Visual river and lake water area ecology supervision method based on big data of Internet of Things

Abstract

The invention belongs to the technical field of water area ecology monitoring, and provides a river and lake water area ecology visual supervision method based on Internet of Things big data, and the method comprises the steps: analyzing the overlapping performance of a monitoring position point and a historical abnormal position point; the priority of adding optimization to the non-overlapping position points is determined through abnormal conditions of the non-overlapping position points and Huffman distance deviation, optimization of the monitoring position points is achieved, the accuracy of ecological monitoring of the river and lake water areas is improved, the closeness degree of water quality parameters and water quality parameter standard values is calculated, and the accuracy of ecological monitoring of the river and lake water areas is improved. According to the method, a plurality of monitoring periods are monitored, prediction analysis of water quality parameter abnormity early warning is carried out, high-frequency abnormal position points are identified according to the repeated occurrence times of abnormal early warning position points in the plurality of monitoring periods, the high-frequency abnormal position points are compared with historical abnormal position points to identify high-frequency non-repeated position points, the historical abnormal position points are updated, and the efficiency of river and lake water area ecological visual supervision is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of water area ecological monitoring, and specifically relates to a method for visual supervision of river and lake water area ecology based on Internet of Things big data. Background Art

[0002] River and lake water areas are an important part of the human social environment and have important significance. For example, as an important water resource source, river and lake water areas provide rich water sources for agricultural irrigation, industrial water use, domestic water use, etc. However, in reality, there are still many deficiencies in the monitoring of river and lake water areas, such as untimely and inaccurate monitoring. Therefore, it is of great significance to study a method for visual supervision of river and lake water area ecology based on Internet of Things big data.

[0003] In the prior art, in the monitoring of river and lake water area ecology, key monitoring points are usually set to monitor the water quality parameters or ecological data of river and lake water areas to achieve the ecological monitoring of river and lake water areas. However, the problem is that when setting monitoring points, there is a lack of optimization analysis of the setting of monitoring points by combining historical monitoring anomalies. For example, specifically, when setting monitoring points, there are some monitoring positions in the river and lake water areas that have shown abnormal conditions in the past history. If the situation of historical abnormal monitoring points is not considered when setting monitoring points, it may lead to the inability to accurately detect or identify the ecological anomalies or water quality anomalies in the river and lake water areas; and in the prior art, the ecological data or water quality parameters of river and lake water areas are usually monitored and compared with corresponding thresholds to discover the ecological anomalies in the river and lake water areas, resulting in untimely ecological monitoring of the river and lake water areas; in the prior art, there is also a lack of targeted prediction when predicting and analyzing the ecology of river and lake water areas. For example, specifically, all ecological data or water quality parameters of river and lake water areas are usually predicted and analyzed, and there is a lack of prediction and analysis when some ecological data or water quality parameters are close to anomalies, resulting in low supervision and prediction efficiency; and in the prior art, without considering historical abnormal monitoring points, there is also a lack of dynamic update and optimization of historical abnormal monitoring points based on monitoring results, and no strategic basis can be provided for setting monitoring points in the ecological monitoring of river and lake water areas.

[0004] Therefore, the present invention provides a method for visual supervision of river and lake water area ecology based on Internet of Things big data. Summary of the Invention

[0005] In order to make up for the deficiencies of the prior art and solve at least one technical problem proposed in the background art.

[0006] The technical solution adopted by the present invention to solve its technical problems is as follows: In the first aspect, the present invention provides a method for visual supervision of river and lake water area ecology based on Internet of Things big data, including: Compare the monitored location points with the historical abnormal location points, identify non-overlapping location points among the historical abnormal location points, and perform processing and analysis to determine whether the set monitored location points need to be optimized. If so, generate an optimization signal; Based on the optimization signal, process and analyze the historical abnormal times of the non-overlapping location points within the historical monitoring period, the Huffman distance between the non-overlapping location points and the nearest adjacent monitored location points, and the water quality parameters when the non-overlapping location points had historical abnormalities. Sort the non-overlapping location points according to the analysis results, and optimize the set monitored location points according to the sorting results; After the optimization of the monitored location points is completed, obtain the water quality parameters of the monitored location points in real time and perform predictive analysis. Identify abnormal warning location points among the monitored location points and determine the predicted remaining warning duration of the abnormal warning location points, and synchronously send them to the river and lake water area visualization supervision platform; According to the number of times the abnormal warning location points appear repeatedly, identify high-frequency abnormal location points among the abnormal warning location points, compare them with the historical abnormal location points, identify high-frequency non-repeated location points, and update the historical abnormal location points through the high-frequency repeated location points.

[0007] In a second aspect, the present invention provides a river and lake water area ecological visualization supervision system based on Internet of Things big data, including: Monitoring optimization analysis module: After the monitored location points in the river and lake water area are set, compare the set monitored location points with the historical abnormal location points, identify non-overlapping location points among the historical abnormal location points, and perform processing and analysis. Judge whether the set monitored location points need to be optimized according to the analysis results. If so, generate an optimization signal; Monitoring optimization processing module: Based on the optimization signal, process and analyze the historical abnormal times of the non-overlapping location points within the historical monitoring period, the Huffman distance between the non-overlapping location points and the nearest adjacent monitored location points, and the water quality parameters when the non-overlapping location points had historical abnormalities. Sort the non-overlapping location points according to the analysis results, and optimize the set monitored location points according to the sorting results; Monitoring warning analysis module: After the optimization of the monitored location points is completed, within the monitoring period, perform predictive analysis on the water quality parameters of the monitored location points obtained in real time through Internet of Things technology. Identify abnormal warning location points among the monitored location points and determine the predicted remaining warning duration of the abnormal warning location points, and synchronously send them to the river and lake water area visualization supervision platform; Monitoring optimization update module: Within multiple monitoring periods, according to the number of times the abnormal warning location points appear repeatedly, identify high-frequency abnormal location points among the abnormal warning location points, and perform comparison and analysis with the historical abnormal location points. Identify high-frequency non-repeated location points, and update the historical abnormal location points through the high-frequency repeated location points.

[0008] The beneficial effects of the present invention are as follows: 1. By analyzing the overlap between the monitoring location points set during the monitoring of river and lake waters and the historical abnormal location points, it is determined whether the setting of the monitoring location points covers the historical abnormal location points. When the overlap between the monitoring location points and the historical abnormal location points does not meet the requirements, the priority of adding and optimizing the non-overlapping location points is determined through the abnormal conditions of the non-overlapping location points and the Huffman distance deviation, and the setting of the monitoring location points is optimized according to the priority. On the one hand, the present invention improves the rationality and comprehensiveness of the setting of monitoring points during the ecological monitoring of river and lake waters. On the other hand, through the addition and optimization of the monitoring location points, the accuracy of the ecological monitoring of river and lake waters is improved.

[0009] 2. Based on the optimization of the monitoring location points, predictive analysis is performed on the water quality parameters obtained in real time at the monitoring location points through the Internet of Things technology. Specifically, when the water quality parameters exceed the warning value of the water quality parameters but do not exceed the standard value of the water quality parameters, the degree of proximity between the water quality parameters and the standard value of the water quality parameters is calculated. When they are relatively close, predictive analysis of the abnormal warning of the water quality parameters is performed. On the one hand, by analyzing the proximity of the water quality parameters, it is beneficial to improve the efficiency of the ecological monitoring of river and lake waters. On the other hand, the ecological monitoring warning of river and lake waters is realized, ensuring the normal operation of the ecological environment of river and lake waters.

[0010] 3. By identifying the high-frequency abnormal location points through the number of times the abnormal warning location points repeatedly appear in multiple monitoring cycles, comparing them with the historical abnormal location points to identify the high-frequency non-repeated location points, and updating the historical abnormal location points through the high-frequency repeated location points. The present invention is beneficial to more reasonably and effectively set the monitoring location points in the future by identifying high-frequency abnormal and new abnormal location points and updating the historical abnormal location points, further improving the efficiency of the ecological visualization supervision of river and lake waters. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The present invention will be further described below with reference to the accompanying drawings.

[0012] Figure 1 is a flowchart of the steps of a method for ecological visualization supervision of river and lake waters based on Internet of Things big data according to an embodiment of the present invention; Figure 2 is a block diagram of a program of a system for ecological visualization supervision of river and lake waters based on Internet of Things big data according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0013] In order to make the technical means, creative features, achieved purposes and functions of the present invention easy to understand, the present invention will be further described below in conjunction with specific embodiments. Embodiment

[0014] As Figure 1 shown, a method for visual supervision of the ecological environment of river and lake waters based on Internet of Things big data according to an embodiment of the present invention includes: Step 1: After the monitoring position points in the river and lake waters are set, compare the set monitoring position points with the historical abnormal position points, identify non-overlapping position points among the historical abnormal position points, and perform processing and analysis to determine whether it is necessary to optimize the set monitoring position points. If necessary, generate an optimization signal; In some embodiments, the set monitoring position points are set in advance by technical personnel, which are not specifically limited herein. The historical abnormal position points refer to the monitoring position points where water quality parameters are abnormal (exceeding the standard values of water quality parameters) during the historical monitoring period. Among them, the water quality parameters include but are not limited to water temperature, pH value, heavy metal content, dissolved oxygen, etc.; the historical abnormal position points are obtained through the historical monitoring record reports of the river and lake waters during the historical monitoring period; Based on any one of the historical abnormal position points; Integrate the historical abnormal position points with the set monitoring position points respectively to obtain multiple position point combinations; For example, assume that the historical abnormal position point is LY(x,y); the existing monitoring position points are JC(x1,y1), JC(x2,y2), JC(x3,y3)......JC(xn,yn); then the obtained position point combinations are: [LY(x,y), JC(x1,y1)], [LY(x,y), JC(x2,y2)], [LY(x,y), JC(x3,y3)]......[LY(x,y), JC(xn,yn)]; where LY(x,y) is the longitude and latitude coordinates of the historical abnormal position point, and JC(xn,yn) is the longitude and latitude coordinates of the nth monitoring position point; If, within all the position point combinations, there is any position point combination where the historical abnormal position point and the monitoring position point have the same position, mark the historical abnormal position point as an overlapping position point; If, within all the position point combinations, there is no position point combination where the historical abnormal position point and all the historical abnormal position points have the same position, mark the historical abnormal position point as a non-overlapping position point; Respectively count the number of non-overlapping position points and the number of monitoring position points, and perform ratio processing to obtain the position overlap quantity deviation value, and mark it as PC; Based on the non-overlapping position points, obtain the longitude and latitude coordinates of the monitoring position point closest to the non-overlapping position point, combine the longitude and latitude coordinates of the non-overlapping position points, and calculate the Huffman distance between the non-overlapping position point and the closest adjacent monitoring position point through the Huffman distance method. Specifically, through the formula: The Huffman distance D is calculated, where r is the radius of the earth, x1 and y1 are the latitude and longitude of the non-overlapping position points respectively, and x2 and y2 are the latitude and longitude of the monitoring position points respectively; Combined with the position overlap quantity deviation value PC and the Huffman distance D between the non-overlapping position points and the nearest adjacent monitoring position points, through the formula: The position overlap performance value WP is obtained, where s1 and s2 are both preset proportionality coefficients, s1 takes the value of 1.216, s2 takes the value of 1.143, Di represents the Huffman distance D between the nearest adjacent monitoring position points of the i-th overlapping position point, and i represents the numbering of the non-overlapping position points; It should be noted that the meaning represented by the position overlap performance value is: by analyzing the position coincidence between the historical abnormal monitoring position points and the set monitoring position points, it is judged whether the set monitoring position points cover the historical abnormal monitoring position points when setting the monitoring position points, so as to prevent the problem that the set monitoring position points do not take into account the historical monitoring anomalies, resulting in inaccurate ecological supervision of river and lake waters; Preferably, the position overlap performance value is compared with the position overlap performance threshold; If the position overlap performance value is greater than the position overlap performance threshold, it means that the set monitoring position points need to be optimized to generate an optimization signal; If the position overlap performance value is less than or equal to the position overlap performance threshold, it means that the set monitoring position points do not need to be optimized; Step two: Based on the optimization signal, process and analyze the historical abnormal times of the non-overlapping position points within the historical monitoring period, the Huffman distance between the non-overlapping position points and the nearest adjacent monitoring position points, and the water quality parameters when the non-overlapping position points have historical anomalies. Sort the non-overlapping position points according to the analysis results, and optimize the set monitoring position points according to the sorting results; In some embodiments, the historical abnormal times of the non-overlapping position points are obtained through the historical monitoring record reports of the non-overlapping position points within the historical monitoring period; Specifically, the specific process of sorting the non-overlapping position points includes: Based on any non-overlapping position point; Count the historical abnormal times of the non-overlapping position points within the historical monitoring period, and perform a ratio process with the total number of times of river and lake water area monitoring within the historical monitoring period to obtain the historical abnormal times ratio; Respectively perform a difference process on the water quality parameters when the non-overlapping position points have historical anomalies and the water quality parameter standard values, and perform a ratio process on the mean value of the differences and the water quality parameter standard values to obtain the historical abnormal degree value; Sum the historical abnormal times ratio and the historical abnormal degree value to obtain the historical abnormal value; Sum the Huffman distance between non - overlapping position points and the nearest adjacent monitoring position points and the historical outliers to obtain the priority value of the non - overlapping position points; It should be noted that the priority value of the non - overlapping position points reflects the number of anomalies, the degree of anomalies of the non - overlapping position points in the historical monitoring period, and the Huffman distance between the non - overlapping position points and the nearest adjacent monitoring position points. Among them, the more the number of anomalies of the non - overlapping position points in the historical monitoring period, the higher the degree of anomalies, and the greater the Huffman distance between the non - overlapping position points and the nearest adjacent monitoring position points, the greater the possibility that the non - overlapping position points are abnormal and the greater the distance deviation between the non - overlapping position points and the nearest adjacent monitoring position points, and the smaller the possibility of being monitored, then the optimization priority is higher; Sort the non - overlapping position points from largest to smallest according to their priority values to obtain a non - overlapping position point sequence; Optimize the set monitoring position points according to the sorting result of the non - overlapping position points. Specifically: Select non - overlapping position points in sequence according to the non - overlapping position point sequence, and add new monitoring position points to the set monitoring position points according to the longitude and latitude coordinates of the non - overlapping position points. During the selection process of non - overlapping position points, after each non - overlapping position point is selected, calculate the position overlap performance value again until the position overlap performance value is less than or equal to the position overlap performance threshold, then stop the selection; The technical solution of the embodiment of the present invention is: By analyzing the overlap between the set monitoring position points and the historical abnormal position points during the monitoring of river and lake waters, to judge whether the setting of the monitoring position points covers the historical abnormal position points, and when the overlap between the monitoring position points and the historical abnormal position points does not meet the requirements, determine the optimization priority of adding non - overlapping position points through the abnormal situation of non - overlapping position points and the Huffman distance deviation, and realize the optimization of the setting of monitoring position points according to the priority. On the one hand, the present invention improves the rationality and comprehensiveness of the setting of monitoring points during the ecological monitoring of river and lake waters. On the other hand, through the addition and optimization of monitoring position points, the accuracy of the ecological monitoring of river and lake waters is improved. Example 1

[0015] As Figure 1 shown, based on Example 1, a method for visual supervision of the ecological environment of river and lake waters based on Internet of Things big data described in the embodiment of the present invention includes: Step 3: After the optimization of the monitoring position points is completed, during the monitoring period, predict and analyze the water quality parameters of the monitoring position points obtained in real time through the Internet of Things technology, identify the abnormal warning position points in the monitoring position points and determine the predicted remaining warning duration of the abnormal warning position points, and synchronously send them to the visual supervision platform of the river and lake waters; In some embodiments, using Internet of Things technology, corresponding water quality monitoring sensors are set at the set monitoring location points to obtain the water quality parameters of the monitoring location points in real time; Specifically, the process of predicting and analyzing water quality parameters is as follows: During the monitoring period, the water quality parameters of the monitoring location points are obtained in real time and compared with the warning values of the water quality parameters; If the water quality parameters exceed the warning values of the water quality parameters but do not exceed the standard values of the water quality parameters, a calculation signal is generated; If the water quality parameters do not exceed the warning values of the water quality parameters, no operation is performed; It should be noted that the warning values of the water quality parameters are pre-set by technical personnel, and the purpose is to perform predictive analysis when the water quality parameters exceed the warning values of the water quality parameters; Based on the calculation signal, the difference between the water quality parameters and the standard values of the water quality parameters is processed to obtain a numerical parameter approximation value, and the water quality parameter approximation value is compared with the water quality parameter approximation threshold; If the water quality parameter approximation value is less than or equal to the water quality parameter approximation threshold, it means that the current water quality parameter is close to the standard value of the water quality parameter, and a prediction signal is generated; If the water quality parameter approximation value is greater than the water quality parameter approximation threshold, it means that the current water quality parameter is not close to the standard value of the water quality parameter, and no operation is performed; It should be noted that the water quality parameter approximation threshold is pre-set, and the purpose is to judge whether the current water quality parameter is close to the standard value of the water quality parameter, so as to provide a basis for the subsequent prediction of the water quality parameter; Based on the prediction signal, according to the time point when the prediction signal appears and the time point when the water quality parameter reaches the warning value of the water quality parameter, the difference is processed to obtain the water quality analysis period; The water quality analysis period is divided into several analysis nodes, and the water quality parameters corresponding to the analysis nodes are integrated into a water quality parameter sequence (SZ 1 、SZ 2 、SZ 3 ......SZ n ), where SZ n represents the water quality parameter of the nth analysis node; Using the moving average method to process the water quality parameter sequence, determining the water quality parameter training feature set and target labels, and combining machine learning algorithms to predict the water quality parameters. Specifically: S1, obtain the water quality parameter sequence (SZ 1 、SZ 2 、SZ 3 ......SZ n ), and perform preprocessing, including but not limited to denoising, smoothing, etc.; S2. Define the moving window size (W = 4): Determine the number of running parameters included in the window; Define the step size (S = 1): Determine the step size for the window to move on the running parameter group. S3. Construct the training feature set and target labels: For each window, use the water quality parameters in the water quality parameter sequence as training features; Specifically, if the window size is W and the step size is S, then the training feature set for the first window is {SZ 1 、SZ 2 、SZ 3 ... SZ w}; Target label: Each training feature set corresponds to a target label; The target label is a running parameter after the window. For example, for the first window, the target label is SZ w +1; Exemplarily, assume the running parameter group is {1, 3, 5, 7, 9, 11, 13, 15, 17, 19}, the window size W = 4, and the step size S = 1. Then the moving window method can be used to construct the following feature set and label set: Feature set 1: {1, 3, 5, 7}, label: 9; Feature set 2: {3, 5, 7, 9}, label: 11; Feature set 3: {5, 7, 9, 11}, label: 13; ... Feature set 7: {13, 15, 17, 19}, label: none; S4. Model training: Use the training feature set and target labels to train the neural network model, and use the trained neural network model to predict the water quality parameters; If during the monitoring period, the predicted water quality parameters exceed the water quality parameter standard value, then mark the monitoring location point corresponding to the water quality parameter as an abnormal warning location point; If during the monitoring period, the predicted water quality parameters do not exceed the water quality parameter standard value, then do not perform any processing on the monitoring location point corresponding to the water quality parameter; Based on the abnormal warning location points, according to the prediction results of the water quality parameters, obtain the time point when the water quality parameters corresponding to the abnormal warning location points exceed the water quality parameter standard value, and combine it with the time point when the prediction signal appears, and perform a difference operation to obtain the predicted remaining warning duration of the abnormal warning location points; During the monitoring period, transmit the longitude and latitude coordinates of the abnormal warning location points and the corresponding predicted remaining warning duration to the river and lake water area visualization monitoring platform for display; The technical solution of the embodiment of the present invention is as follows: Based on the optimization of the monitoring location points, the water quality parameters obtained in real time through the Internet of Things technology at the monitoring location points are predicted and analyzed. Specifically, when the water quality parameters exceed the warning value of the water quality parameters but do not exceed the standard value of the water quality parameters, calculate the proximity of the water quality parameters to the standard value of the water quality parameters. When they are relatively close, conduct predictive analysis of water quality parameter anomaly warnings. On the one hand, by analyzing the proximity of the water quality parameters, it is beneficial to improve the efficiency of river and lake water area ecological monitoring. On the other hand, it realizes the ecological monitoring and early warning of river and lake water areas, ensuring the normal operation of the ecological environment of river and lake water areas.

[0016] Identifying abnormal warning location points and determining the predicted remaining warning duration of the abnormal warning location points in the monitoring location points, and synchronously sending them to the visual supervision platform for river and lake water areas Embodiment 2

[0017] As Figure 1 shown, based on Embodiment 1 and Embodiment 2, a method for visual supervision of the ecological environment of river and lake water areas based on Internet of Things big data according to an embodiment of the present invention includes: Step 4: In multiple monitoring cycles, according to the number of times the abnormal warning location points repeatedly appear, identify high-frequency abnormal location points among the abnormal warning location points, compare and analyze them with historical abnormal location points, identify high-frequency non-repeated location points, and update the historical abnormal location points through the high-frequency repeated location points; Specifically, in multiple monitoring cycles, based on each abnormal warning location point that appears, count the number of times the abnormal warning location point repeatedly appears, and perform a ratio process with the maximum value of the number of times the abnormal warning location point repeatedly appears to obtain the high-frequency occurrence value of the abnormal warning location point; Among them, the maximum value of the number of times the abnormal warning location point repeatedly appears is equal to the number of monitoring cycles; Compare the high-frequency occurrence value of the abnormal warning location point with the high-frequency occurrence threshold; If the high-frequency occurrence value of the abnormal warning location point is greater than or equal to the high-frequency occurrence threshold, mark the abnormal warning location point as a high-frequency abnormal location point; If the high-frequency occurrence value of the abnormal warning location point is less than the high-frequency occurrence threshold, mark the abnormal warning location point as a low-frequency abnormal location point; Compare the high-frequency abnormal location points with the historical abnormal location points. Specifically: If there is any historical abnormal location point among the historical abnormal location points whose longitude and latitude coordinates are the same as those of the high-frequency abnormal location point, do not perform any operation; If there are no longitude and latitude coordinates of historical abnormal location points in the historical abnormal location points that are the same as those of the high-frequency abnormal location points, the high-frequency abnormal location points are marked as high-frequency non-repeated location points, and the high-frequency non-repeated location points are added to the historical abnormal location points as historical abnormal location points; Based on the historical abnormal location points, if there are no longitude and latitude coordinates that are the same as those of any high-frequency abnormal location point during the comparison with the high-frequency abnormal location points, the historical abnormal location points are deleted; It should be noted that the reason for taking the high-frequency non-repeated location points as historical abnormal location points is that in multiple monitoring cycles, high-frequency and new abnormal location points have appeared. Therefore, when setting monitoring location points in subsequent monitoring, the situation of these high-frequency and new abnormal location points needs to be considered. Therefore, taking the high-frequency non-repeated location points as historical abnormal location points realizes the dynamic update of historical abnormal location points. The reason for deleting the historical abnormal location points is that the historical abnormal location points have not shown abnormalities or high frequencies in multiple monitoring cycles. Therefore, when setting monitoring location points in subsequent monitoring, they are deleted; The technical solution of the embodiment of the present invention is as follows: By identifying high-frequency abnormal location points based on the number of times abnormal warning location points repeatedly appear in multiple monitoring cycles, comparing them with historical abnormal location points to identify high-frequency non-repeated location points, and updating historical abnormal location points through high-frequency repeated location points. The present invention identifies high-frequency abnormal and new abnormal location points and updates historical abnormal location points, which is beneficial to setting more reasonable and effective monitoring location points in the future and further improving the efficiency of visual supervision of the ecological environment of river and lake waters. Embodiment 3

[0018] As Figure 2 shown, a visual supervision system for the ecological environment of river and lake waters based on Internet of Things big data according to an embodiment of the present invention includes: Monitoring optimization analysis module: After the monitoring location points in the river and lake waters are set, compare the set monitoring location points with the historical abnormal location points, identify non-overlapping location points in the historical abnormal location points, and perform processing and analysis. According to the analysis results, judge whether it is necessary to optimize the set monitoring location points. If necessary, generate an optimization signal; Monitoring optimization processing module: Based on the optimization signal, perform processing and analysis on the historical abnormal times of non-overlapping location points in the historical monitoring cycle, the Huffman distance between non-overlapping location points and the nearest adjacent monitoring location points, and the water quality parameters when non-overlapping location points have historical abnormalities. Sort the non-overlapping location points according to the analysis results, and optimize the set monitoring location points according to the sorting results; Monitoring and Early Warning Analysis Module: After the optimization of the monitoring location points is completed, during the monitoring period, it predicts and analyzes the water quality parameters of the monitoring location points obtained in real time through the Internet of Things technology, identifies the abnormal warning location points among the monitoring location points, and determines the predicted remaining warning duration of the abnormal warning location points, and synchronously sends them to the visualization supervision platform of the river and lake waters; Monitoring Optimization and Update Module: During multiple monitoring periods, according to the number of times the abnormal warning location points appear repeatedly, it identifies the high-frequency abnormal location points among the abnormal warning location points, compares and analyzes them with the historical abnormal location points, identifies the high-frequency non-repeated location points, and updates the historical abnormal location points through the high-frequency repeated location points.

[0019] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A visual supervision method for river and lake water ecology based on IoT big data, characterized by: include: Compare the monitoring location points with the historical abnormal location points, identify non-overlapping location points in the historical abnormal location points, and perform processing and analysis to determine whether the set monitoring location points need to be optimized. If necessary, generate an optimization signal; Based on the optimization signal, the number of historical anomalies of non-overlapping location points in the historical monitoring cycle, the Huffman distance between the non-overlapping location points and the nearest adjacent monitoring location points, and the water quality parameters when historical anomalies occurred at the non-overlapping location points are processed and analyzed, and the non-overlapping location points are sorted according to the analysis results, and the set monitoring location points are optimized according to the sorting results; After the optimization of the monitoring locations is completed, the water quality parameters of the monitoring locations are obtained in real time and predicted and analyzed. The abnormal warning locations are identified in the monitoring locations, and the predicted remaining warning time of the abnormal warning locations is determined, and sent to the visual supervision platform for rivers and lakes at the same time; According to the number of times the abnormal warning location points appear repeatedly, high-frequency abnormal location points are identified among the abnormal warning location points, and compared with the historical abnormal location points to identify high-frequency non-repeating location points, and the historical abnormal location points are updated through the high-frequency repeating location points.

2. According to claim 1, a method for visualizing the ecological supervision of rivers and lakes based on the Internet of Things big data, characterized in that: The non-overlapping position points are identified as follows: Integrate the historical abnormal location points with the set monitoring location points respectively to obtain multiple location point combinations; If, among all the location point combinations, there is any location point combination in which the historical abnormal location point is at the same position as the monitoring location point, the historical abnormal location point is marked as an overlapping location point.

3. According to claim 1, a method for visualizing the ecological supervision of rivers and lakes based on the Internet of Things big data, characterized in that: The optimization signal is generated in the following manner: The number of non-overlapping position points and the number of monitoring position points are counted respectively, and the ratio processing is performed to obtain the position overlap number deviation value, which is marked as PC; The Huffman distance D between the non-overlapping location point and the nearest adjacent monitoring location point is calculated by the Huffman distance method; Combined with the position overlap number deviation value PC and the Huffman distance D between the non-overlapping position point and the nearest adjacent monitoring position point, the formula is: The position overlap performance value WP is obtained, where s1 and s2 are both preset proportional coefficients, Di represents the Huffman distance D between the nearest monitoring position point adjacent to the i-th overlapping position point, and i represents the number of non-overlapping position points; If the position overlap performance value is greater than the position overlap performance threshold, an optimization signal is generated.

4. According to claim 1, a method for visualizing the ecological supervision of rivers and lakes based on the big data of the Internet of Things, characterized in that: The specific process of sorting the non-overlapping position points includes: Based on any non-overlapping location point; The number of historical anomalies at non-overlapping locations within the historical monitoring period is counted, and the ratio is processed with the total number of river and lake water monitoring within the historical monitoring period to obtain the historical anomaly ratio; The water quality parameters at each non-overlapping location point when a historical anomaly occurs are processed with the standard value of the water quality parameters, and the difference is averaged and then compared with the standard value of the water quality parameters to obtain the historical anomaly degree value; The historical anomaly value is obtained by summing the historical anomaly frequency ratio and the historical anomaly degree value; The Huffman distance and historical anomaly value between the non-overlapping location point and the nearest adjacent monitoring location point are summed to obtain the priority value of the non-overlapping location point, and the non-overlapping location points are sorted from large to small according to the priority value of the non-overlapping location point.

5. According to claim 1, a method for visualizing the ecological supervision of rivers and lakes based on the Internet of Things big data, characterized in that: The process of optimizing the set monitoring location points according to the sorting results is as follows: Non-overlapping position points are selected in sequence according to the non-overlapping position point sequence, and new monitoring position points are added to the set monitoring position points according to the latitude and longitude coordinates of the non-overlapping position points. In the process of selecting non-overlapping position points, after each non-overlapping position point is selected, the position overlap performance value is calculated again until the position overlap performance value is less than or equal to the position overlap performance threshold, then the selection is stopped.

6. According to claim 1, a method for visualizing the ecological supervision of rivers and lakes based on the big data of the Internet of Things, characterized in that: The process of identifying the abnormal warning location point and determining the predicted remaining warning time of the abnormal warning location point is as follows: Acquire water quality parameters at monitoring locations in real time and process and analyze them to obtain a water quality parameter sequence; The water quality parameter sequence is combined with the moving average method and the machine learning algorithm to predict the water quality parameters. If the predicted water quality parameters exceed the standard values ​​of the water quality parameters, the monitoring location points corresponding to the water quality parameters are marked as abnormal warning location points; Based on the abnormal warning location point, according to the prediction results of water quality parameters, the time point when the water quality parameter corresponding to the abnormal warning location point exceeds the water quality parameter standard value is obtained, and combined with the time point when the predicted signal appears, difference processing is performed to obtain the predicted remaining warning time of the abnormal warning location point.

7. The method for visualizing the ecological supervision of rivers and lakes based on the big data of the Internet of Things according to claim 6 is characterized by: The water quality parameter sequence is obtained in the following manner: The water quality parameters of the monitoring location are obtained in real time. If the water quality parameters exceed the water quality parameter warning value but do not exceed the water quality parameter standard value, the water quality parameters are subjected to difference processing with the water quality parameter standard value to obtain the numerical parameter approximation value; If the water quality parameter approximation value is less than or equal to the water quality parameter approximation threshold, a prediction signal is generated; According to the time point when the prediction signal appears and the time point when the water quality parameter reaches the water quality parameter warning value, the water quality analysis period is obtained by difference processing; The water quality analysis period is divided into several analysis nodes, and the water quality parameters corresponding to the analysis nodes are integrated into a water quality parameter sequence in chronological order.

8. According to claim 1, a method for visualizing the ecological supervision of rivers and lakes based on the Internet of Things big data, characterized in that: The method for identifying the high-frequency abnormal position point is: Count the number of times the abnormal warning location point recurs, and perform ratio processing with the maximum number of times the abnormal warning location point recurs to obtain the high-frequency occurrence value of the abnormal warning location point; Among them, the maximum number of times the abnormal warning location point recurs is equal to the number of monitoring cycles; If the high frequency occurrence value of the abnormal warning location point is greater than or equal to the high frequency occurrence threshold, the abnormal warning location point is marked as a high frequency abnormal location point.

9. According to claim 1, a method for visualizing the ecological supervision of rivers and lakes based on the Internet of Things big data, characterized in that: The identification method of the high-frequency non-repetitive position points is: Compare high-frequency anomaly locations with historical anomaly locations: If there is no historical abnormal position point in the historical abnormal position points whose longitude and latitude coordinates are the same as those of the high-frequency abnormal position point, the high-frequency abnormal position point is marked as a high-frequency non-repetitive position point.

10. The method for visualizing the ecological supervision of rivers and lakes based on the Internet of Things big data according to claim 1 is characterized by: include: The process of updating the historical abnormal location points is as follows: Add high-frequency non-repeated position points to the historical abnormal position points as historical abnormal position points; Based on the historical abnormal position point, if in the process of comparing with the high-frequency abnormal position point, there is no latitude and longitude coordinates that are the same as any high-frequency abnormal position point, the historical abnormal position point will be deleted.

Images