Carbon dioxide emission source itinerant monitoring method based on NDIR technology

Abstract

The invention discloses a carbon dioxide emission source itinerant monitoring method based on an NDIR technology, and relates to the technical field of environment monitoring. The method comprises the following steps: a detection vehicle collects carbon dioxide concentration data of each collection point along a planned path; comparing the carbon dioxide concentration data with a set carbon dioxide concentration threshold value, and performing grading treatment; when the carbon dioxide concentration is abnormal, other detection vehicles are cooperated for verification; according to the method, the problems of fixed monitoring points, untimely response and the like in the existing scheme can be effectively solved, and the method is suitable for the requirement of accurately and intelligently monitoring the carbon dioxide concentration in a complex environment.

Description

Technical Field

[0001] The invention relates to the technical field of environmental monitoring, and in particular to a carbon dioxide emission source itinerant monitoring method based on NDIR technology. Background Art

[0002] In today's environmental monitoring field, the traditional way of monitoring carbon dioxide has many limitations. The traditional method is too rigid and can only monitor the carbon dioxide concentration around the sensor. When the monitoring area is large, a sensor needs to be placed at a certain distance, which undoubtedly greatly increases the cost. NDIR technology (Non-Dispersive InfraRed) is a method based on gas absorption theory for detecting and analyzing gas components. Its working principle is to determine the concentration of the gas by measuring the absorption characteristics of specific gas molecules to infrared light of a specific wavelength. The present invention proposes a carbon dioxide emission source patrol monitoring method based on NDIR technology, which aims to achieve efficient and accurate monitoring of carbon dioxide concentration in a specified area, and can quickly locate carbon dioxide emission sources, providing strong support for carbon emission management. Summary of the invention

[0003] Purpose of the invention: The present invention provides a method for patrol monitoring of carbon dioxide emission sources based on NDIR technology, which has the advantages of efficient and accurate monitoring of carbon dioxide concentration in a specified area and rapid positioning of carbon dioxide emission sources, etc., solving the problem that traditional methods of monitoring carbon dioxide have many limitations and that traditional methods are too rigid.

[0004] To achieve the above object, the present invention provides the following technical solution: a carbon dioxide emission source patrol monitoring method based on NDIR technology, comprising the following steps: Step 1: The detection vehicle collects carbon dioxide concentration data at each collection point along the planned route; Step 2: Compare the carbon dioxide concentration data with the set carbon dioxide concentration threshold and perform graded processing; Step 3: When the carbon dioxide concentration is abnormal, cooperate with other detection vehicles to verify: When a detection vehicle finds an abnormality, the detection vehicle transmits its own position to the PC. The PC divides the two-dimensional plane where the detection vehicle moves into four quadrants, sets the detection vehicle as the main node, and searches for other detection vehicles closest to it in the four different quadrants as sub-nodes. Then, according to the shape and size of the search area, the search area is divided into five sub-areas. The main node detection vehicle is responsible for searching the central area, and the four sub-node detection vehicles are responsible for the peripheral sub-areas corresponding to the four quadrants. The data collected by multiple detection vehicles are transmitted to the PC, and the PC performs time series analysis on the data. Step 4: Analyze the relationship between carbon dioxide concentration and other environmental factors.

[0005] Furthermore, when the carbon dioxide concentration collected in step 2 is lower than 1 / 2 of the threshold, a first-level warning is triggered, and the inspection vehicle continues to patrol normally; when the concentration is between 1 / 2 of the threshold and the threshold, a second-level warning is triggered, the location is recorded, and the patrol frequency and abnormality level are adjusted; when the concentration is higher than the threshold, a third-level warning is triggered, and the inspection vehicle starts the camera to search for emission sources.

[0006] Furthermore, the YOLO algorithm with fuzzy strategy is used to search for emission sources: the collected images are preprocessed first; a fuzzy rule base is constructed according to the pre-set target features, such as color, shape, size, etc.; fuzzy reasoning is performed based on the input features and the fuzzy rule base to generate fuzzy decisions, and uncertainty is incorporated into the target detection process to improve the recognition ability of fuzzy targets; YOLOv5 is used as the basic network architecture, which includes a series of convolutional layers, pooling layers, and fully connected layers for feature extraction and target detection; fuzzy reasoning is integrated with YOLO output, and the result of fuzzy reasoning is integrated with the output of the YOLOv5 network, and the final target detection result is obtained through weight adjustment.

[0007] Furthermore, in step 4, the inspection vehicle is equipped with an NDIR sensor, a temperature and humidity sensor, a wind speed sensor, and an air pressure sensor, and a consistent sampling frequency is set to transmit the data of each sensor to the PC in real time, and the PC performs data preprocessing and analysis.

[0008] Compared with the prior art, the present invention has the following beneficial effects: The concentration classification detection mechanism improves the detection efficiency and quickly locates the abnormal area; the four-quadrant collaborative positioning shortens the traceability time and reduces the search area of ​​the abnormal area; the visual positioning with fuzzy strategy enhances the recognition robustness and can achieve accurate positioning in various complex environments. Compared with the traditional method of only monitoring the carbon dioxide concentration, the algorithm proposed in this invention can not only monitor the carbon dioxide concentration, but also find the emission source. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 This is a flow chart of a carbon dioxide emission source patrol monitoring method based on NDIR technology proposed in the present invention. DETAILED DESCRIPTION

[0010] The NDIR sensor is installed at the front of the inspection vehicle, with the air intake facing the direction of the vehicle's movement; the temperature and humidity sensors are installed on both sides of the inspection vehicle; the camera is placed in the middle of the inspection vehicle through the steering gear, the Beidou-GPS positioning module is installed at the rear of the inspection vehicle, the 5G communication module is installed in the middle of the vehicle body, the WIFI module is installed at the right rear of the vehicle body, and the connection of each module is ensured to be stable and reliable. The temperature and humidity sensor is installed on the left side of the vehicle body, and the wind speed sensor and air pressure sensor are installed on the right side of the vehicle body. It starts when the carbon dioxide concentration exceeds the threshold, uses the 360° steering gear to find the emission source, and transmits the results or panoramic photos to the PC.

[0011] Step 1: Set the location of the collection points in the planned path in advance, and the detection vehicle collects carbon dioxide concentration data at each collection point along the planned path.

[0012] Step 2: The CO2 concentration data is compared with the set CO2 concentration threshold and graded. The system defines three warning states: level 1, level 2 and level 3, corresponding to different CO2 concentration ranges and response measures.

[0013] Level 1 warning state: When the monitored carbon dioxide concentration is lower than 1 / 2 of the set threshold, the system is in the level 1 warning state, and the detection vehicle monitors according to the normal patrol route.

[0014] Level 2 warning state: When the monitored carbon dioxide concentration is greater than 1 / 2 of the set threshold but less than the set threshold, the system enters the level 2 warning state, and the detection vehicle will record the location information. When it patrols the location next time, the system will focus on monitoring the concentration of the recorded location: if the concentration is lower than 1 / 2 of the threshold, the level 1 warning state will be restored; if the concentration is still higher than 1 / 2 of the threshold, the warning state will remain at the level 2 warning state, and the number of patrols in the area will be increased.

[0015] Level 3 warning state: When the monitored carbon dioxide concentration exceeds the set threshold, the system enters the level 3 warning state, the detection vehicle records the location information, the camera is activated, and the YOLO algorithm with fuzzy strategy is used to search for emission sources.

[0016] In order to improve the accuracy of target detection, the present invention introduces the YOLO algorithm based on fuzzy strategy, first pre-processes the collected image; builds a fuzzy rule base according to the target features set in advance, such as color, shape, size, etc.; performs fuzzy reasoning based on the input features and the fuzzy rule base, generates fuzzy decisions, incorporates uncertainty into the target detection process, and improves the recognition ability of fuzzy targets; adopts YOLOv5 as the basic network architecture, which includes a series of convolutional layers, pooling layers and fully connected layers for feature extraction and target detection; integrates fuzzy reasoning with YOLO output, integrates the result of fuzzy reasoning with the output of the YOLOv5 network, and obtains the final target detection result through weight adjustment and other methods.

[0017] In the third-level warning state, the system not only records location information, but also captures images of the surrounding environment through the camera, and uses the YOLO algorithm with fuzzy strategy to find possible emission sources. If the algorithm successfully identifies the emission source, the system will upload the type of emission source and the current carbon dioxide concentration to the PC; if the emission source cannot be identified, the system will capture a 360-degree panoramic image of the area and upload it to the PC.

[0018] Step 3: When the carbon dioxide concentration is abnormal, cooperate with other detection vehicles to verify. The detection vehicle monitors the carbon dioxide concentration in real time during driving. When the monitoring value exceeds the preset threshold, the location information is sent to the PC. After the PC receives the location information of the abnormal detection vehicle, it divides the two-dimensional plane of the detection vehicle into four quadrants. Determine the main node and sub-nodes, set the abnormal detection vehicle as the main node, and find the nearest detection vehicle in four different quadrants as the sub-node. The PC divides the square search area with a side length of 1,000 meters into five sub-areas according to the size of the search area. The main node detection vehicle is responsible for searching the square area with a central side length of 200 meters, and the four sub-node detection vehicles are responsible for searching four isosceles right triangle areas with a side length of 400 meters. The detection vehicle searches for emission sources in the area divided by the PC. The PC performs time series analysis on the collected data to identify high emission areas and their changing trends.

[0019] Step 4: Analyze the relationship between carbon dioxide concentration and other environmental factors: The inspection vehicle is equipped with NDIR sensors, temperature and humidity sensors, wind speed sensors, and air pressure sensors. A consistent sampling frequency is set, and the data acquisition system is used to transmit the data of each sensor to the PC in real time. The PC is responsible for checking the data and finding a small number of missing values, which are filled by interpolation; identifying and eliminating obviously abnormal data points. Using the z-score standardization method, for the collected raw data X, its standardized value The calculation is as follows: ,in is the mean value of the environmental factor, The standardized carbon dioxide concentration is used as the dependent variable, and temperature, humidity, wind speed, and air pressure are used as independent variables to construct a multiple linear regression model: ,in is the standardized carbon dioxide concentration, is the standardized temperature, is the standardized humidity, is the normalized wind speed, is the standardized air pressure, is the random error term, is the regression coefficient corresponding to each environmental factor, and the regression coefficient is estimated using the least squares method. Calculate the coefficient of determination : , where m is the number of samples, y i is the true value of carbon dioxide concentration, is the carbon dioxide concentration value predicted by the regression model, is the mean of the observed values ​​of carbon dioxide concentration, and the coefficient of determination is used to evaluate the goodness of fit of the multiple linear regression model.

[0020] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A carbon dioxide emission source patrol monitoring method based on NDIR technology, characterized in that: The following steps are involved: Step 1: The detection vehicle collects carbon dioxide concentration data at each collection point along the planned route; Step 2: Compare the carbon dioxide concentration data with the set carbon dioxide concentration threshold and perform graded processing; Step 3: When the carbon dioxide concentration is abnormal, cooperate with other detection vehicles to verify: When a detection vehicle finds an abnormality, the detection vehicle transmits its own position to the PC. The PC divides the two-dimensional plane where the detection vehicle moves into four quadrants, sets the detection vehicle as the main node, and searches for other detection vehicles closest to it in the four different quadrants as sub-nodes. Then, according to the shape and size of the search area, the search area is divided into five sub-areas. The main node detection vehicle is responsible for searching the central area, and the four sub-node detection vehicles are responsible for the peripheral sub-areas corresponding to the four quadrants. The data collected by multiple detection vehicles are transmitted to the PC, and the PC performs time series analysis on the data. Step 4: Analyze the relationship between carbon dioxide concentration and other environmental factors.

2. According to claim 1, a carbon dioxide emission source patrol monitoring method based on NDIR technology is characterized in that: When the carbon dioxide concentration collected in the step 2 is lower than 1 / 2 of the threshold, a first-level warning is triggered, and the inspection vehicle continues to patrol normally; when the concentration is between 1 / 2 of the threshold and the threshold, a second-level warning is triggered, the location is recorded, and the patrol frequency and abnormality level are adjusted; when the concentration is higher than the threshold, a third-level warning is triggered, and the inspection vehicle starts the camera to search for emission sources.

3. The method for monitoring carbon dioxide emission sources based on NDIR technology according to claim 2 is characterized in that: The YOLO algorithm with fuzzy strategy is used to find emission sources: the collected images are preprocessed first; a fuzzy rule base is built according to the pre-set target object features; fuzzy reasoning is performed based on the input features and the fuzzy rule base to generate fuzzy decisions, and YOLOv5 is used as the basic network architecture for feature extraction and target detection; fuzzy reasoning is integrated with YOLO output, and the results of fuzzy reasoning are integrated with the output of the YOLOv5 network, and the final target detection result is obtained through weight adjustment.

4. The method for monitoring carbon dioxide emission sources based on NDIR technology according to claim 1 is characterized in that: In step 4, the detection vehicle is equipped with an NDIR sensor, a temperature and humidity sensor, a wind speed sensor, and an air pressure sensor, and a consistent sampling frequency is set to transmit the data of each sensor to the PC in real time, and the PC performs data preprocessing and analysis.

Images