12, 0.20, 0.18, 0.15 mg / L, then the difference is 0.08 mg / L. After merging, the amplitude is 0.35 mg / L. Finally, all amplitude results are matched with time periods one by one to obtain the iron and manganese ion fluctuation amplitude values.

[0051] S103. Based on the organic matter degradation rate value and the iron and manganese ion fluctuation amplitude value, compare the numerical change sequence of the two in the same time period, calculate the proportion of times the change direction is consistent, and establish the corresponding coefficient between the two based on the proportion result to obtain the synchronization rate of organic matter and metal fluctuation.

[0052] First, based on the organic matter degradation rate and the fluctuation range of iron and manganese ions, the numerical change sequences of the two are compared point by point within the same time period. The process requires determining the direction of change between adjacent time periods. For example, in time period 1 to time period 2, the organic matter degradation rate changes from 0.05 mg / (g·h) to 0.07 mg / (g·h), which is an increase, while the fluctuation range of iron and manganese ions changes from 0.16 mg / L to 0.35 mg / L, also an increase. Therefore, the comparison result for this time period is consistent. If in the next time period the organic matter degradation rate decreases while the fluctuation range of iron and manganese ions increases, then the comparison... The result is inconsistent direction. The comparison results of all time periods need to be recorded one by one. The number of consistent times needs to be accumulated through counting operations in the whole sequence. Then, the total number of comparisons is called to perform a division operation to obtain the consistency ratio. For example, in the comparison of 10 time periods, the number of consistent times is 7, then the consistency ratio is 7÷10=0.7, which is 70%. In this process, the judgment rules need to be clearly defined. If the difference between the two is zero, it is considered that the direction is consistent. If the two values ​​change in opposite directions, it is considered that they are inconsistent. Finally, the obtained ratio result is directly used as the corresponding coefficient of the two and linked with the time period to obtain the synchronization rate of organic matter and metal fluctuations.

[0053] S201. Based on the synchronization rate of organic matter and metal fluctuations, screen time segments in which the degradation rate and the iron and manganese ion concentration values ​​change upward or downward simultaneously. Number the segments that meet the conditions and match them with the original time axis to obtain the positive change time segments.

[0054] First, the synchronization rate sequence calculated in the previous stage needs to be retrieved. Within each time period, the fluctuation amplitude values ​​of the organic matter degradation rate and iron / manganese ion concentration sequence are extracted one by one. It is then determined whether the direction of change of these two values ​​is consistent between the current and previous periods. If both are increasing or both are decreasing, the period is marked as a positive period. For example, in the time period... The rate of degradation of internal organic matter from Rise to "day" represents the day, and the direction is upward, while the fluctuation range of iron and manganese ions is... Rise to If the direction is also upward, then this segment is selected as a segment that meets the criteria within the time period. If the rate of organic matter degradation is Change to The direction is upward, but the fluctuation range of iron and manganese ions is from Descending to If the direction is downward, the segment does not meet the condition and is excluded. The filtering action needs to cyclically judge all adjacent segments. The judgment criterion is to assign a value of 1 if the direction is the same and to assign a value of 0 if the direction is different. When the value is assigned to 1, the segment number is recorded and matched one by one with the original time axis to finally form a positive change time segment.

[0055] S202. Call the positive change time segments, extract five types of data from these time segments: organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature, and flow rate. Recombine the five types of data into a feature matrix in chronological order, and match the matrix with the segment identifier to obtain the time segment feature set.

[0056] First, within this time period, five types of data were extracted: organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature, and flow rate. During the extraction process, the original monitoring sequences needed to be retrieved. For example, within the time range corresponding to fragment number 1, the organic matter degradation rate was... The iron and manganese ion concentration fluctuated as follows: The pH monitoring records ranged from 7.2 to 7.4, with an average of 7.3. The water temperature monitoring records were as follows: The flow velocity monitoring record is as follows The five types of data are arranged sequentially by time point and stored in a matrix row vector. Then, in the next segment number 2, the data of organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature and flow rate are extracted sequentially, and let them be respectively... , , , , Similarly, the data is combined into row vectors. The action is to generate a row of five-dimensional vectors under each segment number. The row vectors of all segments together form a feature matrix. The index of each row of the matrix corresponds to the time segment number, and finally the time segment feature set is obtained.

[0057] S203. Based on the time segment feature set, calculate the numerical difference between adjacent segments of five types of data in a continuous time period, take the absolute value of each difference and calculate the average value, take the average value as the overall amplitude and correspond it with the time period to obtain the coupling release strength value.

[0058] The specific formula for calculating the numerical difference between adjacent segments of the five types of data within a continuous time period is as follows:

[0059] ,

[0060] in, Represents a time segment The combined magnitude of the differences between adjacent segments of the following five data categories. Representing the Class data in time segments The value, Representing the Class data in the previous time segment The value, +1 represents the amplitude adjustment coefficient at the current moment, where, Representing the Class data in time segments The characteristic amplitude adjustment parameter in the middle; The category index has a value range of 1 to 5. For time segment indexing, The symbols represent the cumulative addition operation on five types of data. To represent the square root operation, This indicates the absolute value operation, and the denominator 5 represents the number of data in the five categories;

[0061] First, the organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature and flow rate of adjacent segments are calculated separately. Subtraction is performed on each type of data to obtain the difference result. Then, the absolute value of the difference is processed to obtain the average value. This average value is the overall amplitude value. During the execution, it is necessary to process all adjacent segments in a loop to generate the overall amplitude value segment by segment. Then, the overall amplitude value is recorded with the time period index to obtain the coupling release intensity value.

[0062] S301. Based on the coupling release intensity value, collect dissolved oxygen concentration data in different water layers, obtain the dissolved oxygen concentration difference between adjacent water layers, calculate the dissolved oxygen gradient difference, and compare the gradient difference with the coupling release intensity value to obtain the dissolved oxygen gradient difference.

[0063] First, the water layers need to be divided according to depth on the monitoring section. A sampling point is set at 1 meter intervals along the water depth direction. Dissolved oxygen concentration values ​​are obtained layer by layer and recorded under a time index. For example, the dissolved oxygen concentration at a water depth of 1 meter is... The dissolved oxygen concentration at a water depth of 2 meters is The dissolved oxygen concentration at a water depth of 3 meters is Then, a subtraction operation needs to be performed between adjacent water layers to obtain the concentration difference value. The difference between the first layer and the second layer is... The difference between the second and third layers is This process is repeated to obtain the difference sequence for all adjacent layers. Then, the difference sequence is compared with the coupling release intensity value. The comparison involves comparing the dissolved oxygen difference of each adjacent layer with the coupling release intensity value for the same time period. For example, if the coupling release intensity value is 0.1733, it is necessary to determine whether the dissolved oxygen difference is greater than or less than this value. A value greater than 0.1733 is recorded as a positive difference, and a value less than 0 is recorded as a negative difference. All judgment operations in this process are based on the difference minus the coupling release intensity value. If the result is greater than zero, it is marked as 1, and if it is less than zero, it is marked as 0. Finally, the dissolved oxygen gradient difference corresponding to the time period is obtained.

[0064] S302. Call the dissolved oxygen gradient difference value to further calculate the average value of the gradient difference between each water layer, and conduct comparative analysis based on these average values ​​to identify the relationship between the gradient difference and the coupling release intensity value, and obtain the average value of the dissolved oxygen gradient difference.

[0065] First, extract the differences between all water layers and calculate the average. The calculation involves adding all gradient differences within the same time period, then dividing by the number of water layers minus one. For example, if data from five water layers were collected within a depth range of 1 to 5 meters, the difference sequence would be... The average value of the dissolved oxygen gradient difference during this time period is This average value needs to be compared with the coupling release intensity value. The comparison is performed by subtracting the value at the same time index. The result is the average dissolved oxygen gradient difference minus the coupling release intensity value. For example, if the coupling release intensity value is 0.1733, then the difference is... If the result is positive, the gradient difference is marked as greater than the coupling release strength value; if it is negative, it is marked as less than. All the marking results are recorded in the time index to form a new dataset, and finally the average difference of dissolved oxygen gradient is obtained.

[0066] S303. Based on the comparison between the dissolved oxygen gradient difference and the average dissolved oxygen gradient difference, calculate the variation range of the longitudinal dissolved oxygen difference and correlate this range with the time period to obtain the longitudinal dissolved oxygen difference.

[0067] First, based on the comparison between the dissolved oxygen gradient difference and the average dissolved oxygen gradient difference, the execution process involves sequentially calling the dissolved oxygen gradient difference and the average dissolved oxygen gradient difference in a continuous time series. First, a subtraction operation is performed to obtain the relative difference, and then the absolute value is processed to represent the magnitude of change. For example, within a time period... The gradient difference is 0.4, the average difference is 0.475, then the magnitude of the longitudinal difference is... During the time period The gradient difference is 0.6, the average difference is 0.475, then the longitudinal difference magnitude is... The same operation was repeated for all time periods to obtain a longitudinal dissolved oxygen difference sequence. Then, the longitudinal dissolved oxygen differences were recorded one-to-one with the time periods, for example, in... Record the absolute value of the vertical difference as 0.075. Record the absolute value of the longitudinal difference as 0.125 to obtain the complete longitudinal dissolved oxygen difference;

[0068] S401. Based on the longitudinal dissolved oxygen difference, nitrogen and phosphorus concentration data are collected at the same water layer location. The difference between the nitrogen concentration value and the phosphorus concentration value corresponding to each water layer is calculated to obtain the nitrogen and phosphorus difference sequence of each water layer. All sequences are then sorted to generate nitrogen and phosphorus concentration difference values.

[0069] First, based on the longitudinal dissolved oxygen difference, the dissolved oxygen concentration of each water layer is measured. For example, using a sensor, the dissolved oxygen is measured to be 8 mg / L at a depth of 0-5 meters, 6 mg / L at 5-10 meters, and 4 mg / L at 10-15 meters. The dissolved oxygen difference between adjacent water layers is calculated. The difference between the surface and middle layers is taken as 8 - 6 = 2 mg / L, and the difference between the middle and bottom layers is taken as 6 - 4 = 2 mg / L. These differences are recorded as the longitudinal dissolved oxygen difference sequence. Subsequently, nitrogen and phosphorus concentration data are obtained at the same water layer location using a data logger. For example, if the nitrogen concentration is measured to be 0.5 mg / L and the phosphorus concentration to be 0.1 mg / L at the surface layer, the difference between the middle and bottom layers is taken as 6 - 4 = 2 mg / L. The nitrogen concentration measured at the first layer was 0.4 mg / L and the phosphorus concentration was 0.08 mg / L. The nitrogen concentration measured at the second layer was 0.3 mg / L and the phosphorus concentration was 0.06 mg / L. For each water layer, the nitrogen concentration value was subtracted from the phosphorus concentration value. For the surface layer, 0.5 minus 0.1 yielded 0.4 mg / L; for the middle layer, 0.4 minus 0.08 yielded 0.32 mg / L; and for the bottom layer, 0.3 minus 0.06 yielded 0.24 mg / L. This formed the nitrogen and phosphorus difference sequence for each water layer. These sequences were arranged into a list, for example, the sequence [0.4, 0.32, 0.24] mg / L, to finally generate the nitrogen and phosphorus concentration difference values.

[0070] S402. Call the nitrogen and phosphorus concentration difference value and the vertical dissolved oxygen difference value, establish paired data at the corresponding positions of the same water layer, calculate the ratio of nitrogen and phosphorus difference to dissolved oxygen difference value of each water layer, obtain the nitrogen and phosphorus to oxygen ratio of each water layer, and take the weighted average of the ratios of all water layers to obtain the nitrogen, phosphorus and oxygen ratio coefficient.

[0071] First, the nitrogen and phosphorus concentration difference values ​​and the vertical dissolved oxygen difference values ​​are retrieved. For example, the nitrogen and phosphorus concentration difference value sequence is [0.4, 0.32, 0.24] mg / L, and the vertical dissolved oxygen difference value sequence is [2, 2] mg / L (corresponding to the differences between the surface and middle layers and between the middle and bottom layers). Paired data are established at corresponding positions in the same water layer. The surface nitrogen and phosphorus difference of 0.4 is paired with the vertical dissolved oxygen difference of 2, the middle layer nitrogen and phosphorus difference of 0.32 is paired with the vertical dissolved oxygen difference of 2, and there is no direct pairing for the bottom layer because the vertical dissolved oxygen difference only has two values. Assuming that the water layer pairing is based on depth order, the surface and middle layers are paired. The nitrogen and phosphorus differences of each water layer are then calculated. The ratio of dissolved oxygen difference is calculated as follows: the surface layer ratio is 0.4 divided by 2, which equals 0.2; the middle layer ratio is 0.32 divided by 2, which equals 0.16. This yields the nitrogen, phosphorus, and oxygen ratio sequence for each water layer [0.2, 0.16]. The ratios of all water layers are then weighted and averaged. The weights are set based on the water layer depth ratio, for example, 0.33 for the surface layer and 0.33 for the middle layer (assuming equal depth). The weighted average is calculated as (0.2*0.33+0.16*0.33) / (0.33+0.33)=(0.066+0.0528) / 0.66=0.18, thus obtaining the nitrogen, phosphorus, and oxygen ratio coefficient.

[0072] S403. Based on the nitrogen-phosphorus-oxygen ratio coefficient and the longitudinal dissolved oxygen difference, perform joint calculations on each water layer, calculate the comprehensive influence of the nitrogen-phosphorus concentration difference and the longitudinal dissolved oxygen difference, and perform overall statistics on this comprehensive value to generate the nitrogen-phosphorus-oxygen difference joint coefficient.

[0073] First, based on the nitrogen-phosphorus-oxygen ratio coefficient and the longitudinal dissolved oxygen difference, for example, if the nitrogen-phosphorus-oxygen ratio coefficient is 0.18 and the longitudinal dissolved oxygen difference sequence is [2,2] mg / L, a joint calculation is performed on each water layer to calculate the comprehensive influence of the nitrogen-phosphorus concentration difference and the longitudinal dissolved oxygen difference. The formula is: comprehensive influence = nitrogen-phosphorus-oxygen ratio coefficient * longitudinal dissolved oxygen difference, where the nitrogen-phosphorus-oxygen ratio coefficient is a scalar and the longitudinal dissolved oxygen difference is a sequence. For each difference, a multiplication is performed. For example, the comprehensive influence of the surface layer = 0.18 * 2 = 0.36, and the comprehensive influence of the middle layer = 0.18 * 2 = 0.36, resulting in the comprehensive influence sequence [0.36, 0.36]. The comprehensive value is then statistically analyzed as a whole, and the average value is (0.36 + 0.36) / 2 = 0.36, generating the joint coefficient of nitrogen-phosphorus and oxygen difference.

[0074] S501. Call the nitrogen-phosphorus-oxygen difference joint coefficient, compare the value with the set diffusion judgment threshold, mark the status of each monitoring time point according to the comparison result, mark the diffusion state when the nitrogen-phosphorus-oxygen difference joint coefficient is greater than the diffusion judgment threshold, and statistically analyze the continuous marked time intervals to obtain the diffusion state duration value.

[0075] First, based on real-time numerical data collected at monitoring points, such as nitrogen concentration (N) in milligrams per liter (mg / L), phosphorus concentration (P) in milligrams per liter, surface dissolved oxygen (DO_s) in milligrams per liter, and bottom dissolved oxygen (DO_b) in milligrams per liter, the absolute value of the oxygen difference ΔDO is calculated as the absolute value of surface dissolved oxygen minus the absolute value of bottom dissolved oxygen, i.e., ΔDO = |DO_s - DO_b|. The weights are set as follows: nitrogen weight w_N = 0.4, phosphorus weight w_P = 0.4, and oxygen difference weight w_DO = 0.2. These weights are determined based on the relative importance of nitrogen, phosphorus, and oxygen difference on diffusion in historical monitoring data. The joint coefficient C_joint of nitrogen, phosphorus, and oxygen difference is then calculated through a weighted sum, i.e., C... _joint = w_N*N + w_P*P + w_DO*ΔDO. For example, at time point t1, nitrogen concentration N = 0.5 mg / L, phosphorus concentration P = 0.05 mg / L, surface dissolved oxygen DO_s = 8 mg / L, and bottom dissolved oxygen DO_b = 6 mg / L. Then ΔDO = |8 - 6| = 2 mg / L, C_joint = 0.4*0.5 + 0.4*0.05 + 0.2*2 = 0.2 + 0.02 + 0.4 = 0.62. The diffusion determination threshold T_diff is set to 0.5. This threshold is based on the distribution of C_joint values ​​in historical data from multiple monitoring points, taking the median or a specific percentile, such as 7. The 5% quantile is determined to ensure the identification of significant diffusion events. C_joint and T_diff are compared using a numerical comparison operation. If C_joint is greater than T_diff, the time point is marked as a diffusion state; otherwise, it is marked as a non-diffusion state. For example, C_joint = 0.62 is greater than T_diff = 0.5, therefore t1 is marked as a diffusion state. This process is repeated for each monitoring time point. Assuming the time point sequence is from t1 to t10, with a time interval Δt = 1 hour, the C_joint values ​​are 0.62, 0.7, 0.3, 0.8, 0.9, 0.4, 0.6, 0.55, 0.75, and 0. 5. T_diff=0.5, the comparison results are labeled as diffusion, diffusion, non-diffusion, diffusion, diffusion, non-diffusion, diffusion, diffusion, diffusion, non-diffusion, and the time intervals of continuous diffusion states are identified, such as t1-t2 continuous diffusion, t4-t5 continuous diffusion, and t7-t9 continuous diffusion. The duration of each interval is calculated. The interval duration is equal to the number of continuous diffusion time points multiplied by the time interval Δt. The t1-t2 interval has 2 points, so the duration is 2*1=2 hours. The t4-t5 interval has 2 points, so the duration is 2*1=2 hours. The t7-t9 interval has 3 points, so the duration is 3*1=3 hours. The duration value of each continuous diffusion state interval is obtained.

[0076] S502. Based on the duration value of the diffusion state, calculate the cumulative duration within the monitoring interval, add up the duration of all continuous diffusion states to obtain the total duration, and use it as a characterization of the longitudinal diffusion intensity to generate the longitudinal diffusion duration.

[0077] First, based on the duration data of each continuous diffusion state interval, such as interval duration values ​​d1=2 hours, d2=2 hours, d3=3 hours, the monitoring interval is from time point t1 to t10, with a total time span of 10 hours. The cumulative duration is calculated by adding the duration of all continuous diffusion states, that is, the total duration D_total=d1+d2+d3=2+2+3=7 hours. This total is used as a characterization of the longitudinal diffusion intensity to generate the longitudinal diffusion duration.

[0078] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for monitoring rural water pollution based on multi-source heterogeneous data, characterized in that, include: The organic matter content of the sediment surface and the iron and manganese ion concentration in the water were obtained. The degradation rate of organic matter at adjacent time points was calculated. The fluctuation range of degradation rate and iron and manganese ion concentration were compared to generate the synchronization rate of organic matter and metal fluctuation. Based on the synchronization rate of organic matter and metal fluctuations, segments with positive changes in degradation rate and iron and manganese ion concentration are screened. The corresponding segments are recombined as feature sets, including organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature, and flow rate. The average change amplitude of the feature set over a continuous time period is calculated to obtain the coupling release intensity value. The coupling release intensity value is used to collect dissolved oxygen concentrations in different water layers, calculate the dissolved oxygen gradient difference between adjacent water layers, and compare the gradient difference with the coupling release intensity value to obtain the longitudinal dissolved oxygen difference value. Based on the longitudinal dissolved oxygen difference, nitrogen and phosphorus concentrations of the same water layer are collected, the nitrogen and phosphorus concentration difference is calculated, and the nitrogen and phosphorus concentration difference is calculated together with the longitudinal dissolved oxygen difference to generate a nitrogen-phosphorus-oxygen difference joint coefficient. The nitrogen-phosphorus-oxygen difference joint coefficient is called, compared with the diffusion determination threshold, and the diffusion state is marked when the threshold is exceeded. The duration of this state in the monitoring interval is calculated to obtain the longitudinal diffusion duration. The specific method for screening degradation rate and iron-manganese ion concentration positively changing fragments based on the synchronization rate of organic matter and metal fluctuations, recombining the corresponding fragments' organic matter degradation rate, iron-manganese ion concentration fluctuations, pH value, water temperature, and flow rate as a feature set, and calculating the average change amplitude of the feature set over a continuous time period to obtain the coupling release intensity value includes: Based on the synchronization rate of organic matter and metal fluctuations, time segments in which the degradation rate and iron and manganese ion concentration values ​​change upward or downward simultaneously are screened. The segments that meet the conditions are numbered and matched with the original time axis to obtain the positive change time segments. The positive change time segments are called, and five types of data are extracted from these time segments: organic matter degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature, and flow rate. The five types of data are recombined into a feature matrix according to the time order, and the matrix is ​​matched with the segment identifier to obtain the time segment feature set. Based on the time segment feature set, calculate the numerical difference of five types of data between adjacent segments within a continuous time period, take the absolute value of each difference and calculate the average value, use the average value as the overall amplitude and correspond it with the time period to obtain the coupling release strength value. The specific method for collecting nitrogen and phosphorus concentrations of the same water layer based on the longitudinal dissolved oxygen difference, calculating the nitrogen and phosphorus concentration difference, and jointly calculating the nitrogen-phosphorus-oxygen difference joint coefficient with the longitudinal dissolved oxygen difference includes: Based on the longitudinal dissolved oxygen difference, nitrogen and phosphorus concentration data are collected at the same water layer location. The difference between the nitrogen and phosphorus concentration values ​​corresponding to each water layer is calculated to obtain the nitrogen and phosphorus difference sequence of each water layer. All sequences are then sorted to generate nitrogen and phosphorus concentration difference values. The nitrogen and phosphorus concentration difference value and the longitudinal dissolved oxygen difference value are called to establish paired data at the corresponding positions of the same water layer. The ratio of nitrogen and phosphorus difference to dissolved oxygen difference value of each water layer is calculated to obtain the nitrogen and phosphorus to oxygen ratio of each water layer. The ratio of all water layers is weighted and averaged to obtain the nitrogen, phosphorus and oxygen ratio coefficient. Based on the nitrogen-phosphorus-oxygen ratio coefficient and the longitudinal dissolved oxygen difference, a joint calculation is performed on each water layer to calculate the comprehensive influence of the nitrogen-phosphorus concentration difference and the longitudinal dissolved oxygen difference. The comprehensive values ​​are then statistically analyzed to generate a joint coefficient of nitrogen-phosphorus and oxygen difference.

2. The rural water pollution monitoring method based on multi-source heterogeneous data according to claim 1, characterized in that, The organic matter and metal fluctuation synchronization rate includes the degree of synchronization between the organic matter degradation rate and the iron and manganese ion concentration fluctuation amplitude; the coupling release intensity value includes the average change amplitude of degradation rate, iron and manganese ion concentration fluctuation, pH value, water temperature, and flow rate within a specific time period; the longitudinal dissolved oxygen difference value includes the dissolved oxygen concentration gradient difference between adjacent water layers; the nitrogen-phosphorus and oxygen difference joint coefficient includes the joint calculation result of nitrogen-phosphorus concentration difference and longitudinal dissolved oxygen gradient difference; the longitudinal diffusion duration includes the duration of diffusion state within the monitoring interval.

3. The rural water pollution monitoring method based on multi-source heterogeneous data according to claim 2, characterized in that, The specific method for obtaining the organic matter content of the sediment surface layer and the iron and manganese ion concentration in the water, calculating the degradation rate of organic matter at adjacent time points, comparing the fluctuation range of degradation rate and iron and manganese ion concentration, and generating the synchronization rate of organic matter and metal fluctuations includes: Data on organic matter content in the sediment surface and data on iron and manganese ion concentrations in the water were obtained. The organic matter content difference between adjacent time points was subtracted, and the difference was divided by the corresponding time interval to obtain the degradation rate. The rate was then matched with the time interval to generate an organic matter degradation rate value. The organic matter degradation rate value is called, and the difference between the maximum and minimum values ​​of the iron ion concentration and manganese ion concentration in the same time period is calculated. The difference is used as the fluctuation amplitude in that time period, and the amplitude value is recorded in correspondence with the time period to obtain the iron and manganese ion fluctuation amplitude value. Based on the organic matter degradation rate value and the iron and manganese ion fluctuation amplitude value, the numerical change sequence of the two within the same time period is compared, the proportion of times the change direction is consistent is calculated, and the corresponding coefficient between the two is established based on the proportion result to obtain the synchronization rate of organic matter and metal fluctuation.

4. The rural water pollution monitoring method based on multi-source heterogeneous data according to claim 3, characterized in that, The specific formula for calculating the numerical difference between adjacent segments of the five types of data within the continuous time period is as follows: ， in, Represents a time segment The combined magnitude of the differences between adjacent segments of the following five data categories. Representing the Class data in time segments The value, Representing the Class data in the previous time segment The value, +1 represents the amplitude adjustment coefficient at the current moment. The category index has a value range of 1 to 5. For time segment indexing, The symbols represent the cumulative addition operation on five types of data. To represent the square root operation, This indicates the absolute value operation, and the denominator 5 represents the quantity of the five types of data.

5. The rural water pollution monitoring method based on multi-source heterogeneous data according to claim 4, characterized in that, The specific method for obtaining the longitudinal dissolved oxygen difference by calling the coupling release intensity value, collecting dissolved oxygen concentrations in different water layers, calculating the dissolved oxygen gradient difference between adjacent water layers, and comparing the gradient difference with the coupling release intensity value includes: Based on the coupling release intensity value, dissolved oxygen concentration data are collected in different water layers to obtain the dissolved oxygen concentration difference between adjacent water layers. The dissolved oxygen gradient difference is calculated and compared with the coupling release intensity value to obtain the dissolved oxygen gradient difference. The dissolved oxygen gradient difference is called up to further calculate the average value of the gradient difference between each water layer, and these average values ​​are compared and analyzed to identify the relationship between the gradient difference and the coupling release intensity value, so as to obtain the average value of the dissolved oxygen gradient. Based on the comparison between the dissolved oxygen gradient difference and the average dissolved oxygen gradient difference, the variation range of the longitudinal dissolved oxygen difference is calculated, and this range is mapped to a time period to obtain the longitudinal dissolved oxygen difference.

6. The rural water pollution monitoring method based on multi-source heterogeneous data according to claim 5, characterized in that, The specific method for calling the nitrogen-phosphorus-oxygen difference joint coefficient, comparing it with the diffusion determination threshold, marking the diffusion state when it exceeds the threshold, and calculating the duration of this state in the monitoring interval to obtain the longitudinal diffusion duration includes: The nitrogen-phosphorus-oxygen difference joint coefficient is called, and the value is compared with the set diffusion judgment threshold. The status of each monitoring time point is marked according to the comparison result. When the nitrogen-phosphorus-oxygen difference joint coefficient is greater than the diffusion judgment threshold, it is marked as diffusion state. The time interval of continuous marking is statistically analyzed to obtain the diffusion state duration value. Based on the diffusion state duration value, the cumulative duration within the monitoring interval is calculated, the duration of all continuous diffusion states is added together to obtain the total overall duration, and this is used as a characterization of the longitudinal diffusion intensity to generate the longitudinal diffusion duration.

Images