Intelligent distributed stockpile pollution source monitoring system and method

Abstract

The invention provides an intelligent distributed stockpile pollution source monitoring system and method.The intelligent distributed stockpile pollution source monitoring system comprises a plurality of stockpile pollution auxiliary monitoring devices, a stockpile pollution main monitoring device, a 5G signal transmission terminal and a monitoring server, and the plurality of stockpile pollution auxiliary monitoring devices are in communication connection with the stockpile pollution main monitoring device through Bluetooth; the stockpile pollution main monitoring device is connected with the 5G signal transmission terminal through 5G communication, and the 5G signal transmission terminal is connected with the monitoring server through 5G communication. According to the invention, a plurality of auxiliary stockyard pollution monitoring devices work cooperatively and are combined with a plurality of sensors, so that all-around and multi-parameter real-time monitoring of the environment of the stockyard is realized, pollution discharge of the stockyard can be effectively controlled, pollution to soil and underground water caused by the fact that sewage permeates underground is reduced, pollution to air caused by raised dust is reduced, and the surrounding ecological environment is protected.

Description

Technical Field

[0001] The present invention relates to the technical field of pollution source monitoring, and in particular to an intelligent distributed stockpile pollution source monitoring system and method. Background Art

[0002] With the development of various industrial and agricultural productions, the number of stockpiles (such as organic fertilizer stockpiles, ore stockpiles, construction waste stockpiles, etc.) is increasing continuously. During the process of material stacking in many stockpiles, due to poor management, there are many environmental problems. For example, when the materials are stacked outdoors and it rains, the sewage generated by rainwater scouring the stockpiles may directly seep into the ground, polluting the soil and groundwater and affecting the surrounding ecological environment and the safety of residents' water use. If it is powdery stockpiles, such as coal and minerals, when the materials are stacked outdoors and there is a strong wind, it will cause dust in the surrounding area and pollute the air at the same time. At present, stockpiles generally lack effective monitoring measures and it is difficult to give timely warnings and handle pollution risks. Therefore, it is of great practical significance to develop an efficient and intelligent stockpile pollution source monitoring system. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide an intelligent distributed stockpile pollution source monitoring system and method, which can effectively control the pollution emissions of stockpiles, reduce the pollution of soil and groundwater caused by sewage seeping into the ground, reduce the pollution of air by dust, and protect the surrounding ecological environment through real-time and accurate pollution monitoring and early warning.

[0004] To solve the above technical problem, an embodiment of the present invention provides the following technical solution: An intelligent distributed stockpile pollution source monitoring system includes a plurality of stockpile pollution sub-monitoring devices, a stockpile pollution main monitoring device, a 5G signal transmission terminal, and a monitoring server. The plurality of stockpile pollution sub-monitoring devices are communicatively connected to the stockpile pollution main monitoring device via Bluetooth, the stockpile pollution main monitoring device is communicatively connected to the 5G signal transmission terminal via 5G, and the 5G signal transmission terminal is communicatively connected to the monitoring server via 5G;

[0005] The stockpile pollution sub-monitoring device includes a battery module and a first processor, a rainfall sensor, a humidity sensor, a water flow sensor, an acoustic-optic alarm, a wind sensor, a Beidou positioning module, a dust sensor, and a first Bluetooth communication module that are electrically connected to the battery module. The first processor is communicatively connected to the rainfall sensor, the humidity sensor, the water flow sensor, the acoustic-optic alarm, the wind sensor, the Beidou positioning module, the dust sensor, and the first Bluetooth communication module by wire;

[0006] The stockpile pollution main monitoring device includes a power supply module and a second processor, a second Bluetooth communication module, and a 5G communication module that are electrically connected to the power supply module. The second processor is communicatively connected to the second Bluetooth communication module and the 5G communication module by wire;

[0007] The bottom and side of the housing of the auxiliary monitoring device for stockpile pollution are provided with a densely distributed water seepage layer, and the pore diameter of the water seepage layer is smaller than the pore diameter of the stockpile materials.

[0008] Preferably, the rainfall sensor is located at the top of the housing of the main monitoring device for stockpile pollution, the humidity sensor is located on the inner side wall of the middle part of the main monitoring device for stockpile pollution, and the water flow sensor is located at the lower inner part of the main monitoring device for stockpile pollution.

[0009] Preferably, the water flow sensor includes a water flow sensing probe, a funnel-shaped water collecting surface, and a water collecting pipe. The edge of the funnel-shaped water collecting surface is connected to the inner side of the housing of the main monitoring device for stockpile pollution. The bottom of the funnel-shaped water collecting surface is communicated with the water collecting pipe, and the water flow sensing probe is arranged on the inner wall of the water collecting pipe.

[0010] Preferably, the sound and light alarm is located at the top of the housing of the auxiliary monitoring device for stockpile pollution, and includes a sound alarm and a light alarm. The light alarm includes yellow, orange, and red LED lights.

[0011] Preferably, the wind sensor and the dust sensor are located on the outer side of the housing of the auxiliary monitoring device for stockpile pollution. The dust sensor includes a parabolic dust converging surface, a suction fan blade, a rotating shaft, a dust gathering cavity, a motor, and a dust sensing probe. The motor is installed in the dust gathering cavity. The rotating shaft of the motor is mechanically connected to one end of the rotating shaft. The other end of the rotating shaft is connected to the suction fan blade. The dust sensing probe is arranged on the inner wall of the dust gathering cavity. The connection point between the rotating shaft and the suction fan blade is located at the focal point of the parabolic dust converging surface.

[0012] Preferably, the battery module, the first processor, the Beidou positioning module, and the first Bluetooth communication module are arranged on a control board inside a control board housing, and the control board housing is located at the top of the housing of the auxiliary monitoring device for stockpile pollution.

[0013] The present invention also provides a working method of the intelligent distributed stockpile pollution source monitoring system as described above, including the following steps:

[0014] The rainfall sensors, humidity sensors, and water flow sensors of multiple auxiliary monitoring devices for stockpile pollution detect the sensing data in the stockpile in real time. Among them: when the first processor perceives the existence of precipitation through the data sent by the rainfall sensor, the yellow light of the sound and light alarm flashes, and a sound warning is given to prevent sewage discharge; when the first processor perceives that the humidity of the stockpile is 100% through the data sent by the humidity sensor, the orange light of the sound and light alarm flashes, and a sound warning is given that the stockpile will discharge sewage; when the first processor perceives the sewage flow through the data sent by the water flow sensor, the red light of the sound and light alarm flashes, and an alarm sound is sounded.

[0015] Meanwhile, the wind sensor monitors the wind data at the stockpiling site. When the first processor monitors the wind data through the wind sensor, the first processor controls the dust sensor to collect the dust data around the stockpile. When dust is detected, the red light of the audible and visual alarm flashes and an alarm sound is sounded; the Beidou positioning module monitors the displacement change of the stockpile pollution secondary monitoring device to monitor whether it has been removed;

[0016] The first processor wirelessly sends the monitored stockpile sewage and dust sensor data to the second Bluetooth communication module of the stockpile pollution main monitoring device through the first Bluetooth communication module. The second Bluetooth communication module transmits the received sensor data to the second processor. The second processor processes the data sent by Bluetooth communication and stores it in the storage module, and wirelessly sends the data to the 5G signal transmission terminal through the 5G communication module for 5G communication. Finally, the monitoring server obtains the pollution monitoring data of the stockpile.

[0017] Preferably, the first processor sets a dynamic threshold model according to historical precipitation data, the variation law of stockpile humidity, and sewage discharge conditions, specifically:

[0018] Let x 1 、x 2 , ……, x n be input features. The input features at least include historical precipitation data, historical stockpile humidity data, historical sewage discharge data, and historical dust data, where x i represents the i-th feature, y is the corresponding threshold, and the prediction formula of the decision tree regression model is:

[0019]

[0020] where J is the number of leaf nodes of the decision tree, R j is the feature space region corresponding to the j-th leaf node, c j is the predicted value of leaf node j, I(x∈R j ) is an exponential function, which is 1 when x belongs to R j and 0 otherwise;

[0021] When training the model, the predicted value c j

[0022]

[0023] of the leaf node is determined by the minimum mean square error, where N j is the number of samples belonging to leaf node j, and y j is the true threshold of the corresponding sample;

[0024] The overall objective function of the model is

[0025]

[0026] Among them, N is the number of training samples, is the prediction threshold of the model for the i-th sample. During real-time threshold adjustment, the feature data collected in real time is input into the trained model to obtain the dynamic threshold y d .

[0027] Preferably, after the main monitoring device for stockpile pollution processes the sensing data monitored by multiple secondary monitoring devices for stockpile pollution through a fusion algorithm, it is wirelessly transmitted to the 5G signal transmission terminal, and finally the monitoring server obtains the pollution monitoring data of the stockpile. Specifically: for each monitoring parameter, corresponding weights are allocated according to the historical data stability of the sensors of the secondary monitoring devices. Suppose there are n secondary monitoring devices for stockpile pollution. For a certain monitoring parameter P, the measured value of the i-th device is P i , and its corresponding weight is ω i , then the fused measured value P f is:

[0028] Preferably, the monitoring server predicts the pollution trend of the stockpile yard in the future period based on multiple linear regression and exponential smoothing method, where:

[0029] Taking the fused measured value as the independent variable and the pollution index as the dependent variable, the fused measured value includes stockpile humidity, precipitation intensity, sewage discharge situation, air temperature, wind direction, and wind speed;

[0030] Establish a multiple linear regression model: Y = β 0 + β 1 X 1 + β 2 X 2 + ··· + β n X n + ε, where Y is the predicted value of the pollution index, β 0 is the constant term, β i is the regression coefficient corresponding to each factor, X i is the fused measured value, and ε is the error term;

[0031] Estimate the regression coefficient by the least squares method to minimize the sum of the squares of the errors between the predicted value and the actual value;

[0032] Perform exponential smoothing on the preliminary prediction result obtained from the multiple linear regression model. Let F t be the predicted value at time t, Y t be the actual value at time t, and α be the smoothing coefficient, 0 < α < 1;

[0033] The exponential smoothing formula is: F t+1 = αY t+(1 - α)F t , through continuous iteration, dynamically adjust the prediction result according to the new actual data;

[0034] Compare the smoothed prediction result F t+1 with the corresponding dynamic threshold y d and classify the early warning mechanism according to the comparison result. The classification formula is:

[0035]

[0036] When the classification is low pollution, initiate a yellow early warning, and the monitoring server sends an alarm rectification notice to the on-site staff;

[0037] When the classification is medium pollution, initiate an orange early warning, the monitoring server sends an alarm rectification notice to the on-site staff, and automatically sends pollution alarm information to the environmental protection department;

[0038] When the classification is high pollution, initiate a red early warning, the monitoring server sends an alarm notice to the on-site staff, and automatically sends information requesting law enforcement action to the environmental protection department.

[0039] The beneficial effects of the above technical solutions of the present invention are as follows:

[0040] Multiple stockpile pollution sub-monitoring devices of the present invention work together, combined with a variety of sensors, to achieve all-round and multi-parameter real-time monitoring of the stockpile yard environment. The rainfall sensor can timely sense precipitation, providing a basis for preventing sewage generation; the humidity sensor can accurately master the change of stockpile humidity, and early warn of possible sewage discharge risks; the dust sensor can effectively capture the dust situation around the stockpile, and timely discover potential air pollution hazards in windy weather. The data of each sensor complement each other, and can more comprehensively and accurately reflect the actual pollution situation of the stockpile yard, improving the monitoring accuracy. Through real-time and accurate pollution monitoring and early warning, it is possible to effectively control the pollution emissions of the stockpile yard, reduce the pollution of sewage seeping into the ground to the soil and groundwater, reduce the pollution of dust to the air, and protect the surrounding ecological environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is the principle block diagram of the intelligent distributed stockpile pollution source monitoring system of the present invention;

[0042] Figure 2 is the principle block diagram of the stockpile pollution sub-monitoring device of the intelligent distributed stockpile pollution source monitoring system of the present invention;

[0043] Figure 3 is the principle block diagram of the stockpile pollution main-monitoring device of the intelligent distributed stockpile pollution source monitoring system of the present invention;

[0044] Figure 4Schematic structural diagram of the intelligent distributed stockpile pollution source monitoring system of the present invention placed for pollution monitoring of the stockpile;

[0045] Figure 5 Schematic diagram of the dust monitoring principle of the intelligent distributed stockpile pollution source monitoring system of the present invention. Specific implementation manners

[0046] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0047] As Figure 1 shown, the present invention proposes an intelligent distributed stockpile pollution source monitoring system, including a plurality of stockpile pollution sub-monitoring devices 1, a stockpile pollution main monitoring device 2, a 5G signal transmission terminal 3, and a monitoring server 4. The plurality of stockpile pollution sub-monitoring devices 1 are communicatively connected to the stockpile pollution main monitoring device 2 via Bluetooth. The stockpile pollution main monitoring device 2 is communicatively connected to the 5G signal transmission terminal 3 via 5G. The 5G signal transmission terminal 3 is communicatively connected to the monitoring server 4 via 5G;

[0048] As Figure 2 shown, the stockpile pollution sub-monitoring device 1 includes a battery module 11 and a first processor 12, a rainfall sensor 13, a humidity sensor 14, a water flow sensor 15, an acoustic-optic alarm 16, a wind sensor 17, a Beidou positioning module 18, a dust sensor 19, and a first Bluetooth communication module 110 that are electrically connected to the battery module 11. The first processor 12 is in wired communication connection with the rainfall sensor 13, the humidity sensor 14, the water flow sensor 15, the acoustic-optic alarm 16, the wind sensor 17, the Beidou positioning module 18, the dust sensor 19, and the first Bluetooth communication module 110;

[0049] As Figure 3 shown, the stockpile pollution main monitoring device 2 includes a power supply module 21 and a second processor 22, a second Bluetooth communication module 23, and a 5G communication module 24 that are electrically connected to the power supply module 21. The second processor 22 is in wired communication connection with the second Bluetooth communication module 23 and the 5G communication module 24;

[0050] As Figure 4 shown, the battery module 11, the first processor 12, the Beidou positioning module 18, and the first Bluetooth communication module 110 are arranged on a control board within a control board housing 112, and the control board housing 112 is located at the top of the housing of the stockpile pollution sub-monitoring device 1.

[0051] The bottom and side of the housing of the stockpile pollution sub-monitoring device 1 are provided with a dense water seepage layer, and the pore diameter of the water seepage layer is smaller than the pore diameter of the stockpile materials.

[0052] The rainfall sensor 13 is located on the top of the housing of the main monitoring device for stockpile pollution 2 , the humidity sensor 14 is located on the inner side wall of the middle part of the main monitoring device for stockpile pollution 2 , and the water flow sensor 15 is located on the lower part of the main monitoring device for stockpile pollution 2 .

[0053] The water flow sensor 15 includes a water flow sensing probe 151, a funnel-shaped water collecting surface 152, and a water collecting pipe 153. The edge of the funnel-shaped water collecting surface 152 is connected to the inner side of the shell of the main monitoring device 2 for stockpile pollution, the bottom of the funnel-shaped water collecting surface 152 is connected to the water collecting pipe 153, and the water flow sensing probe 151 is arranged on the inner wall of the water collecting pipe 153.

[0054] The sound and light alarm 16 is located on the top of the housing of the stockpile pollution auxiliary monitoring device 1, and includes a sound alarm and a light alarm. The light alarm includes yellow, orange and red LED lights.

[0055] like Figure 5 As shown, the wind sensor 17 and the dust sensor 19 are located on the outer side of the shell of the stockpile pollution auxiliary monitoring device 1. The dust sensor 19 includes a parabolic dust gathering surface 191, exhaust blades 192, a rotating shaft 193, a dust gathering chamber 194, a motor 195, and a dust sensing probe 196. The motor 195 is installed in the dust gathering chamber 194. The rotating shaft of the motor 195 is mechanically connected to one end of the rotating shaft 193, and the other end of the rotating shaft 193 is connected to the exhaust blades 192. The dust sensing probe 196 is arranged on the inner wall of the dust gathering chamber 194, and the connection point between the rotating shaft 193 and the exhaust blades 192 is located at the focal position of the parabolic dust gathering surface 191.

[0056] The working method of the above-mentioned intelligent distributed stockpile pollution source monitoring system includes the following steps:

[0057] The rainfall sensors 13, humidity sensors 14, and water flow sensors 15 of the multiple stockpile pollution auxiliary monitoring devices 1 detect the sensor data in the stockpile in real time, wherein: when the first processor 12 senses the presence of precipitation through the data sent by the rainfall sensor 13, the yellow light of the sound and light alarm 16 flashes, and a sound warning is given to prevent sewage discharge; when the first processor 12 senses the stockpile humidity is 100% through the data sent by the humidity sensor 14, the orange light of the sound and light alarm 16 flashes, and a sound warning is given that the stockpile will discharge sewage; when the first processor 12 senses the sewage flow through the data sent by the water flow sensor 15, the red light of the sound and light alarm 16 flashes, and an alarm sounds;

[0058] At the same time, the wind sensor 17 monitors the wind data at the stockpile site. When the first processor 12 monitors the wind data through the wind sensor 17, the first processor 12 controls the dust sensor 19 to collect dust data around the stockpile. When dust is sensed, the red light of the sound and light alarm 16 flashes and an alarm sounds; the Beidou positioning module 18 monitors the displacement change of the stockpile pollution auxiliary monitoring device 1 to monitor whether it has been removed;

[0059] The first processor 12 wirelessly sends the monitored stockpile sewage and dust sensing data to the second Bluetooth communication module 23 of the stockpile pollution main monitoring device 2 through the first Bluetooth communication module 110. The second Bluetooth communication module 23 transmits the received sensing data to the second processor 22. The second processor 22 processes the data sent via Bluetooth and stores it in the storage module 25, and wirelessly sends the data to the 5G signal transmission terminal 3 via 5G communication through the 5G communication module 24. Finally, the monitoring server 4 obtains the pollution monitoring data of the stockpile.

[0060] In this embodiment, the first processor 12 sets a dynamic threshold model according to historical precipitation data, the variation law of stockpile humidity, and sewage discharge conditions, specifically:

[0061] Let x 1 、x 2 , ……, x n be input features. The input features at least include historical precipitation data, historical stockpile humidity data, historical sewage discharge data, and historical dust data, where x i represents the i-th feature, y is the corresponding threshold, and the prediction formula of the decision tree regression model is:

[0062]

[0063] where J is the number of leaf nodes of the decision tree, R j is the feature space region corresponding to the j-th leaf node, c j is the predicted value of leaf node j, I(x∈R j ) is an exponential function, which is 1 when x belongs to R j and 0 otherwise;

[0064] When training the model, the predicted value c j

[0065]

[0066] of the leaf node is determined by the minimum mean square error. j where N j is the number of samples belonging to leaf node j, and y j is the true threshold of the corresponding sample;

[0067] The overall objective function of the model is

[0068]

[0069] where N is the number of training samples, is the predicted threshold of the model for the i-th sample. When adjusting the real-time threshold, the feature data collected in real time is brought into the trained model to obtain the dynamic threshold y d. Precipitation may cause a sudden increase in the sewage volume. By constructing a model to set dynamic thresholds by combining precipitation and sewage discharge data, it is possible to predict in advance whether the sewage volume will exceed the carrying capacity of the treatment facilities, avoiding the problem that setting a fixed threshold cannot judge potential hazards according to the actual situation on site. When approaching or reaching the threshold, timely warnings can be issued to avoid pollution of the surrounding environment caused by sewage overflow, such as polluting surface water and soil, and endangering the ecosystem and human health.

[0070] In this embodiment, after the main monitoring device for stockpile pollution processes the sensing data monitored by multiple secondary monitoring devices for stockpile pollution through a fusion algorithm, it is wirelessly transmitted to the 5G signal transmission terminal, and finally the monitoring server obtains the pollution monitoring data of the stockpile. Specifically: for each monitoring parameter, corresponding weights are allocated according to the historical data stability of the sensors of the secondary monitoring devices. Suppose there are n secondary monitoring devices for stockpile pollution. For a certain monitoring parameter P, the measured value of the i-th device is P i , and its corresponding weight is ω i , then the fused measured value P f is:

[0071] In this embodiment, the monitoring server predicts the pollution trend of the stockpile yard in the future for a period of time based on multiple linear regression and exponential smoothing method, where:

[0072] Taking the fused measured value as the independent variable and the pollution index as the dependent variable, the fused measured value includes stockpile humidity, precipitation intensity, sewage discharge situation, air temperature, wind direction, and wind speed;

[0073] Establish a multiple linear regression model: Y = β 0 + β 1 X 1 + β 2 X 2 + ··· + β n X n + ε, where Y is the predicted value of the pollution index, β 0 is the constant term, β i is the regression coefficient corresponding to each factor, X i is the fused measured value, and ε is the error term;

[0074] Estimate the regression coefficients by the least squares method to minimize the sum of the squares of the errors between the predicted value and the actual value;

[0075] Perform exponential smoothing on the preliminary prediction results obtained from the multiple linear regression model. Suppose F t is the predicted value at time t, Y t is the actual value at time t, and α is the smoothing coefficient, 0 < α < 1;

[0076] The exponential smoothing formula is: Ft+1 = αY t + (1 - α)F t , through continuous iteration, dynamically adjust the prediction results according to the new actual data;

[0077] Compare the smoothed prediction result F t+1 with the corresponding dynamic threshold y d and classify the early warning mechanism according to the comparison result. The classification formula is:

[0078]

[0079] When the classification is low pollution, start the yellow early warning, and the monitoring server sends a warning and rectification notice to the on-site staff;

[0080] When the classification is medium pollution, start the orange early warning, the monitoring server sends a warning and rectification notice to the on-site staff, and automatically sends pollution warning information to the environmental protection department;

[0081] When the classification is high pollution, start the red early warning, the monitoring server sends a warning notice to the on-site staff, and automatically sends information requesting law enforcement to the environmental protection department.

[0082] Multiple stockpile pollution sub-monitoring devices of the present invention work together, combined with a variety of sensors (such as rainfall, humidity, water flow, dust, wind sensors, etc.), to achieve all-round and multi-parameter real-time monitoring of the stockpile yard environment. For example, the rainfall sensor can timely sense precipitation to provide a basis for preventing sewage generation; the humidity sensor can accurately master the humidity change of the stockpile to early warn of possible sewage discharge risks; the dust sensor can effectively capture the dust situation around the stockpile to timely detect potential air pollution hazards in windy weather. The data of each sensor complement each other, can more comprehensively and accurately reflect the actual pollution situation of the stockpile yard, and improve the monitoring accuracy.

[0083] The sound and light alarm emits intuitive warning signals according to different pollution risk levels (such as yellow light early warning during precipitation, orange light early warning when the stockpile humidity reaches 100%, red light early warning when sewage discharge or dust exceeds the standard). The sound warning can also timely attract the attention of on-site personnel, ensure that corresponding measures are taken at the first time when pollution risks appear, effectively avoid the occurrence or expansion of pollution accidents, and protect the surrounding ecological environment and the safety of residents' lives.

[0084] The main monitoring device and the secondary monitoring device are connected through Bluetooth communication, which has the characteristics of low power consumption and stable short-distance communication. It is suitable for deployment in the complex environment of the stockyard to ensure the stability of data transmission. The main monitoring device then quickly transmits the data to the 5G signal transmission terminal through the 5G communication module. By utilizing the advantages of high speed and low latency of 5G, it ensures that the monitoring server can obtain the latest pollution monitoring data in real time, realizes the efficient transmission and timely processing of data, and provides strong support for subsequent analysis and decision-making.

[0085] The monitoring server predicts the pollution trend based on multiple linear regression and exponential smoothing method, fully considering various key factors such as the humidity of the stockpile, precipitation intensity, sewage discharge, temperature, wind direction, and wind speed. It can not only accurately evaluate the current pollution situation but also predict the pollution trend of the stockyard in the future for a period of time in advance. This helps the management personnel formulate countermeasures in advance. For example, when it is predicted that precipitation or increased humidity may lead to an increase in pollution risk, measures such as arranging stockpile coverage and checking drainage facilities in advance can be taken to improve the scientific nature and initiative of stockyard management and reduce the environmental pollution risk.

[0086] The main monitoring device uses a fusion algorithm (such as the weighted average method) to process the sensing data of multiple secondary monitoring devices, and assigns weights by comprehensively considering factors such as the stability of sensor historical data, effectively reducing the impact of individual sensor errors or abnormal data on the overall monitoring results, and improving the reliability and accuracy of the data. For example, a higher weight is given to sensors with better long-term stability, so that the fused data can better reflect the real environmental conditions and provide a solid data basis for accurate decision-making.

[0087] The densely arranged water-permeable layers are set at the bottom and side of the secondary monitoring device housing, and the pore diameter is smaller than the pore diameter of the stockpile materials. It can not only ensure that water can flow in and converge during precipitation for detection by the water flow sensor but also prevent the materials from entering the device interior and damaging components such as sensors, effectively improving the stability and service life of the equipment in the complex stockyard environment.

[0088] The Beidou positioning module equipped in the secondary monitoring device monitors the displacement change of the equipment in real time, can promptly detect whether the equipment has been removed or damaged, prevent the loss of monitoring equipment or malicious tampering of data, and ensure the security and integrity of the monitoring system. At the same time, this also helps with the management and maintenance of the equipment. Once an abnormal equipment position is found, measures can be quickly taken to handle it to ensure the continuity of the monitoring work.

[0089] This monitoring system adopts a distributed architecture, and multiple secondary monitoring devices can be flexibly arranged according to the scale, terrain, and pollution risk distribution of the stockyard, and can adapt to the needs of different types and scales of stockyards. Whether it is a small-scale organic fertilizer stockyard or a large-scale ore or construction waste stockyard, comprehensive and effective pollution monitoring can be achieved by reasonably arranging the secondary monitoring devices, which has strong flexibility.

[0090] Through real-time and accurate pollution monitoring and early warning, the pollution emissions of the stockyard can be effectively controlled, the pollution of soil and groundwater caused by sewage seepage into the ground can be reduced, the air pollution caused by dust can be reduced, the surrounding ecological environment can be protected, which helps to maintain the ecological balance and promote the sustainable development of the region.

[0091] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent distributed stockpile pollution source monitoring system, characterized in that: The system comprises a plurality of pile material pollution auxiliary monitoring devices (1), a pile material pollution main monitoring device (2), a 5G signal transmission terminal (3), and a monitoring server (4), wherein the plurality of pile material pollution auxiliary monitoring devices (1) are connected to the pile material pollution main monitoring device (2) via Bluetooth communication, the pile material pollution main monitoring device (2) is connected to the 5G signal transmission terminal (3) via 5G communication, and the 5G signal transmission terminal (3) is connected to the monitoring server (4) via 5G communication; The stockpile pollution auxiliary monitoring device (1) comprises a battery module (11) and a first processor (12) electrically connected to the battery module (11), a rainfall sensor (13), a humidity sensor (14), a water flow sensor (15), an audible and visual alarm (16), a wind sensor (17), a Beidou positioning module (18), a dust sensor (19), and a first Bluetooth communication module (110); the first processor (12) is connected to the rainfall sensor (13), the humidity sensor (14), the water flow sensor (15), the audible and visual alarm (16), the wind sensor (17), the Beidou positioning module (18), the dust sensor (19), and the first Bluetooth communication module (110) by wired communication; The stockpile pollution main monitoring device (2) comprises a power module (21) and a second processor (22) electrically connected to the power module (21), a second Bluetooth communication module (23), and a 5G communication module (24), wherein the second processor (22) is connected to the second Bluetooth communication module (23) and the 5G communication module (24) by wired communication; The bottom and side of the shell of the pile pollution auxiliary monitoring device (1) are provided with densely distributed water-permeable layers, and the pore size of the water-permeable layer is smaller than the pore size of the pile material.

2. The intelligent distributed stockpile pollution source monitoring system according to claim 1 is characterized in that: The rainfall sensor (13) is located at the top of the housing of the main monitoring device for stockpile pollution (2), the humidity sensor (14) is located at the inner side wall of the middle part of the main monitoring device for stockpile pollution (2), and the water flow sensor (15) is located at the lower part of the main monitoring device for stockpile pollution (2).

3. The intelligent distributed stockpile pollution source monitoring system according to claim 1 is characterized in that: The water flow sensor (15) comprises a water flow sensing probe (151), a funnel-shaped water collecting surface (152), and a water collecting pipe (153); the edge of the funnel-shaped water collecting surface (152) is connected to the inner side of the shell of the stockpile pollution main monitoring device (2); the bottom of the funnel-shaped water collecting surface (152) is connected to the water collecting pipe (153); and the water flow sensing probe (151) is arranged on the inner wall of the water collecting pipe (153).

4. The intelligent distributed stockpile pollution source monitoring system according to claim 1 is characterized in that: The sound and light alarm (16) is located on the top of the housing of the stockpile pollution auxiliary monitoring device (1), and comprises a sound alarm and a light alarm, wherein the light alarm comprises a yellow, orange and red LED light.

5. The intelligent distributed stockpile pollution source monitoring system according to claim 1 is characterized in that: The wind force sensor (17) and the dust sensor (19) are located on the outer side of the housing of the stockpile pollution auxiliary monitoring device (1); the dust sensor (19) comprises a parabolic dust gathering surface (191), an exhaust blade (192), a rotating shaft (193), a dust gathering chamber (194), a motor (195), and a dust sensing probe (196); the motor (195) is installed in the dust gathering chamber (194); the rotating shaft of the motor (195) is mechanically connected to one end of the rotating shaft (193); the other end of the rotating shaft (193) is connected to the exhaust blade (192); the dust sensing probe (196) is arranged on the inner wall of the dust gathering chamber (194); and the connection point between the rotating shaft (193) and the exhaust blade (192) is located at the focal position of the parabolic dust gathering surface (191).

6. The intelligent distributed stockpile pollution source monitoring system according to claim 1 is characterized in that: The battery module (11), the first processor (12), the Beidou positioning module (18), and the first Bluetooth communication module (110) are arranged on a control board inside a control board housing (112), and the control board housing (112) is located on the top of the housing of the stockpile pollution auxiliary monitoring device (1).

7. A working method of the intelligent distributed stockpile pollution source monitoring system according to any one of claims 1 to 6, characterized in that: The following steps are involved: The rainfall sensors (13), humidity sensors (14), and water flow sensors (15) of the plurality of pile pollution auxiliary monitoring devices (1) detect sensor data in the pile in real time, wherein: when the first processor (12) senses the presence of precipitation through the data sent by the rainfall sensor (13), the yellow light of the sound and light alarm (16) flashes, and a sound warning is issued to prevent sewage discharge; when the first processor (12) senses the presence of 100% humidity of the pile through the data sent by the humidity sensor (14), the orange light of the sound and light alarm (16) flashes, and a sound warning is issued that sewage will be discharged from the pile; when the first processor (12) senses the flow of sewage through the data sent by the water flow sensor (15), the red light of the sound and light alarm (16) flashes, and an alarm sounds; At the same time, the wind sensor (17) monitors the wind data at the stockpile site. When the first processor (12) detects the wind data through the wind sensor (17), the first processor (12) controls the dust sensor (19) to collect dust data around the stockpile. When dust is sensed, the red light of the sound and light alarm (16) flashes and an alarm sounds. The Beidou positioning module (18) monitors the displacement change of the stockpile pollution auxiliary monitoring device (1) to monitor whether it has been removed. The first processor (12) wirelessly transmits the monitored pile sewage and dust sensor data to the second Bluetooth communication module (23) of the pile pollution main monitoring device (2) through the first Bluetooth communication module (110); the second Bluetooth communication module (23) transmits the received sensor data to the second processor (22); the second processor (22) processes the data sent by Bluetooth communication and stores it in the storage module (25), and wirelessly transmits the data to the 5G signal transmission terminal (3) through the 5G communication module (24) using 5G communication, and finally the monitoring server (4) obtains the pollution monitoring data of the pile.

8. The working method of the intelligent distributed stockpile pollution source monitoring system according to claim 7 is characterized in that: The first processor (12) sets a dynamic threshold model according to historical precipitation data, the change law of stockpile humidity and sewage discharge conditions, specifically: Let x1, x2, ..., x n is an input feature, which includes at least historical precipitation data, historical stockpile humidity data, historical sewage discharge data, and historical dust data, where x i represents the i-th feature, y is the corresponding threshold, and the prediction formula of the decision tree regression model is: Among them, J is the number of leaf nodes in the decision tree, R j is the feature space area corresponding to the jth leaf node, c j is the predicted value of leaf node j, I(x∈R j ) is an exponential function. When x belongs to R j 1 when it is, otherwise 0; When training the model, the predicted value c of the leaf node is determined by the minimum mean square error j Among them, N j is the number of samples belonging to leaf node j, y j is the true threshold of the corresponding sample; The overall objective function of the model is Where N is the number of training samples, is the prediction threshold of the model for the i-th sample. When the real-time threshold is adjusted, the feature data collected in real time is brought into the trained model to obtain the dynamic threshold y d .

9. The working method of the intelligent distributed stockpile pollution source monitoring system according to claim 8 is characterized in that: The main material pile pollution monitoring device (2) processes the sensor data monitored by the multiple material pile pollution secondary monitoring devices (1) through a fusion algorithm, and then wirelessly sends the data to the 5G signal transmission terminal (3). Finally, the monitoring server (4) obtains the pollution monitoring data of the material pile. Specifically, for each monitoring parameter, a corresponding weight is assigned according to the historical data stability of the sensor of the secondary monitoring device. Assuming that there are n material pile pollution secondary monitoring devices in total, for a certain monitoring parameter P, the measurement value of the i-th device is P i , and its corresponding weight is ω i , then the fused measurement value P f for:

10. The working method of the intelligent distributed stockpile pollution source monitoring system according to claim 9 is characterized in that: The monitoring server (4) predicts the pollution trend of the stockpile yard in the future based on multiple linear regression and exponential smoothing method, where: Taking the fused measured values ​​as independent variables and the pollution index as dependent variables, the fused measured values ​​include pile humidity, precipitation intensity, sewage discharge, temperature, wind direction, and wind speed; Establish a multiple linear regression model: Y = β0 + β1X1 + β2X2 + ··· + β n X n +ε, where Y is the predicted value of the pollution index, β0 is a constant term, β i is the regression coefficient corresponding to each factor, X i is the measured value after fusion, ε is the error term; The regression coefficient is estimated by the least square method to minimize the sum of squares of the error between the predicted value and the actual value; The initial prediction results obtained by the multivariate linear regression model are subjected to exponential smoothing. Let F t is the predicted value at time t, Y t is the actual value at time t, α is the smoothing coefficient, 0<α<1; The exponential smoothing formula is: F t+1 =αY t +(1-α)F t ,Through continuous iteration, the prediction results are dynamically adjusted according to new actual data; The smoothed prediction result F t+1 and the corresponding dynamic threshold y d Compare and classify the early warning mechanisms according to the comparison results. The classification formula is: When classified as low pollution, a yellow warning is activated, and the monitoring server (4) sends an alarm rectification notice to the on-site staff; When classified as medium pollution, an orange warning is activated, and the monitoring server (4) sends an alarm rectification notice to the on-site staff and automatically sends pollution warning information to the environmental protection department; When classified as high pollution, a red alert is activated, and the monitoring server (4) sends an alarm notification to the on-site staff and automatically sends a request for law enforcement to the environmental protection department.

Images