Sudden environmental pollution monitoring system

Abstract

The invention relates to the technical field of environmental monitoring, and discloses a sudden environmental pollution monitoring system which is composed of a monitoring module, a data processing module, a data analysis module, a pollution identification module, a source positioning module and a response module. Water quality sensors including dissolved oxygen, pH value, turbidity and heavy metal sensors are laid in a target water body, water quality changes are monitored in real time, continuity and timeliness of data are ensured, the sensors gather the data to a data processing module through wireless or satellite communication, and the data are acquired through a support vector machine and a decision tree algorithm. And the processed data are deeply analyzed, abnormal data which do not accord with a normal range are timely detected, a pollution mode can be rapidly identified, support is provided for timely early warning, and the beneficial effects of continuously monitoring the target water body and rapidly determining the sudden pollution area of the water body are achieved.

Description

Technical Field

[0001] The invention relates to the technical field of environmental monitoring, and in particular to a sudden environmental pollution monitoring system. Background Art

[0002] Sudden water pollution refers to sudden water pollution, the source of which is hazardous chemicals or other pollutants. Sudden water pollution has the characteristics of rapid pollution and difficulty in controlling the pollution source. Many sudden water pollution accidents occurred around the world because the pollution spread was not effectively controlled in time at the initial stage of the pollution leakage. By the time the pollution source was cut off, serious pollution consequences had already occurred. Therefore, timely blocking the spread of pollutants in the early stage of pollution is an effective measure to prevent and control sudden water pollution.

[0003] After searching, it was found that the publication number is CN102424447A, which is a device for controlling sudden water environmental pollution. When sudden water environmental pollution occurs, the control device is placed into the polluted waters, and then the driving device is turned on. The foldable device is quickly unfolded on the water surface under the driving action of the driving device to form a cofferdam or cover the water surface to intercept pollution.

[0004] Although the aforementioned technical solutions have effectively solved the problem of managing sudden water environmental pollution, they still cannot solve the problem of how to continuously monitor and quickly determine the areas of sudden water pollution. At the same time, the existing technical solutions are also unable to determine the specific source of pollution when sudden water environmental pollution is discovered during monitoring. Summary of the invention

[0005] Technical issues solved:

[0006] In view of the shortcomings of the existing technology, the present invention provides a sudden environmental pollution monitoring system, which has the advantages of continuously monitoring the target water body to quickly determine the sudden water pollution area, and determining the specific pollution source when sudden water environmental pollution is monitored, thereby solving the problems of the above-mentioned technology.

[0007] Technical solution:

[0008] To achieve the above object, the present invention provides the following technical solutions: a sudden environmental pollution monitoring system, the detection system is composed of a monitoring module, a data processing module, a data analysis module, a pollution identification module, a source location module and a response module;

[0009] The monitoring module is used to monitor the regional target water body, the regional target water body is mapped and zoned by satellite, water quality sensors are laid in the regional target water body, and the monitoring module also includes a data collection unit, the data collection unit sensor data is gathered by wireless or satellite communication, and the data is transmitted to the data processing module;

[0010] The data processing module is used to receive the regional target water body data collected by the data collection unit and perform real-time data processing on it: using a sliding window algorithm to process and analyze sensor data;

[0011] The data analysis module uses a type of support vector machine to detect data that does not conform to the normal range; and uses a decision tree algorithm for pollution pattern recognition and early warning, setting thresholds for water quality parameters. When the data in a certain area exceeds the normal range, it is marked as an abnormal area:

[0012] The source location module uses the K-means clustering algorithm to identify the concentrated areas of the polluted area, analyze the spatial distribution of abnormal data points, and determine the boundaries of the polluted area:

[0013] The response module is used to automatically activate an alarm when contamination is detected:

[0014] The water quality sensors in the monitoring module are: dissolved oxygen sensor, pH value sensor, turbidity sensor and heavy metal sensor.

[0015] Preferably, the regional target water body calculation expression in the monitoring module is:

[0016] Regional target water body zoning:

[0017] R = {(x,y)|x min ≤x≤x max ,y min ≤y≤y max}

[0018] Where: R is the target water area, which defines a rectangular area; (x, y) is the coordinate of a point in the area; x min and x max is the minimum and maximum coordinate value of the region on the x-axis; y min and max are the minimum and maximum coordinate values ​​of the region on the y-axis;

[0019] The water quality sensor data expression in the monitoring module is:

[0020] d i(t) =f i(pi,t)

[0021] Where: d i(t) is the data value of the i-th sensor at time t; f i is the sensor measurement function, which varies according to the sensor; p i are the water quality parameters monitored by sensor i: dissolved oxygen, pH value, turbidity and heavy metals; t is time.

[0022] Preferably, the data collection and transmission expressions in the monitoring module are:

[0023] D(t)=d1(t),d2(t),…,d n (t)

[0024] Where: D(t) is the data set collected from all sensors at time t; d1(t), d2(t), ..., d n (t) is the data value of each sensor at time t;

[0025] The real-time data processing expression is:

[0026]

[0027] in: is the moving average at time t; W is the size of the sliding window, that is, the number of time points used to calculate the average; d(tk) is the sensor data value at time tk, where k ranges from 0 to W-1.

[0028] Preferably, the data processing module uses a sliding window algorithm to process data as follows:

[0029]

[0030] in: is the sliding window average of the i-th sensor at time t; W is the size of the sliding window, that is, the number of time points used to calculate the average; d i (tk) is the measurement value of the i-th sensor at time tk; k ranges from 0 to W-1, representing the data points in the sliding window.

[0031] Preferably, the real-time data processing process of the data processing module is expressed as:

[0032]

[0033] in: is the sliding window average set of all sensor data at time t; Sliding window average of the ith sensor.

[0034] Preferably, the expression for detecting abnormal data by the support vector machine in the data analysis module is:

[0035] f(x)=sign(w T x+b)

[0036] Where: x is the input feature vector; w is the weight vector; b is the bias; f(x) is the classification result, +1 indicates normal, and 1 indicates abnormal.

[0037] Preferably, the decision tree algorithm in the data analysis module is used for pollution pattern recognition and early warning expression:

[0038]

[0039] p j is the value of the jth water quality parameter; θ j is the threshold of the jth water quality parameter; Class1 and Class2 are the class labels in the decision tree;

[0040] The expression for marking abnormal regions is:

[0041]

[0042] Among them: the abnormal area is the set of areas marked as abnormal; p j is the value of the jth water quality parameter; θ j is the threshold value of the jth water quality parameter; To indicate that there exists a j such that p j Exceeding the threshold value θ j .

[0043] Preferably, the spatial distribution of the abnormal data points and the contaminated area boundary expression are:

[0044]

[0045] Where: x i is the i-th data point; C j is the jth cluster; μ j is the center of the jth cluster; |x i μ j | 2 is the square distance from the i-th data point to the j-th cluster center;

[0046] The cluster allocation is:

[0047]

[0048] c i is the cluster label of the i-th data point;

[0049] To find the index of the cluster center that minimizes the distance;

[0050] Update the cluster centers to:

[0051]

[0052] |C j | is the number of samples in the jth cluster; μ j is the new center of the jth cluster.

[0053] Preferably, the abnormal data point detection expression in the source positioning module is:

[0054] Exception = {x i ∣|x i μ ci |>Threshold}

[0055] Where: threshold is the distance threshold defined as anomaly; μ ci is the center of the cluster to which the i-th data point belongs;

[0056] The boundary expression of the contaminated area is:

[0057]

[0058] Where: R j is the boundary radius of the jth cluster; is the maximum distance from a data point in a cluster to the cluster center.

[0059] Preferably, in the response module, p is the monitored pollution parameter, θ is the pollution threshold, and if the pollution parameter p exceeds the threshold θ, the system will detect pollution: p>θ.

[0060] Compared with the prior art, the present invention provides a sudden environmental pollution monitoring system, which has the following beneficial effects:

[0061] 1. The present invention monitors water quality changes in real time by laying water quality sensors: dissolved oxygen, pH value, turbidity and heavy metal sensors in the target water body, ensuring the continuity and timeliness of data. The sensors gather data to the data processing module through wireless or satellite communication, realizing the high efficiency of remote monitoring and data transmission. The sliding window algorithm is used to process the collected data in real time, effectively smooth the noise data, and reflect the water quality changes in time, so that the system can quickly respond to sudden pollution events. The support vector machine and decision tree algorithm are used to conduct in-depth analysis of the processed data, and abnormal data that does not meet the normal range is detected in time. The pollution pattern can be quickly identified to provide support for timely warning, achieving the beneficial effect of continuously monitoring the target water body and quickly determining the sudden pollution area of ​​the water body.

[0062] 2. The present invention analyzes the spatial distribution of abnormal data points by using the K-means clustering algorithm to identify the concentrated areas of the polluted area. This method determines the specific location of the pollution source by clustering the data points into clusters, and draws the boundaries of the polluted area. By calculating the distance from the abnormal data points to the cluster center, the system can identify the specific scope of the polluted area. The detection of abnormal data points and the division of clusters help determine the specific location of the pollution source. By using the K-means clustering algorithm and calculating the maximum distance from the data points in the cluster to the cluster center, the system can determine the boundaries of the polluted area, which helps to define the specific area of ​​the pollution source, and achieves the beneficial effect of determining the specific pollution source when sudden water environmental pollution is monitored. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a schematic diagram of the system of the present invention; DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] See also Figure 1 ,A sudden environmental pollution monitoring system ,the detection system is composed of a monitoring module, a data processing module, a data analysis module, a pollution identification module, a source ,location module and a response module;

[0066] The monitoring module is used to monitor the regional target water body. The regional target water body is mapped and zoned by satellite. Water quality sensors are laid in the regional target water body. The monitoring module also includes a data collection unit. The data collection unit sensor data is gathered by wireless or satellite communication and the data is transmitted to the data processing module.

[0067] The data processing module is used to receive the regional target water body data collected by the data collection unit and perform real-time data processing on it: using a sliding window algorithm to process and analyze sensor data;

[0068] The data analysis module uses a type of support vector machine to detect data that does not conform to the normal range; and uses a decision tree algorithm for pollution pattern recognition and early warning, setting thresholds for water quality parameters. When the data in a certain area exceeds the normal range, it is marked as an abnormal area:

[0069] The source location module uses the K-means clustering algorithm to identify the concentrated areas of the polluted area, analyze the spatial distribution of abnormal data points, and determine the boundaries of the polluted area:

[0070] The response module is used to automatically initiate an alarm when contamination is detected:

[0071] The water quality sensors in the monitoring module are: dissolved oxygen sensor, pH sensor, turbidity sensor and heavy metal sensor.

[0072] By laying water quality sensors (such as dissolved oxygen, pH, turbidity and heavy metal sensors) in the target water body, water quality changes are monitored in real time. This real-time monitoring ensures the continuity and timeliness of data.

[0073] The sensors gather data to the data processing module through wireless or satellite communication, achieving high efficiency of remote monitoring and data transmission.

[0074] The collected data is processed in real time using a sliding window algorithm, which can effectively smooth out noisy data and reflect water quality changes in a timely manner, enabling the system to quickly respond to sudden pollution events.

[0075] Using support vector machines and decision tree algorithms, the processed data is deeply analyzed to detect abnormal data that does not conform to the normal range in a timely manner. This analysis mechanism can quickly identify pollution patterns and provide support for timely warnings.

[0076] The K-means clustering algorithm is used to identify concentrated areas of contaminated areas and analyze the spatial distribution of abnormal data points. This function helps determine the source of contamination and quickly locate the affected areas, so that countermeasures can be implemented more effectively.

[0077] Once pollution is detected, the module automatically activates the alarm and quickly issues an early warning to relevant personnel. This automated response mechanism ensures that measures can be taken quickly to reduce damage to the environment when pollution occurs.

[0078] Decision tree algorithm and threshold setting: Setting thresholds for water quality parameters helps to quickly mark abnormal areas. This decision support based on data analysis enables the system to respond in the shortest possible time.

[0079] Specifically, the calculation expression of the regional target water body in the monitoring module is:

[0080] Regional target water body zoning:

[0081] R = {(x,y)|x min ≤x≤x max ,y min ≤y≤y max}

[0082] Where: R is the target water area, which defines a rectangular area; (x, y) is the coordinate of a point in the area; x min and x maxis the minimum and maximum coordinate value of the region on the x-axis; y min and max are the minimum and maximum coordinate values ​​of the region on the y-axis;

[0083] The water quality sensor data expression in the monitoring module is:

[0084]

[0085] Where: d i(t) is the data value of the i-th sensor at time t; f i is the sensor measurement function, which varies according to the sensor; p i are the water quality parameters monitored by sensor i: dissolved oxygen, pH value, turbidity and heavy metals; t is time.

[0086] Specifically, the data collection and transmission expressions in the monitoring module are:

[0087] D(t)=d1(t),d2(t),…,d n (t)

[0088] Where: D(t) is the data set collected from all sensors at time t; d1(t), d2(t), ..., d n (t) is the data value of each sensor at time t;

[0089] The real-time data processing expression is:

[0090]

[0091] in: is the moving average at time t; W is the size of the sliding window, that is, the number of time points used to calculate the average; d(tk) is the sensor data value at time tk, where k ranges from 0 to W-1.

[0092] Specifically, the data processing expression using the sliding window algorithm in the data processing module is:

[0093]

[0094] in: is the sliding window average of the i-th sensor at time t; W is the size of the sliding window, that is, the number of time points used to calculate the average; d i (tk) is the measurement value of the i-th sensor at time tk; k ranges from 0 to W-1, representing the data points in the sliding window.

[0095] Specifically, the real-time data processing process of the data processing module is expressed as:

[0096]

[0097] in: is the sliding window average set of all sensor data at time t; Sliding window average of the ith sensor.

[0098] Specifically, the expression for detecting abnormal data by the support vector machine in the data analysis module is:

[0099] f(x)=sign(w T x+b)

[0100] Where: x is the input feature vector; w is the weight vector; b is the bias; f(x) is the classification result, +1 indicates normal, and 1 indicates abnormal.

[0101] Specifically, the decision tree algorithm in the data analysis module is used for pollution pattern recognition and early warning expression:

[0102]

[0103] p j is the value of the jth water quality parameter; θ j is the threshold of the jth water quality parameter; Class1 and Class2 are the class labels in the decision tree;

[0104] The expression for marking abnormal areas is:

[0105]

[0106] Among them: the abnormal area is the set of areas marked as abnormal; p j is the value of the jth water quality parameter; θ j is the threshold value of the jth water quality parameter; To indicate that there exists a j such that p j Exceeding the threshold value θ j .

[0107] Specifically, the spatial distribution of abnormal data points and the boundary expression of the contaminated area are:

[0108]

[0109] Where: x i is the i-th data point; C j is the jth cluster; μ j is the center of the jth cluster; |x i μ j | 2 is the square distance from the i-th data point to the j-th cluster center;

[0110] The cluster allocation is:

[0111]

[0112] c i is the cluster label of the i-th data point;

[0113] To find the index of the cluster center that minimizes the distance;

[0114] Update the cluster centers to:

[0115]

[0116] |C j | is the number of samples in the jth cluster; μ j is the new center of the jth cluster.

[0117] Specifically, the expression for detecting abnormal data points in the source location module is:

[0118]

[0119] Where: threshold is the distance threshold defined as anomaly; is the center of the cluster to which the i-th data point belongs;

[0120] The boundary expression of the contaminated area is:

[0121]

[0122] Where: R j is the boundary radius of the jth cluster; is the maximum distance from a data point in a cluster to the cluster center.

[0123] Specifically, in the response module, p is the monitored pollution parameter, θ is the pollution threshold, and if the pollution parameter p exceeds the threshold θ, the system will detect pollution: p>θ.

[0124] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and alterations may be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A sudden environmental pollution monitoring system, characterized in that: The detection system is composed of a monitoring module, a data processing module, a data analysis module, a pollution identification module, a source location module and a response module; The monitoring module is used to monitor the regional target water body, the regional target water body is mapped and zoned by satellite, water quality sensors are laid in the regional target water body, and the monitoring module also includes a data collection unit, the data collection unit sensor data is gathered by wireless or satellite communication, and the data is transmitted to the data processing module; The data processing module is used to receive the regional target water body data collected by the data collection unit and perform real-time data processing on it: using a sliding window algorithm to process and analyze sensor data; The data analysis module uses a type of support vector machine to detect data that does not conform to the normal range; and uses a decision tree algorithm for pollution pattern recognition and early warning, setting thresholds for water quality parameters. When the data in a certain area exceeds the normal range, it is marked as an abnormal area: The source location module uses the K-means clustering algorithm to identify the concentrated areas of the polluted area, analyze the spatial distribution of abnormal data points, and determine the boundaries of the polluted area: The response module is used to automatically activate an alarm when contamination is detected: The water quality sensors in the monitoring module are: dissolved oxygen sensor, pH value sensor, turbidity sensor and heavy metal sensor.

2. The sudden environmental pollution monitoring system according to claim 1 is characterized in that: The calculation expression of the regional target water body in the monitoring module is: Regional target water body zoning: R={(x,y)∣x min ≤x≤x max ,and min ≤y≤y max } Where: R is the target water area, which defines a rectangular area; (x, y) is the coordinate of a point in the area; x min and x max is the minimum and maximum coordinate value of the region on the x-axis; y min and max are the minimum and maximum coordinate values ​​of the region on the y-axis; The water quality sensor data expression in the monitoring module is: d i(t) =f i(pi,t) Where: d i(t) is the data value of the i-th sensor at time t; f i is the sensor measurement function, which varies according to the sensor; p i are the water quality parameters monitored by sensor i: dissolved oxygen, pH value, turbidity and heavy metals; t is time.

3. A sudden environmental pollution monitoring system according to claim 2, characterized in that: The data collection and transmission expressions in the monitoring module are as follows: D(t)=d1(t),d2(t),…,d n (t) Where: D(t) is the data set collected from all sensors at time t; d1(t), d2(t), ..., d n (t) is the data value of each sensor at time t; The real-time data processing expression is: in: is the moving average at time t; W is the size of the sliding window, that is, the number of time points used to calculate the average; d(tk) is the sensor data value at time tk, where k ranges from 0 to W-1.

4. The sudden environmental pollution monitoring system according to claim 1 is characterized in that: The data processing module uses a sliding window algorithm to process data. The expression is: in: is the sliding window average of the i-th sensor at time t; W is the size of the sliding window, that is, the number of time points used to calculate the average; d i (tk) is the measurement value of the i-th sensor at time tk; k ranges from 0 to W-1, representing the data points in the sliding window.

5. The sudden environmental pollution monitoring system according to claim 3 is characterized in that: The real-time data processing process of the data processing module is expressed as: in: is the sliding window average set of all sensor data at time t; Sliding window average of the ith sensor.

6. A sudden environmental pollution monitoring system according to claim 5, characterized in that: The expression for abnormal data detection by the support vector machine in the data analysis module is: f(x)=sign(w T x+b) Where: x is the input feature vector; w is the weight vector; b is the bias; f(x) is the classification result, +1 indicates normal, and 1 indicates abnormal.

7. A sudden environmental pollution monitoring system according to claim 6, characterized in that: The decision tree algorithm in the data analysis module is used for pollution pattern recognition and early warning expression: p j is the value of the jth water quality parameter; θ j is the threshold of the jth water quality parameter; Class1 and Class2 are the class labels in the decision tree; The expression for marking abnormal regions is: Among them: the abnormal area is the set of areas marked as abnormal; p j is the value of the jth water quality parameter; θ j is the threshold value of the jth water quality parameter; To indicate that there exists a j such that p j Exceeding the threshold value θ j .

8. The system for monitoring sudden environmental pollution according to claim 7, characterized in that: The spatial distribution of the abnormal data points and the boundary expression of the contaminated area are: Where: x i is the i-th data point; C j is the jth cluster; μ j is the center of the jth cluster; |x i μ j | 2 is the square distance from the i-th data point to the j-th cluster center; The cluster allocation is: c i is the cluster label of the i-th data point; To find the index of the cluster center that minimizes the distance; Update the cluster centers to: |C j | is the number of samples in the jth cluster; μ j is the new center of the jth cluster.

9. A sudden environmental pollution monitoring system according to claim 8, characterized in that: The abnormal data point detection expression in the source positioning module is: Where: threshold is the distance threshold defined as anomaly; is the center of the cluster to which the i-th data point belongs; The boundary expression of the contaminated area is: Where: R j is the boundary radius of the jth cluster; is the maximum distance from a data point in a cluster to the cluster center.

10. A sudden environmental pollution monitoring system according to claim 9, characterized in that: In the response module, p is the monitored pollution parameter, and θ is the pollution threshold. If the pollution parameter p exceeds the threshold θ, the system will detect pollution: p>θ.

Images