A fog cannon eagle eye intelligent linkage control system

Abstract

The present invention discloses a fog cannon eagle eye intelligent linkage control system, which relates to the field of environmental monitoring and governance technology. The intelligent linkage control system is composed of a high-magnification eagle eye camera, an environmental sensor, a fog cannon device, an NPU module, a fog cannon device linkage control module, an Internet of Things gateway and a central control system. The NPU is a neural network processing unit; wherein the central control system is centered on the Internet of Things gateway and is connected to the high-magnification eagle eye camera, the environmental sensor and the fog cannon device in communication. The present invention realizes efficient collaboration and automatic response between devices by combining high-performance AI processors, intelligent linkage control algorithms and Internet of Things technologies. The system can not only identify pollution sources in real time, but also dynamically adjust spraying strategies according to environmental conditions, greatly improving governance efficiency and resource utilization. This intelligent improvement effectively overcomes the shortcomings of traditional equipment and realizes refined and low-cost pollution control.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring and control, in particular to air pollution control and dust control, and in particular to a fog cannon Hawkeye intelligent linkage control system. Background Art

[0002] With the advancement of science and technology, especially the rapid development of machine vision, data analysis, artificial intelligence and other technologies, it has provided strong technical support for the research and development and application of the fog cannon Eagle Eye intelligent linkage control system.

[0003] For example, Chinese Patent Publication No. CN115780122A discloses a redundant control system and method for automatic rotation safety protection of a fog cannon, including: a hydraulic control module, a limit sensing module and a controller; the hydraulic control module includes: a rotation hydraulic reversing valve of a fog cannon rotation motor, a limit oil circuit cut-off valve, an oil circuit limit release valve, a check valve, and a hydraulically controlled check valve.

[0004] In the prior art, system faults are judged by means of fault detection programs and time control; redundant protection is provided to the system from three aspects: control logic, limit sensors, and hydraulic systems, so as to solve the problem that there are certain visual blind spots in the control of fog cannons. The fog cannons can only be observed through cameras, and faults cannot be discovered in time, which easily causes accidents of fog cannons. However, since traditional environmental monitoring equipment is often separated from fog cannon devices and lacks an intelligent linkage mechanism, data exchange and control scheduling between various devices are not efficient enough, and automated linkage control cannot be achieved. In addition, there is a lack of computer vision and deep learning algorithms for pollution source identification and analysis, which makes automated identification and positioning difficult to achieve, and it is impossible to provide accurate pollution control strategies, and thus it is impossible to achieve real-time linkage and scheduling of different equipment and areas. Summary of the invention

[0005] The purpose of the present invention is to provide a fog cannon Hawkeye intelligent linkage control system to solve the problems raised in the above background technology.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] A fog cannon Hawkeye intelligent linkage control system, the intelligent linkage control system is composed of a high-magnification Hawkeye camera, an environmental sensor, a fog cannon device, an NPU module, a fog cannon device linkage control module, an Internet of Things gateway and a central control system, and the NPU is a neural network processing unit;

[0008] The central control system is centered on the Internet of Things gateway and is connected to the high-magnification Hawkeye camera, environmental sensor and fog cannon equipment in communication;

[0009] The high-magnification eagle-eye camera is used to capture video stream data in real time;

[0010] The environmental sensor is used to collect air quality data, the air quality data includes PM2.5, PM10 concentration and wind speed, wind direction environmental parameters, and the environmental sensor includes a PM2.5 concentration sensor, a PM10 concentration sensor, a wind speed concentration sensor and a wind direction sensor;

[0011] The NPU module analyzes the location and type of pollution sources in the video stream data through a deep learning algorithm, and combines the results with environmental sensor data to identify the intensity and distribution of pollution sources;

[0012] The fog cannon equipment linkage control module automatically adjusts the spraying strategy of the fog cannon equipment according to the output of the NPU module, performs precise spraying, and realizes fixed-point and quantitative spraying operation;

[0013] The IoT gateway is responsible for data transmission and remote control.

[0014] A further improvement of the technical solution of the present invention is that the collection process of the video stream data and the air quality data includes:

[0015] Install high-magnification Hawkeye cameras in key monitoring areas, start the high-magnification Hawkeye cameras, monitor pollution sources in the area in real time, collect high-resolution video streams, capture image details of pollution sources, and connect video stream data to the central control system through the IoT gateway;

[0016] Using the deployed PM2.5 concentration sensor, PM10 concentration sensor, wind speed sensor and wind direction sensor, real-time air quality data including PM2.5 and PM10 concentrations and wind speed and wind direction environmental parameters are collected, and the air quality data is connected to the central control system through the IoT gateway;

[0017] In the central control system, the video stream data and air quality data received are verified and corrected to ensure the accuracy and integrity of the data, and the data of the pollution source data set are preprocessed. The preprocessing operation includes data cleaning and data format conversion steps. Through data cleaning, invalid or abnormal data generated due to equipment failure, transmission error, etc. are corrected, and obviously erroneous data are converted. Through data format conversion, the video stream data collected by the camera is converted into a format suitable for storage, and the data collected by the environmental sensor is converted into a unified format for subsequent processing;

[0018] The preprocessed video stream data is integrated with the air quality data to form a complete pollution source data set.

[0019] A further improvement of the technical solution of the present invention is that: the NPU module includes an edge computing unit, a data processing and analysis unit, and an intelligent decision control unit, wherein the electrical signals between the units are connected;

[0020] The edge computing unit, by running a deep learning model, performs local data processing based on video stream data, identifies pollution sources in real time, processes data quickly locally, reduces data transmission delays, and improves the response speed and processing efficiency of the system;

[0021] The data processing and analysis unit, based on the pollution source identification result of the edge computing unit, combines the result with the environmental sensor data to identify the intensity and distribution of the pollution source, realizes accurate identification and positioning of the pollution source, and provides accurate decision-making basis for intelligent control;

[0022] The intelligent decision-making control unit dynamically adjusts the spray angle, intensity and coverage of the fog cannon based on air quality data and pollution source analysis results using fuzzy control and PID control algorithms. This module ensures the system's response speed and spraying accuracy, achieving precise pollution control.

[0023] A further improvement of the technical solution of the present invention is that in the edge computing unit, the real-time identification process of the pollution source includes:

[0024] Extract environmental image data of the monitored area from the public database, use the image annotation tool (LabelImg) to annotate the pollution source images, including the pollution source types of smoke, dust, and emissions, and divide the pollution source image dataset into a training set and a test set;

[0025] Use the training set data to train the deep learning model and the pollution source identification model so that it can identify the pollution source types of smoke, dust, and emissions. Use the test set data to evaluate the pollution source identification model and calculate indicators such as accuracy, recall rate, and F1 score. Adjust the model parameters based on the evaluation results.

[0026] Deploy the trained pollution source identification model to the NPU, load the preprocessed video stream data to the edge computing unit, and extract the color, shape, and texture features related to pollution source identification from the preprocessed video stream data;

[0027] Based on the extracted features, the pollution source identification model is used for real-time analysis, the pollution source matching index is calculated, the location of the pollution source in the video stream is identified, and the identification results are output, classified, and the pollution type is determined.

[0028] A further improvement of the technical solution of the present invention is that the calculation expression of the pollution source matching index is:

[0029] ;

[0030] in, is the pollution source matching index, It is the color feature matching score, which is a value between 0 and 1. The higher the value, the more the color feature matches the color of the known pollution source. It is the shape feature matching score, which is a value between 0 and 1. The higher the value, the more the shape feature matches the known pollution source shape. It is the texture feature matching score, which is a value between 0 and 1. The higher the value, the more the texture feature matches the texture of the known pollution source. is a constant used to adjust the sensitivity of the matching index. is the benchmark score of the color feature, which is used to normalize the color feature matching degree. The value range is between 0 and 1. When , it means no pollution source is detected. , it means that the detected pollution source characteristics completely match the known pollution source.

[0031] A further improvement of the technical solution of the present invention is that in the data processing and analysis unit, the process of identifying the intensity and distribution of pollution sources includes:

[0032] Receive the pollution source identification results sent by the edge computing unit, including the location, type and pollution degree evaluation index of the pollution source, extract the real-time PM2.5 concentration and PM10 concentration data from the air quality data, and traverse the historical database to determine the baseline concentration value and maximum concentration value of PM2.5 concentration and PM10 concentration;

[0033] The pollution source matching index is combined with PM2.5 and PM10 concentration-related data to calculate the pollution level evaluation index and further analyze the pollution level of the pollution source;

[0034] Extract real-time wind speed and direction data from air quality data, analyze the impact of wind speed and wind direction on pollutant diffusion, determine the possible transmission path of pollutants, and calculate the pollution source distribution coefficient by combining the pollution degree evaluation index and wind speed and direction data;

[0035] According to the pollution source distribution coefficient, the diffusion direction and speed of pollutants are determined to reflect the spatial distribution trend of pollution sources, and geographic information system tools are used to visualize the diffusion path and potential impact areas of pollutants.

[0036] A further improvement of the technical solution of the present invention is that the calculation expression of the pollution degree evaluation index is:

[0037] ;

[0038] in, is the pollution degree evaluation index, is the pollution source matching index, For real-time monitoring of PM2.5 concentration, For real-time monitoring of PM10 concentration, is the baseline concentration value of PM2.5, used to normalize the concentration value. is the baseline concentration value of PM10, used to normalize the concentration value. is the maximum concentration value of PM2.5, which is used to calculate the exponential decay term. is the maximum concentration value of PM10, which is used to calculate the exponential decay term. The value range is between 0 and 1. When , it means that no pollution source is detected or the pollution concentration is very low. , it means that the detected pollution source characteristics completely match the known pollution source.

[0039] A further improvement of the technical solution of the present invention is that the calculation expression of the pollution source distribution coefficient is:

[0040] ;

[0041] in, is the pollution source distribution coefficient, is the pollution degree evaluation index, is the wind speed, which is used to measure the diffusion rate of pollutants. is the reference wind speed value, used to normalize the wind speed. It is the angle between wind direction and the diffusion direction of pollutants, and is used to measure the effect of wind direction on the diffusion of pollutants. is the maximum wind direction angle, which is used to limit the influence of the exponential attenuation term. It is the angle between the wind direction and the north direction, which is used to calculate the trigonometric effect of the pollutant diffusion direction. When the wind is in the north (or south), When the wind is due east or west, the sine value is 0, indicating that the pollutants will not spread in the east-west direction. When the angle is 90 degrees, the sine value reaches the maximum, indicating that the diffusion intensity of pollutants in the east-west direction is the largest. The value range is between 0 and 1. When , it means that there is no pollution diffusion or the pollution level is very low. When , it means that the pollutant diffusion conditions are optimal, the pollution level is high, and the wind speed and direction are most conducive to the diffusion of pollutants.

[0042] A further improvement of the technical solution of the present invention is that: in the intelligent decision-making control unit, the process of dynamically adjusting the spray angle, intensity and coverage of the fog cannon includes:

[0043] Integrate real-time air quality data and pollution source analysis results, including the location, type, pollution level, pollutant diffusion direction and speed of the pollution source, and then preset the initial values ​​of the spray angle, intensity and coverage of the fog cannon equipment according to environmental conditions and pollution level. The initial value serves as the starting point of the control process and is used to initially respond to pollution conditions. The spray angle is determined according to the location, wind direction and wind speed of the pollution source, the spray intensity is adjusted according to the pollution level and pollution source type, and the coverage ensures that the pollution source and its potential diffusion area are covered;

[0044] Design fuzzy control rules to convert qualitative air quality data and pollution source analysis results into fuzzy logic inputs for fog cannon equipment operation, calculate preliminary spray parameters of spray angle, intensity and coverage of fog cannon equipment based on real-time data, and fine-tune preliminary spray parameters of fog cannon equipment in combination with PID control algorithm to achieve more accurate pollution control, where the fuzzy controller provides qualitative control decisions and the PID controller provides quantitative adjustments;

[0045] The output results of the intelligent decision-making control unit are converted into specific parameters of the spray angle, intensity and coverage range of the fog cannon, and a control command is generated and sent to the linkage control module of the fog cannon equipment.

[0046] A further improvement of the technical solution of the present invention is that in the fog cannon equipment linkage control module, the process of adjusting the spraying strategy of the fog cannon equipment includes:

[0047] Receive specific parameters of the spray angle, intensity and coverage of the fog cannon equipment from the intelligent decision control module, and parse the received control instructions and convert them into specific parameters that the fog cannon equipment understands and executes;

[0048] According to the analyzed parameters, the fog cannon equipment linkage control module automatically adjusts the status of the fog cannon equipment, including starting or stopping the equipment, adjusting the spray angle, controlling the spray intensity, etc. After the adjustment is completed, the fog cannon equipment performs fixed-point and quantitative spray operations according to the preset spraying strategy;

[0049] The fog cannon equipment linkage control module monitors the operating status and spraying effect of the fog cannon equipment in real time during the spraying process, and feeds back the relevant information to the intelligent decision-making control module for further optimization and adjustment.

[0050] Due to the adoption of the above technical solution, the present invention has the following technical advances compared with the prior art:

[0051] The present invention provides a fog cannon Hawkeye intelligent linkage control system, which realizes efficient collaboration and automatic response between devices by combining high-performance AI processors, intelligent linkage control algorithms and Internet of Things technology. The system can not only identify pollution sources in real time, but also dynamically adjust spraying strategies according to environmental conditions, greatly improving the control efficiency and resource utilization. This intelligent improvement effectively overcomes the shortcomings of traditional equipment and realizes refined and low-cost pollution control.

[0052] The present invention provides a fog cannon Hawkeye intelligent linkage control system. Through deep learning and computer vision technology, the system can accurately identify and locate pollution sources in complex scenarios, rather than simply spraying in an area. By performing video stream analysis, high-dimensional features can be extracted from the video to achieve accurate identification of smoke and dust. Accurately locating the pollution source enables the fog cannon equipment to concentrate on spraying the pollution source area, reducing water resource waste and the problem of insufficient spraying coverage. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0054] Figure 1 It is a product structure diagram of the present invention;

[0055] Figure 2 A flow chart for real-time identification of pollution sources of the present invention;

[0056] Figure 3 This is a flow chart for identifying the intensity and distribution of pollution sources of the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0058] Embodiment 1, as Figure 1 , Figure 2 As shown, the present invention provides a fog cannon Hawkeye intelligent linkage control system, which is composed of a high-magnification Hawkeye camera, an environmental sensor, a fog cannon device, an NPU module, a fog cannon device linkage control module, an Internet of Things gateway and a central control system, and the NPU is a neural network processing unit;

[0059] The central control system is centered on the IoT gateway, which is connected to the high-magnification Hawkeye camera, environmental sensors and fog cannon equipment;

[0060] High-magnification Eagle Eye cameras are used to capture video streaming data in real time;

[0061] Environmental sensors are used to collect air quality data, including PM2.5, PM10 concentration, wind speed, and wind direction environmental parameters. Environmental sensors include PM2.5 concentration sensor, PM10 concentration sensor, wind speed concentration sensor, and wind direction sensor.

[0062] The NPU module uses deep learning algorithms to analyze the location and type of pollution sources in video stream data and combines the results with environmental sensor data to identify the intensity and distribution of pollution sources;

[0063] The fog cannon equipment linkage control module automatically adjusts the spraying strategy of the fog cannon equipment according to the output of the NPU module, performs precise spraying, and realizes fixed-point and quantitative spraying operations;

[0064] The IoT gateway is responsible for data transmission and remote control;

[0065] Furthermore, the process of collecting video stream data and air quality data includes:

[0066] High-magnification Hawkeye cameras are installed in key monitoring areas. When the high-magnification Hawkeye cameras are started, pollution sources in the area are monitored in real time, high-resolution video streams are collected, image details of pollution sources are captured, and video stream data are connected to the central control system through the Internet of Things gateway. The deployed PM2.5 concentration sensor, PM10 concentration sensor, wind speed sensor and wind direction sensor are used to collect air quality data including PM2.5 and PM10 concentrations and wind speed and wind direction environmental parameters in real time, and the air quality data is connected to the central control system through the Internet of Things gateway. The accessed video stream data and air quality data are verified and corrected in the central control system to ensure the accuracy and completeness of the data, and the data of the pollution source data set is preprocessed. The preprocessing operation includes data cleaning and data format conversion steps. Through data cleaning, invalid or abnormal data generated due to equipment failure, transmission error, etc. are cleared, and obviously erroneous data are corrected. Through data format conversion, the video stream data collected by the camera is converted into a format suitable for storage, and the data collected by the environmental sensor is converted into a unified format for subsequent processing. The preprocessed video stream data is integrated with the air quality data to form a complete pollution source data set.

[0067] Embodiment 2, as Figure 3As shown, based on Example 1, the present invention provides a technical solution: preferably, the NPU module includes an edge computing unit, a data processing and analysis unit, and an intelligent decision control unit, wherein the electrical signals between the units are connected;

[0068] The edge computing unit runs a deep learning model and performs local data processing based on video stream data to identify pollution sources in real time, quickly process data locally, reduce data transmission delays, and improve the system's response speed and processing efficiency. It extracts environmental image data of the monitored area from the public database, uses the image annotation tool (LabelImg) to annotate the pollution source images, including the pollution source types of smoke, dust, and emissions, and divides the pollution source image data set into a training set and a test set. The training set data is used to test the deep learning model (YOLO or Faster R-CNN) is used for training, and the pollution source identification model is trained to enable it to identify the pollution source types of smoke, dust, and emissions. The pollution source identification model is evaluated using the test set data, and the accuracy, recall rate, F1 score and other indicators are calculated. According to the evaluation results, the model parameters are adjusted, and the trained pollution source identification model is deployed on the NPU. The pre-processed video stream data is loaded into the edge computing unit, and the color, shape, and texture features related to pollution source identification are extracted from the pre-processed video stream data. Based on the extracted features, the pollution source identification model is used for real-time analysis, the pollution source matching index is calculated, the pollution source location in the video stream is identified, and the identification results are output, classified, and the pollution type is determined;

[0069] Furthermore, the calculation expression of the pollution source matching index is:

[0070] ;

[0071] in, is the pollution source matching index, It is the color feature matching score, which is a value between 0 and 1. The higher the value, the more the color feature matches the color of the known pollution source. It is the shape feature matching score, which is a value between 0 and 1. The higher the value, the more the shape feature matches the known pollution source shape. It is the texture feature matching score, which is a value between 0 and 1. The higher the value, the more the texture feature matches the texture of the known pollution source. is a constant used to adjust the sensitivity of the matching index. is the benchmark score of the color feature, which is used to normalize the color feature matching degree. The value range is between 0 and 1. When , it means no pollution source is detected. When , it means that the detected pollution source characteristics completely match the known pollution source. , and The increase in pollution source matching index Increase, indicating that the confidence in the identification of the pollution source is improved. The use of radical and logarithmic functions increases the nonlinearity of the formula, making the pollution source matching index more sensitive to changes in the characteristic value;

[0072] The data processing and analysis unit, based on the pollution source identification results of the edge computing unit, combines the results with the environmental sensor data, identifies the intensity and distribution of the pollution source, realizes the accurate identification and positioning of the pollution source, and provides accurate decision-making basis for intelligent control. It receives the pollution source identification results sent by the edge computing unit, including the location, type and pollution degree evaluation index of the pollution source, and extracts real-time PM2.5 concentration and PM10 concentration data from the air quality data, traverses the historical database to determine the baseline concentration value and maximum concentration value of PM2.5 concentration and PM10 concentration, combines the pollution source matching index with the PM2.5 and PM10 concentration related data, calculates the pollution degree evaluation index, further analyzes the pollution degree of the pollution source, extracts real-time wind speed and wind direction data from the air quality data, analyzes the influence of wind speed and wind direction on the diffusion of pollutants, determines the possible transmission path of pollutants, and calculates the pollution source distribution coefficient based on the pollution degree evaluation index and wind speed and wind direction data. According to the pollution source distribution coefficient, the diffusion direction and speed of pollutants are determined to reflect the spatial distribution trend of pollution sources, and uses geographic information system tools to visualize the diffusion path and potential impact area of ​​pollutants;

[0073] Furthermore, the calculation expression of the pollution degree evaluation index is:

[0074] ;

[0075] in, is the pollution degree evaluation index, is the pollution source matching index, For real-time monitoring of PM2.5 concentration, For real-time monitoring of PM10 concentration, is the baseline concentration value of PM2.5, used to normalize the concentration value. is the baseline concentration value of PM10, used to normalize the concentration value. is the maximum concentration value of PM2.5, which is used to calculate the exponential decay term. is the maximum concentration value of PM10, which is used to calculate the exponential decay term. The value range is between 0 and 1. When , it means that no pollution source is detected or the pollution concentration is very low. , it means that the detected pollution source characteristics completely match the known pollution source, and and are very high, with The increase in pollution level evaluation index Increases, indicating that the confidence in identifying the pollution source increases. and As the square term in the denominator increases, decrease, indicating that the pollution concentration has a negative impact on the evaluation index. The exponential decay term decreases with and decreases with the increase of The value of indicates that high pollution concentration has a significant negative impact on the evaluation index;

[0076] Furthermore, the calculation expression of the pollution source distribution coefficient is:

[0077] ;

[0078] in, is the pollution source distribution coefficient, is the pollution degree evaluation index, is the wind speed, which is used to measure the diffusion rate of pollutants. is the reference wind speed value, used to normalize the wind speed. It is the angle between wind direction and the diffusion direction of pollutants, and is used to measure the effect of wind direction on the diffusion of pollutants. is the maximum wind direction angle, which is used to limit the influence of the exponential attenuation term. It is the angle between the wind direction and the north direction, which is used to calculate the trigonometric effect of the pollutant diffusion direction. When the wind is in the north (or south), When the wind is due east or west, the sine value is 0, indicating that the pollutants will not spread in the east-west direction. When the angle is 90 degrees, the sine value reaches the maximum, indicating that the diffusion intensity of pollutants in the east-west direction is the largest. The value range is between 0 and 1. When , it means that there is no pollution diffusion or the pollution level is very low. When , it means that the pollutant diffusion conditions are optimal, the pollution level is high, and the wind speed and wind direction are most conducive to the diffusion of pollutants. The increase in pollution source distribution coefficient Increase, indicating that the higher the pollution level, the greater the possibility of pollutant diffusion. The increase of radical terms leads to Increase, indicating that the higher the wind speed, the faster the pollutants diffuse. As the exponential decay term increases, the exponential decay term decreases, resulting in decreases, indicating that the greater the angle between the wind direction and the diffusion direction of pollutants, the more restricted the diffusion of pollutants. Indicates the effect of wind direction on the diffusion direction of pollutants. When the wind direction is due north, this item is 1, indicating that the diffusion direction of pollutants is the best;

[0079] The intelligent decision-making control unit dynamically adjusts the spray angle, intensity and coverage of the fog cannon based on air quality data and pollution source analysis results, using fuzzy control and PID control algorithms. This module ensures the system's response speed and spraying accuracy, achieves precise pollution control, and integrates real-time air quality data and pollution source analysis results, including the location, type, degree of pollution, direction and speed of pollutant diffusion of the pollution source, and then presets the initial values ​​of the spray angle, intensity and coverage of the fog cannon equipment according to environmental conditions and pollution degree. The initial value serves as the starting point of the control process and is used to initially respond to pollution conditions. The spray angle is determined according to the location, wind direction and wind speed of the pollution source, and the spray intensity is determined according to the pollution degree and Adjust the pollution source type and ensure that the coverage covers the pollution source and its potential diffusion area. Design fuzzy control rules to convert qualitative air quality data and pollution source analysis results into fuzzy logic inputs for fog cannon equipment operation. Calculate the preliminary spray parameters of the spray angle, intensity and coverage of the fog cannon equipment based on real-time data, and fine-tune the preliminary spray parameters of the fog cannon equipment in combination with the PID control algorithm to achieve more accurate pollution control. The fuzzy controller provides qualitative control decisions, while the PID controller provides quantitative adjustments. Convert the output results of the intelligent decision-making control unit into specific parameters of the spray angle, intensity and coverage of the fog cannon, and generate control commands to be sent to the linkage control module of the fog cannon equipment.

[0080] In the fog cannon equipment linkage control module, the process of adjusting the fog cannon equipment spraying strategy includes:

[0081] The specific parameters of the spray angle, intensity and coverage range of the fog cannon equipment are received from the intelligent decision-making control module, and the received control instructions are parsed to convert them into specific parameters that the fog cannon equipment understands and executes. According to the parsed parameters, the fog cannon equipment linkage control module automatically adjusts the status of the fog cannon equipment, including starting or stopping the equipment, adjusting the spray angle, controlling the spray intensity, etc. After the adjustment is completed, the fog cannon equipment performs fixed-point and quantitative spray operations according to the preset spraying strategy. The fog cannon equipment linkage control module monitors the operating status and spraying effect of the fog cannon equipment in real time during the spraying process, and feeds back the relevant information to the intelligent decision-making control module for further optimization and adjustment.

[0082] Example 3, based on Examples 1-2, the present invention provides the following practical application effects and test data examples:

[0083] A. In a test at an industrial park, compared with the traditional manually controlled fog cannon equipment, the response speed of this system was shortened by about 50% when the pollution source suddenly appeared, from 3 minutes to 1.5 minutes to complete the spraying start. In the dust control test, the system reduced the PM10 concentration from 300 micrograms / cubic meter to 120 micrograms / cubic meter within 10 minutes, and the dust reduction efficiency reached more than 60%. Through automatic control and self-learning optimization, the system saved 30% of water resource usage and about 40 hours of manual operation time in a one-month test cycle;

[0084] B. In a simulated test environment, compared with traditional fog cannon equipment, within 10 minutes after the system was started at the same dust concentration, the concentration of PM2.5 and PM10 decreased by an average of 60%, while the traditional equipment decreased by 40%. In construction site applications, the system reduced water usage by about 30% within 1 month and increased spraying efficiency by 25%. In actual application in an industrial park, the system was able to start spraying measures within 1 minute during sudden pollution (such as increased smoke and dust emissions), which doubled the response speed of traditional systems. In urban road dust control projects, the system achieved remote automated control, reducing manual maintenance time by more than 70%, ensuring the continuous and stable operation of the equipment at night and in severe weather conditions;

[0085] In addition, based on the Internet of Things technology and edge computing, real-time fusion and processing of multiple sensor data is achieved, which greatly shortens the response time of pollution detection and governance. Edge computing technology enables the equipment to process data and execute control locally without relying on cloud computing, reducing data transmission delay. In the experimental scenario, compared with the traditional manual control method, the system can start spraying within 1-2 minutes of the pollution concentration change, which is more than 50% faster than the traditional method.

[0086] The system's automated operation and self-learning capabilities significantly reduce the need for manual operation and maintenance costs. The intelligent control algorithm combined with automated control devices can operate unattended, reducing the manpower investment in long-term operations. According to field application feedback, the application of the system on construction sites can reduce manual operation time by more than 50 hours per month and save about 30% of water consumption.

[0087] Traditional fog cannon equipment relies on timed spraying and manual operation and cannot be adjusted in real time according to the pollution situation. This system can dynamically adjust the spraying angle and intensity through intelligent analysis and adaptive control to achieve more refined governance. Under the same environment, the pollutant concentration of this system decreases more than 20% faster than traditional fog cannon equipment.

[0088] Other smart devices usually need to rely on cloud processing and have certain network delays, especially in areas with unstable networks. This system is based on edge computing and performs fast processing locally, ensuring the device's immediate response capability. In industrial area applications with unstable networks, the response delay of this system is reduced by about 30% compared to other devices, greatly improving processing efficiency.

[0089] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A fog cannon Hawkeye intelligent linkage control system, characterized in that: The intelligent linkage control system is composed of a high-magnification Hawkeye camera, an environmental sensor, a fog cannon device, an NPU module, a fog cannon device linkage control module, an Internet of Things gateway and a central control system, and the NPU is a neural network processing unit; The central control system is centered on the Internet of Things gateway and is connected to the high-magnification Hawkeye camera, environmental sensor and fog cannon equipment in communication; The high-magnification eagle-eye camera is used to capture video stream data in real time; The environmental sensor is used to collect air quality data, which includes PM2.5, PM10 concentration and wind speed and wind direction environmental parameters; The NPU module analyzes the location and type of pollution sources in the video stream data through a deep learning algorithm, and combines the results with the environmental sensor data to identify the intensity and distribution of the pollution sources. The NPU module includes an edge computing unit, a data processing and analysis unit, and an intelligent decision control unit, wherein the electrical signals between the units are connected; The edge computing unit performs local data processing based on video stream data by running a deep learning model to identify pollution sources in real time. In the edge computing unit, the real-time identification process of pollution sources includes: Extract environmental image data of the monitored area from the public database, use image annotation tools to annotate the pollution source images, including the pollution source types of smoke, dust, and emissions, and divide the pollution source image data set into training and test sets; Use the training set data to train the deep learning model and the pollution source identification model, and use the test set data to evaluate the pollution source identification model, and adjust the model parameters according to the evaluation results; Deploy the trained pollution source identification model to the NPU, load the preprocessed video stream data to the edge computing unit, and extract the color, shape, and texture features related to pollution source identification from the preprocessed video stream data; Based on the extracted features, the pollution source identification model is used for real-time analysis, the pollution source matching index is calculated, the location of the pollution source in the video stream is identified, and the identification results are output, classified, and the pollution type is determined; The calculation expression of the pollution source matching index is: ； in, is the pollution source matching index, is the color feature matching score, is the shape feature matching score, is the texture feature matching score, is a constant used to adjust the sensitivity of the matching index. is the benchmark score of the color feature, which is used to normalize the color feature matching degree. The value range is between 0 and 1. When , it means no pollution source is detected. When , it means that the detected pollution source characteristics completely match the known pollution source; The data processing and analysis unit identifies the intensity and distribution of the pollution source based on the pollution source identification result of the edge computing unit and combines the result with the environmental sensor data. In the data processing and analysis unit, the identification process of the intensity and distribution of the pollution source includes: Receive the pollution source identification results sent by the edge computing unit, including the location, type and pollution source matching index of the pollution source, extract the real-time PM2.5 concentration and PM10 concentration data from the air quality data, and traverse the historical database to determine the baseline concentration value and maximum concentration value of PM2.5 concentration and PM10 concentration; The pollution source matching index is combined with PM2.5 and PM10 concentration-related data to calculate the pollution level evaluation index and further analyze the pollution level of the pollution source; Extract real-time wind speed and wind direction data from air quality data, analyze the impact of wind speed and wind direction on pollutant diffusion, determine the propagation path of pollutants, and calculate the pollution source distribution coefficient by combining the pollution degree evaluation index and wind speed and wind direction data; According to the pollution source distribution coefficient, determine the diffusion direction and speed of pollutants, reflect the spatial distribution trend of pollution sources, and use geographic information system tools to visualize the diffusion path and potential impact area of ​​pollutants; The calculation expression of the pollution degree evaluation index is: ； in, is the pollution degree evaluation index, is the pollution source matching index, For real-time monitoring of PM2.5 concentration, For real-time monitoring of PM10 concentration, is the baseline concentration value of PM2.5, is the baseline concentration value of PM10, is the maximum concentration value of PM2.5, which is used to calculate the exponential decay term. is the maximum concentration value of PM10, which is used to calculate the exponential decay term. The value range is between 0 and 1. When , it means no pollution source is detected. When , it means that the detected pollution source characteristics completely match the known pollution source; The calculation expression of the pollution source distribution coefficient is: ； in, is the pollution source distribution coefficient, is the pollution degree evaluation index, is the wind speed, which is used to measure the diffusion rate of pollutants. is the reference wind speed value, is the angle between wind direction and pollutant diffusion direction, is the maximum wind direction angle, is the angle between the wind direction and due north, The value range is between 0 and 1. When , it means that there is no pollution diffusion, when When , it means that the pollutant diffusion conditions are optimal, the pollution level is high, and the wind speed and direction are most conducive to the diffusion of pollutants; The intelligent decision control unit dynamically adjusts the spray angle, intensity and coverage of the fog cannon based on air quality data and pollution source analysis results using fuzzy control and PID control algorithms; The fog cannon equipment linkage control module automatically adjusts the spraying strategy of the fog cannon equipment according to the output of the NPU module; The IoT gateway is responsible for data transmission and remote control.

2. According to claim 1, the intelligent linkage control system of the fog cannon Hawkeye is characterized in that: The process of collecting the video stream data and the air quality data includes: A high-magnification Hawkeye camera is installed in the monitoring area, the high-magnification Hawkeye camera is started, real-time monitoring of pollution sources in the area, collecting high-resolution video streams, and connecting the video stream data to the central control system through the Internet of Things gateway; Using the deployed PM2.5 concentration sensor, PM10 concentration sensor, wind speed sensor and wind direction sensor, real-time air quality data including PM2.5 and PM10 concentrations and wind speed and wind direction environmental parameters are collected, and the air quality data is connected to the central control system through the IoT gateway; Verify and correct the accessed video stream data and air quality data in the central control system, and preprocess the data of the pollution source data set, wherein the preprocessing operation includes data cleaning and data format conversion steps; The preprocessed video stream data is integrated with the air quality data to form a complete pollution source data set.

3. According to claim 2, a fog cannon Hawkeye intelligent linkage control system is characterized in that: In the intelligent decision control unit, the process of dynamically adjusting the spray angle, intensity and coverage of the fog cannon includes: Integrate real-time air quality data and pollution source analysis results, including the location, type, pollution level, pollutant diffusion direction and speed of the pollution source, and then preset the initial values ​​of the spray angle, intensity and coverage of the fog cannon equipment according to environmental conditions and pollution level; Design fuzzy control rules to convert qualitative air quality data and pollution source analysis results into fuzzy logic inputs for fog cannon equipment operation, calculate preliminary spray parameters of fog cannon equipment such as spray angle, intensity and coverage based on real-time data, and fine-tune preliminary spray parameters of fog cannon equipment in combination with PID control algorithm; The output results of the intelligent decision-making control unit are converted into specific parameters of the spray angle, intensity and coverage range of the fog cannon, and a control command is generated and sent to the linkage control module of the fog cannon equipment.

4. The intelligent linkage control system of the fog cannon Hawkeye according to claim 3 is characterized in that: In the fog cannon equipment linkage control module, the process of adjusting the spraying strategy of the fog cannon equipment includes: Receive specific parameters of the spray angle, intensity and coverage of the fog cannon equipment from the intelligent decision control module, and parse the received control instructions and convert them into specific parameters that the fog cannon equipment understands and executes; According to the analyzed parameters, the linkage control module of the fog cannon equipment automatically adjusts the state of the fog cannon equipment. After the adjustment is completed, the fog cannon equipment performs fixed-point and quantitative spraying operations according to the preset spraying strategy; The fog cannon equipment linkage control module monitors the operating status and spraying effect of the fog cannon equipment in real time during the spraying process, and feeds back the relevant information to the intelligent decision-making control module for further optimization and adjustment.

Images