An on-line automatic monitoring system and method for carbon dioxide gas concentration in soil

Abstract

The present application relates to the field of gas detection, and particularly relates to a soil carbon dioxide gas concentration online automatic monitoring system and method. A CNN convolutional neural network model is established, carbon dioxide gas data is input into the CNN convolutional neural network model for training, and carbon dioxide gas concentration data is obtained. Whether the carbon dioxide gas concentration data is greater than a warning threshold is determined. If the carbon dioxide gas concentration data is greater than the warning threshold, real-time carbon dioxide gas data is input into the CNN convolutional neural network model for identification, carbon dioxide is transmitted to the soil through a carbon dioxide releasing device. Real-time monitoring of carbon dioxide gas data in the soil is performed to obtain target carbon dioxide gas concentration data. When the target carbon dioxide gas concentration data meets a preset standard value, transmission of carbon dioxide to the soil by the carbon dioxide releasing device is stopped. Real-time monitoring of carbon dioxide gas concentration data in the soil can be performed.

Description

TECHNICAL FIELD

[0001] The application relates to the field of gas detection, in particular to an online automatic monitoring system and method for carbon dioxide gas concentration in soil. BACKGROUND

[0002] Carbon dioxide is an important raw material for plants to carry out photosynthesis, and through photosynthesis, carbon dioxide, water and light energy can be converted into organic matter and oxygen. With the increase of carbon dioxide, the photosynthetic efficiency and nutrient utilization rate of plants will also be improved, so the concentration of carbon dioxide in soil is usually related to the production of plants, but the concentration of carbon dioxide in soil is usually detected by a sensor, but the detection of the sensor is usually the concentration of carbon dioxide in local soil, and the concentration of a single sensor cannot indicate the total carbon dioxide content in soil, and the concentration of carbon dioxide in soil can be affected by factors such as temperature and illumination, so how to detect the concentration of carbon dioxide in soil and automatically adjust the concentration of carbon dioxide in soil is a technical problem to be solved at present. SUMMARY

[0003] The application aims to solve the above problems, and designs an online automatic monitoring system and method for carbon dioxide gas concentration in soil.

[0004] To achieve the above-mentioned purpose, the technical scheme of the application is further provided in the online automatic monitoring system for carbon dioxide gas concentration in soil, and the online automatic monitoring system comprises the following steps:

[0005] A gas data acquisition module is used to acquire historical carbon dioxide gas data in soil through a carbon dioxide gas sensor, to perform data preprocessing on the historical carbon dioxide gas data, and to obtain training carbon dioxide gas data.

[0006] A first gas data identification module is used to establish a CNN convolutional neural network model, to input the training carbon dioxide gas data into the CNN convolutional neural network model for training, and to obtain carbon dioxide gas concentration data.

[0007] A carbon dioxide neutralization module is used to judge whether the carbon dioxide gas concentration data is greater than a warning threshold, if the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed through a carbon dioxide neutralization device, and real-time carbon dioxide gas data in soil is collected in real time.

[0008] A second gas data identification module is used to input the real-time carbon dioxide gas data into the CNN convolutional neural network model for identification, and to obtain initial carbon dioxide gas concentration data.

[0009] A gas data transmission module is configured to determine whether the initial carbon dioxide gas concentration data meets a preset standard value, and if the initial carbon dioxide gas concentration data meets the preset standard value of the soil, the initial carbon dioxide gas concentration data is transmitted to a display terminal.

[0010] A carbon dioxide supplement module is configured to transmit carbon dioxide to the soil through a carbon dioxide release device if the initial carbon dioxide gas concentration data is less than the preset standard value.

[0011] A gas data monitoring module is configured to monitor the carbon dioxide gas data in the soil in real time to obtain target carbon dioxide gas concentration data, and stop transmitting carbon dioxide to the soil by using the carbon dioxide release device when the target carbon dioxide gas concentration data meets the preset standard value.

[0012] Further, in the online automatic monitoring system, the gas data acquisition module comprises the following sub-modules:

[0013] A sensor sub-module is configured to acquire historical carbon dioxide gas data in the soil by using a carbon dioxide gas sensor, and to perform data preprocessing on the historical carbon dioxide gas data. The carbon dioxide gas sensor comprises at least a solid electrolytic CO2 sensor, an optical fiber CO2 sensor, and a capacitive CO2 sensor.

[0014] A sensor setting sub-module is configured to determine that the carbon dioxide gas sensor is inserted into the soil by 10 cm, and the arrangement density of the carbon dioxide gas sensor is 1 / 2 m 2 ;

[0015] A data determination sub-module is configured to determine that the historical carbon dioxide gas data is electrical signal data and digital signal data of the carbon dioxide gas sensor, and the historical carbon dioxide gas data comprises at least 500 or more electrical signal data of the carbon dioxide gas sensor.

[0016] A data denoising sub-module is configured to perform denoising processing on the historical carbon dioxide gas data by using a wavelet threshold denoising method to obtain denoised historical carbon dioxide gas data.

[0017] An obtaining sub-module is configured to randomly extract 80% of the historical carbon dioxide gas data to obtain training carbon dioxide gas data.

[0018] Further, in the online automatic monitoring system, the first gas data recognition module comprises the following units:

[0019] A model establishing unit is configured to establish a CNN convolutional neural network model, and to optimize the CNN convolutional neural network model by using a back propagation algorithm.

[0020] A model training unit is configured to keep the model weight unchanged after training the training carbon dioxide gas data in the CNN convolutional neural network model.

[0021] A back propagation calculation unit is configured to calculate the predicted value of the CNN convolutional neural network model by using a back propagation algorithm, and to calculate the error of the output layer by comparing the predicted value with the expected output.

[0022] A recursive calculation unit is configured to calculate the local gradient of each neuron of the CNN convolutional neural network model in a recursive manner by using error reverse layer-by-layer propagation, to calculate the change of each weight, and to update the weight of the CNN convolutional neural network model to obtain the carbon dioxide gas concentration data.

[0023] Further, in the online automatic monitoring system, the carbon dioxide neutralization module comprises the following sub-modules:

[0024] A judgment sub-module is configured to judge whether the carbon dioxide gas concentration data is greater than a warning threshold, and the warning threshold at least comprises that the carbon dioxide gas concentration data is greater than 3%.

[0025] A threshold sub-module is configured to determine that if the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed by the carbon dioxide neutralization device, and when the carbon dioxide gas concentration data is 3%-5%, the spraying concentration of calcium hydroxide is 6%, and 100 ml is sprayed per square meter.

[0026] A numerical sub-module is configured to determine that when the carbon dioxide gas concentration data is greater than 5%, the spraying concentration of calcium hydroxide is 8%, and 200 ml is sprayed per square meter.

[0027] A category sub-module is configured to determine that the carbon dioxide neutralization device at least comprises a calcium hydroxide spraying device, a calcium oxide powder spraying device, and a sodium hydroxide spraying device.

[0028] An identification sub-module is configured to collect real-time carbon dioxide gas data in the soil in real time, input the real-time carbon dioxide gas data into the CNN convolutional neural network model for identification, and obtain initial carbon dioxide gas concentration data.

[0029] Further, in the online automatic monitoring system, the gas data transmission module comprises the following sub-modules:

[0030] A judgment sub-module is configured to judge whether the initial carbon dioxide gas concentration data meets a preset standard value, and the preset standard value is that the initial carbon dioxide gas concentration data is less than 0.5%.

[0031] transmitting sub-module, configured to transmit the initial carbon dioxide gas concentration data to a display terminal if the initial carbon dioxide gas concentration data meets the preset standard value of the soil;

[0032] warning sub-module, configured to transmit the initial carbon dioxide gas concentration data and the spraying data of the carbon dioxide neutralizing device to a server for warning if the initial carbon dioxide gas concentration data does not meet the preset standard value of the soil.

[0033] Further, in the online automatic monitoring system, the carbon dioxide supplement module comprises the following sub-modules:

[0034] a gas releasing unit, configured to determine that carbon dioxide is transmitted to the soil by the carbon dioxide releasing device if the initial carbon dioxide gas concentration data is less than the preset standard value;

[0035] a gas numerical value unit, configured to determine that carbon dioxide gas with a concentration of 3% is released by the carbon dioxide releasing device, and 300 ml per square meter is released when the initial carbon dioxide gas concentration data is 0.2%-0.3%; and carbon dioxide gas with a concentration of 5% is released by the carbon dioxide releasing device, and 400 ml per square meter is released when the initial carbon dioxide gas concentration data is less than 0.2%;

[0036] a releasing arrangement unit, configured to determine that the carbon dioxide releasing device is inserted into the soil by 6 cm, and the arrangement density of the carbon dioxide releasing device is 1 / 4 m 2 .

[0037] Further, in the online automatic monitoring system, the gas data monitoring module comprises the following units:

[0038] a gas data acquisition unit, configured to monitor the carbon dioxide gas data in the soil in real time, acquire the carbon dioxide gas data in the soil by the carbon dioxide sensor every 5 min, and obtain target carbon dioxide gas concentration data;

[0039] a gas data transmission unit, configured to record the carbon dioxide concentration releasing data corresponding to the target carbon dioxide gas concentration data, and input the target carbon dioxide gas concentration data and the carbon dioxide concentration releasing data into a database for storage;

[0040] a gas releasing control unit, configured to stop transmitting carbon dioxide to the soil by the carbon dioxide releasing device when the target carbon dioxide gas concentration data meets the preset standard value.

[0041] The technical scheme of the present application for achieving the above-mentioned purpose is that, further, in the online automatic monitoring method for the carbon dioxide gas concentration in the soil, the online automatic monitoring method comprises:

[0042] a data acquisition module configured to acquire historical seedling image data of seedlings in the mangrove forest through an image acquisition device, and to obtain training mangrove growth state data by performing data preprocessing on the historical seedling image data;

[0043] acquire historical carbon dioxide gas data in the soil through the carbon dioxide gas sensor, and obtain training carbon dioxide gas data by performing data preprocessing on the historical carbon dioxide gas data;

[0044] establish a CNN convolutional neural network model, input the training carbon dioxide gas data into the CNN convolutional neural network model for training, and obtain carbon dioxide gas concentration data;

[0045] determine whether the carbon dioxide gas concentration data is greater than a warning threshold value, and if the carbon dioxide gas concentration data is greater than the warning threshold value, spray calcium hydroxide through a carbon dioxide neutralization device and collect real-time carbon dioxide gas data in the soil in real time;

[0046] input the real-time carbon dioxide gas data into the CNN convolutional neural network model for identification, and obtain initial carbon dioxide gas concentration data;

[0047] determine whether the initial carbon dioxide gas concentration data meets a preset standard value, and if the initial carbon dioxide gas concentration data meets the soil preset standard value, transmit the initial carbon dioxide gas concentration data to a display terminal;

[0048] if the initial carbon dioxide gas concentration data is less than the preset standard value, transmit carbon dioxide to the soil through a carbon dioxide release device;

[0049] monitor the carbon dioxide gas data in the soil in real time to obtain target carbon dioxide gas concentration data; and when the target carbon dioxide gas concentration data meets the preset standard value, stop transmitting carbon dioxide to the soil using the carbon dioxide release device.

[0050] Further, in the above-mentioned method for online automatic monitoring of carbon dioxide gas concentration in soil, the establishment of the CNN convolutional neural network model, the input of the training carbon dioxide gas data into the CNN convolutional neural network model for training, and the obtaining of carbon dioxide gas concentration data, comprise:

[0051] establish a CNN convolutional neural network model, and optimize the CNN convolutional neural network model using a backpropagation algorithm;

[0052] input the training carbon dioxide gas data into the CNN convolutional neural network model for training, and keep the model weight unchanged after the training;

[0053] The prediction value of the CNN convolutional neural network model is calculated by using the back propagation algorithm, and the error of the output layer is calculated by comparing the prediction value with the expected output;

[0054] The local gradient of each neuron of the CNN convolutional neural network model is calculated in a recursive manner by using error reverse layer-by-layer propagation, and the change of each weight value is calculated to update the weight value of the CNN convolutional neural network model, thereby obtaining carbon dioxide gas concentration data.

[0055] Further, in the above-mentioned soil carbon dioxide gas concentration online automatic monitoring method, if the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed through the carbon dioxide neutralization device, and real-time carbon dioxide gas data in the soil is collected in real time, comprising:

[0056] If the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed through the carbon dioxide neutralization device, and real-time carbon dioxide gas data in the soil is collected in real time, comprising:

[0057] If the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed through the carbon dioxide neutralization device, and real-time carbon dioxide gas data in the soil is collected in real time, comprising:

[0058] If the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed through the carbon dioxide neutralization device, and real-time carbon dioxide gas data in the soil is collected in real time, comprising:

[0059] The carbon dioxide neutralization device at least comprises a calcium hydroxide spraying device, a calcium oxide powder spraying device, and a sodium hydroxide spraying device.

[0060] Real-time carbon dioxide gas data in the soil is collected, and the real-time carbon dioxide gas data is input into the CNN convolutional neural network model for identification to obtain initial carbon dioxide gas concentration data.

[0061] Its beneficial effect lies in, through the gas data acquisition module, the historical carbon dioxide gas data in the soil is acquired through the carbon dioxide gas sensor, the historical carbon dioxide gas data is preprocessed to obtain training carbon dioxide gas data; the first gas data identification module is used for establishing CNN convolution neural network model, the training carbon dioxide gas data is input into the CNN convolution neural network model to train, and carbon dioxide gas concentration data is obtained; carbon dioxide neutralization module is used for judging whether the carbon dioxide gas concentration data is greater than the early warning threshold, if the carbon dioxide gas concentration data is greater than the early warning threshold, then through the calcium hydroxide is sprayed by carbon dioxide neutralization device, and real-time carbon dioxide gas data in soil is collected; the second gas data identification module is used for inputting the real-time carbon dioxide gas data into the CNN convolution neural network model to identify, and initial carbon dioxide gas concentration data is obtained; the gas data transmission module is used for judging whether the initial carbon dioxide gas concentration data meets the preset standard value, if the initial carbon dioxide gas concentration data meets the soil preset standard value, the initial carbon dioxide gas concentration data is transmitted to the display terminal; the carbon dioxide supplement module is used for if the initial carbon dioxide gas concentration data is less than the preset standard value, then the carbon dioxide is transmitted to the soil through the carbon dioxide release device; the gas data monitoring module is used for real-time monitoring of carbon dioxide gas data in soil, and target carbon dioxide gas concentration data is obtained; when the target carbon dioxide gas concentration data meets the preset standard value, stop transmitting carbon dioxide to the soil by using the carbon dioxide release device. The carbon dioxide gas concentration data in the soil of a large area can be monitored in real time, the carbon dioxide concentration in the soil of a large area is automatically monitored and warned, the carbon dioxide concentration in the soil of a large area can be effectively automatically adjusted, the management efficiency is improved, and the investment of manpower and material resources is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0062] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments with reference made to the accompanying drawings. The drawings are for purposes of illustration only and are not considered as limiting the application.

[0063] Figure 1 It is a first embodiment schematic diagram of a kind of soil carbon dioxide gas concentration online automatic monitoring system in the embodiment of the present application;

[0064] Figure 2 It is a second embodiment schematic diagram of a kind of soil carbon dioxide gas concentration online automatic monitoring system in the embodiment of the present application;

[0065] Figure 3 It is a third embodiment schematic diagram of a kind of soil carbon dioxide gas concentration online automatic monitoring system in the embodiment of the present application;

[0066] Figure 4 Figure 1 is a schematic diagram of a first embodiment of a method for online automatic monitoring of carbon dioxide gas concentration in soil according to the present application. DETAILED DESCRIPTION

[0067] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0068] Those skilled in the art can understand that the singular forms "a", "an" and "the" used herein include plural forms unless specifically stated otherwise. It should be further understood that the use of the term "include" in the specification of the present application means that the features, integers, steps, operations, elements and / or components are present, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0069] The present application will be specifically described below with reference to the accompanying drawings, such as Figure 1 Figure 1 is a schematic diagram of a system for online automatic monitoring of carbon dioxide gas concentration in soil according to the present application. The online automatic monitoring system includes the following modules:

[0070] A gas data acquisition module is used to acquire historical carbon dioxide gas data in soil through a carbon dioxide gas sensor, and to perform data preprocessing on the historical carbon dioxide gas data to obtain training carbon dioxide gas data. Specifically, the present embodiment further includes a sensor submodule for acquiring historical carbon dioxide gas data in soil through a carbon dioxide gas sensor and performing data preprocessing on the historical carbon dioxide gas data. The carbon dioxide gas sensor at least includes a solid electrolytic CO2 sensor, an optical fiber CO2 sensor and a capacitive CO2 sensor. A sensor setting submodule is used to determine that the carbon dioxide gas sensor is inserted into soil by 10 cm, and the arrangement density of the carbon dioxide gas sensor is 1 per 2 m2. A data determination submodule is used to determine that the historical carbon dioxide gas data is electrical signal data and digital signal data of the carbon dioxide gas sensor. The historical carbon dioxide gas data at least includes more than 500 electrical signal data of the carbon dioxide gas sensor. A data denoising submodule is used to perform denoising processing on the historical carbon dioxide gas data by using a wavelet threshold denoising method to obtain denoised historical carbon dioxide gas data. A obtaining submodule is used to randomly extract 80% of the historical carbon dioxide gas data to obtain training carbon dioxide gas data. 2

[0071] ​The first gas data identification module is configured to establish a CNN convolutional neural network model, input training carbon dioxide gas data into the CNN convolutional neural network model for training, and obtain carbon dioxide gas concentration data.

[0072] Specifically, the embodiment further includes a model establishing unit configured to establish a CNN convolutional neural network model and optimize the CNN convolutional neural network model by using a back propagation algorithm; a model training unit configured to input training carbon dioxide gas data into the CNN convolutional neural network model for training and keep the model weight unchanged; a back propagation calculation unit configured to calculate a predicted value of the CNN convolutional neural network model by using the back propagation algorithm, calculate an error of an output layer by comparing the predicted value with an expected output; and a recursive calculation unit configured to calculate a local gradient of each neuron of the CNN convolutional neural network model in a recursive manner by using the error to propagate back layer by layer, calculate a change amount of each weight, update the weight of the CNN convolutional neural network model, and obtain carbon dioxide gas concentration data.

[0073] The carbon dioxide neutralization module is configured to determine whether the carbon dioxide gas concentration data is greater than a warning threshold value, and if the carbon dioxide gas concentration data is greater than the warning threshold value, spray calcium hydroxide through a carbon dioxide neutralization device and collect real-time carbon dioxide gas data in the soil in real time.

[0074] Specifically, the embodiment further includes a judgment sub-module configured to determine whether the carbon dioxide gas concentration data is greater than a warning threshold value, wherein the warning threshold value at least includes that the carbon dioxide gas concentration data is greater than 3%; a threshold sub-module configured to determine that if the carbon dioxide gas concentration data is greater than the warning threshold value, calcium hydroxide is sprayed through a carbon dioxide neutralization device, the spraying concentration of calcium hydroxide is 6% and 100 ml per square meter when the carbon dioxide gas concentration data is 3%-5%; a numerical sub-module configured to determine that the spraying concentration of calcium hydroxide is 8% and 200 ml per square meter when the carbon dioxide gas concentration data is greater than 5%; a category sub-module configured to determine that the carbon dioxide neutralization device at least includes a calcium hydroxide spraying device, a calcium oxide powder spraying device, and a sodium hydroxide spraying device; and an identification sub-module configured to collect real-time carbon dioxide gas data in the soil in real time, input the real-time carbon dioxide gas data into the CNN convolutional neural network model for identification, and obtain initial carbon dioxide gas concentration data.

[0075] The second gas data identification module is configured to input real-time carbon dioxide gas data into the CNN convolutional neural network model for identification, and obtain initial carbon dioxide gas concentration data.

[0076] The gas data transmission module is configured to determine whether the initial carbon dioxide gas concentration data meets a preset standard value, and if the initial carbon dioxide gas concentration data meets the preset standard value of the soil, transmit the initial carbon dioxide gas concentration data to a display terminal.

[0077] Specifically, the embodiment further includes a determination submodule configured to determine whether the initial carbon dioxide gas concentration data meets a preset standard value, the preset standard value being that the initial carbon dioxide gas concentration data is less than 0.5%; a transmission submodule configured to, if the initial carbon dioxide gas concentration data meets the preset standard value of the soil, transmit the initial carbon dioxide gas concentration data to a display terminal; and a warning submodule configured to, if the initial carbon dioxide gas concentration data does not meet the preset standard value of the soil, transmit the initial carbon dioxide gas concentration data and carbon dioxide neutralization device spraying data to a server for warning.

[0078] The carbon dioxide supplement module is configured to, if the initial carbon dioxide gas concentration data is less than the preset standard value, transmit carbon dioxide to the soil through a carbon dioxide release device.

[0079] Specifically, the embodiment further includes a gas release unit configured to determine, if the initial carbon dioxide gas concentration data is less than the preset standard value, to transmit carbon dioxide to the soil through the carbon dioxide release device; a gas value unit configured to determine, when the initial carbon dioxide gas concentration data is 0.2%-0.3%, to release carbon dioxide gas with a concentration of 3% through the carbon dioxide release device, 300 ml per square meter; and when the initial carbon dioxide gas concentration data is less than 0.2%, to release carbon dioxide gas with a concentration of 5% through the carbon dioxide release device, 400 ml per square meter; a release arrangement unit configured to determine that the carbon dioxide release device is inserted into the soil by 6 cm, and the arrangement density of the carbon dioxide release device is 1 / 4 m 2 .

[0080] The gas data monitoring module is configured to monitor the carbon dioxide gas data in the soil in real time to obtain target carbon dioxide gas concentration data, and stop transmitting carbon dioxide to the soil by using the carbon dioxide release device when the target carbon dioxide gas concentration data meets a preset standard value.

[0081] Specifically, the embodiment further includes a gas data acquisition unit for monitoring carbon dioxide gas data in the soil in real time, obtaining target carbon dioxide gas concentration data by acquiring carbon dioxide gas data in the soil every 5 minutes through a carbon dioxide sensor; a gas data transmission unit for recording carbon dioxide concentration release data corresponding to the target carbon dioxide gas concentration data, inputting the target carbon dioxide gas concentration data and the carbon dioxide concentration release data into a database for storage; and a gas release control unit for stopping transmission of carbon dioxide into the soil by using the carbon dioxide release device when the target carbon dioxide gas concentration data meets a preset standard value.

[0082] The beneficial effects are that the gas data acquisition module is used to acquire historical carbon dioxide gas data in the soil through a carbon dioxide gas sensor, data preprocessing is performed on the historical carbon dioxide gas data to obtain training carbon dioxide gas data; the first gas data identification module is used to establish a CNN convolutional neural network model, input the training carbon dioxide gas data into the CNN convolutional neural network model for training to obtain carbon dioxide gas concentration data; the carbon dioxide neutralization module is used to judge whether the carbon dioxide gas concentration data is greater than a warning threshold value, if the carbon dioxide gas concentration data is greater than the warning threshold value, calcium hydroxide is sprayed through a carbon dioxide neutralization device, and real-time carbon dioxide gas data in the soil is collected in real time; the second gas data identification module is used to input the real-time carbon dioxide gas data into the CNN convolutional neural network model for identification to obtain initial carbon dioxide gas concentration data; the gas data transmission module is used to judge whether the initial carbon dioxide gas concentration data meets a preset standard value, if the initial carbon dioxide gas concentration data meets the soil preset standard value, the initial carbon dioxide gas concentration data is transmitted to a display terminal; the carbon dioxide supplement module is used to transmit carbon dioxide into the soil through a carbon dioxide release device if the initial carbon dioxide gas concentration data is less than the preset standard value; the gas data monitoring module is used to monitor carbon dioxide gas data in the soil in real time to obtain target carbon dioxide gas concentration data; and when the target carbon dioxide gas concentration data meets the preset standard value, transmission of carbon dioxide into the soil by using the carbon dioxide release device is stopped. The carbon dioxide gas concentration data in a large area of soil can be monitored in real time, the carbon dioxide concentration in a large area of soil can be monitored and warned automatically, the carbon dioxide concentration in a large area of soil can be adjusted automatically, and the management efficiency is improved to reduce the investment of manpower and material resources.

[0083] In the embodiment, please refer to Figure 2 The first gas data identification module in the second embodiment of the soil carbon dioxide gas concentration online automatic monitoring system includes the following units:

[0084] The model establishing unit is configured to establish a CNN convolutional neural network model and optimize the CNN convolutional neural network model by using a back propagation algorithm.

[0085] The model training unit is configured to input training carbon dioxide gas data into the CNN convolutional neural network model for training and keep the model weight unchanged after the training.

[0086] The back propagation calculation unit is configured to calculate a predicted value of the CNN convolutional neural network model by using the back propagation algorithm, and calculate an error of an output layer by comparing the predicted value with an expected output.

[0087] The recursive calculation unit is configured to calculate a local gradient of each neuron of the CNN convolutional neural network model in a recursive manner by using the error to propagate reversely layer by layer, calculate a change amount of each weight, update the weight of the CNN convolutional neural network model, and obtain carbon dioxide gas concentration data.

[0088] In this embodiment, please refer to Figure 3 In the third embodiment of the soil carbon dioxide gas concentration online automatic monitoring system in the embodiment of the present application, the carbon dioxide neutralization module comprises the following sub-modules.

[0089] The judgment sub-module is configured to judge whether the carbon dioxide gas concentration data is greater than a warning threshold, and the warning threshold at least comprises that the carbon dioxide gas concentration data is greater than 3%.

[0090] The threshold sub-module is configured to determine that, if the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed by the carbon dioxide neutralization device, and when the carbon dioxide gas concentration data is 3%-5%, the spraying concentration of the calcium hydroxide is 6%, and 100 ml is sprayed per square meter.

[0091] The numerical value sub-module is configured to determine that, when the carbon dioxide gas concentration data is greater than 5%, the spraying concentration of the calcium hydroxide is 8%, and 200 ml is sprayed per square meter.

[0092] The category sub-module is configured to determine that the carbon dioxide neutralization device at least comprises a calcium hydroxide spraying device, a calcium oxide powder spraying device, and a sodium hydroxide spraying device.

[0093] The recognition sub-module is configured to collect real-time carbon dioxide gas data in the soil in real time, input the real-time carbon dioxide gas data into the CNN convolutional neural network model for recognition, and obtain initial carbon dioxide gas concentration data.

[0094] The soil carbon dioxide gas concentration online automatic monitoring system provided in the embodiment of the present application is described above, and the soil carbon dioxide gas concentration online automatic monitoring method provided in the embodiment of the present application is described below. Please refer to Figure 4The online automatic monitoring method in the embodiment of the application comprises the following steps:

[0095] The historical carbon dioxide gas data in the soil is acquired by a carbon dioxide gas sensor, the historical carbon dioxide gas data is preprocessed to obtain training carbon dioxide gas data;

[0096] A CNN convolutional neural network model is established, the training carbon dioxide gas data is input into the CNN convolutional neural network model for training to obtain carbon dioxide gas concentration data;

[0097] It is judged whether the carbon dioxide gas concentration data is greater than a warning threshold value, if the carbon dioxide gas concentration data is greater than the warning threshold value, calcium hydroxide is sprayed by a carbon dioxide neutralizing device, and real-time carbon dioxide gas data in the soil is collected in real time;

[0098] The real-time carbon dioxide gas data is input into the CNN convolutional neural network model for identification to obtain initial carbon dioxide gas concentration data;

[0099] It is judged whether the initial carbon dioxide gas concentration data meets a preset standard value, if the initial carbon dioxide gas concentration data meets the preset standard value of the soil, the initial carbon dioxide gas concentration data is transmitted to a display terminal;

[0100] If the initial carbon dioxide gas concentration data is less than the preset standard value, carbon dioxide is transmitted to the soil by a carbon dioxide releasing device;

[0101] The carbon dioxide gas data in the soil is monitored in real time to obtain target carbon dioxide gas concentration data; when the target carbon dioxide gas concentration data meets the preset standard value, the transmission of carbon dioxide to the soil by the carbon dioxide releasing device is stopped.

[0102] The basic principle, main features and advantages of the application are shown and described above. The skilled in the art should understand from the above, the application is not limited by the above-mentioned embodiments, the above-mentioned embodiments and the description in the specification are only preferred examples of the application, and are not used to limit the application, various modifications and improvements of the application can be made without departing from the spirit and scope of the application, and these modifications and improvements all fall within the scope of the claimed application. The scope of protection of the application is defined by the appended claims and their equivalents.

Claims

1. An online automatic monitoring system for carbon dioxide gas concentration in soil, characterized in that: The online automatic monitoring system includes the following modules: A gas data acquisition module is used to acquire historical carbon dioxide gas data in the soil through a carbon dioxide gas sensor, and perform data preprocessing on the historical carbon dioxide gas data to obtain training carbon dioxide gas data; A first gas data recognition module is used to establish a CNN convolutional neural network model, input the training carbon dioxide gas data into the CNN convolutional neural network model for training, and obtain carbon dioxide gas concentration data; A carbon dioxide neutralization module is used to determine whether the carbon dioxide gas concentration data is greater than a warning threshold. If the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed through a carbon dioxide neutralization device and real-time carbon dioxide gas data in the soil is collected in real time; A second gas data recognition module is used to input the real-time carbon dioxide gas data into the CNN convolutional neural network model for recognition to obtain initial carbon dioxide gas concentration data; A gas data transmission module, configured to determine whether the initial carbon dioxide gas concentration data meets a preset standard value, and if so, transmit the initial carbon dioxide gas concentration data to a display terminal; a carbon dioxide replenishing module, configured to transmit carbon dioxide into the soil through a carbon dioxide releasing device if the initial carbon dioxide gas concentration data does not meet a preset standard value; The gas data monitoring module is used to monitor the carbon dioxide gas data in the soil in real time and obtain the target carbon dioxide gas concentration data; when the target carbon dioxide gas concentration data meets the preset standard value, the carbon dioxide release device is stopped from transmitting carbon dioxide into the soil.

2. The online automatic monitoring system for carbon dioxide gas concentration in soil according to claim 1, characterized in that: The gas data acquisition module includes the following submodules: The sensor submodule is used to obtain historical carbon dioxide gas data in the soil through a carbon dioxide gas sensor and perform data preprocessing on the historical carbon dioxide gas data; the carbon dioxide gas sensor includes at least a solid electrolytic Sensors, optical fibers Sensor, capacitive sensor; The sensor setting submodule is used to determine that the carbon dioxide gas sensor is inserted into the soil 10cm and the arrangement density of the carbon dioxide gas sensor is 1 / ; a data determination submodule, configured to determine that the historical carbon dioxide gas data is electrical signal data and digital signal data of a carbon dioxide gas sensor; the historical carbon dioxide gas data includes electrical signal data of at least 500 or more carbon dioxide gas sensors; a data denoising submodule, configured to perform denoising on the historical carbon dioxide gas data using a wavelet threshold denoising method to obtain denoised historical carbon dioxide gas data; A submodule is obtained, which is used to randomly extract 80% of the historical carbon dioxide gas data to obtain training carbon dioxide gas data.

3. The online automatic monitoring system for carbon dioxide gas concentration in soil according to claim 1, characterized in that: The first gas data identification module includes the following units: A model building unit, used to build a CNN convolutional neural network model and optimize the CNN convolutional neural network model using a back propagation algorithm; A model training unit, configured to input the training carbon dioxide gas data into the CNN convolutional neural network model for training and keep the model weights unchanged; A back propagation calculation unit is used to calculate the predicted value of the CNN convolutional neural network model using a back propagation algorithm, and calculate the error of the output layer by comparing the predicted value with the expected output; The recursive calculation unit is used to use the error to propagate layer by layer in reverse order to recursively calculate the local gradient of each neuron of the CNN convolutional neural network model, calculate the change of each weight, update the weight of the CNN convolutional neural network model, and obtain carbon dioxide gas concentration data.

4. The online automatic monitoring system for carbon dioxide gas concentration in soil according to claim 1, characterized in that: The carbon dioxide neutralization module includes the following submodules: A judgment submodule is used to judge whether the carbon dioxide gas concentration data is greater than a warning threshold, and the warning threshold is 3%; The threshold submodule is used to determine whether to spray calcium hydroxide through the carbon dioxide neutralization device if the carbon dioxide gas concentration data is greater than the warning threshold. When the carbon dioxide gas concentration data is 3%-5%, the spraying concentration of calcium hydroxide is 6%, and 100 ml is sprayed per square meter. The numerical submodule is used to determine that when the carbon dioxide gas concentration data is greater than 5%, the spraying concentration of calcium hydroxide is 8%, and 200 ml is sprayed per square meter; A category submodule, configured to determine whether the carbon dioxide neutralization device comprises at least a calcium hydroxide spraying device, a calcium oxide powder spraying device, and a sodium hydroxide spraying device; Identification submodule, used to collect real-time carbon dioxide gas data in the soil.

5. The online automatic monitoring system for carbon dioxide gas concentration in soil according to claim 1, characterized in that: The gas data transmission module includes the following submodules: A judgment submodule, configured to judge whether the initial carbon dioxide gas concentration data meets a preset standard value, wherein the preset standard value is that the initial carbon dioxide gas concentration data is greater than 0.5%; a transmission submodule, configured to transmit the initial carbon dioxide gas concentration data to a display terminal if the initial carbon dioxide gas concentration data meets a preset standard value; The early warning submodule is used to transmit the initial carbon dioxide gas concentration data and the carbon dioxide neutralization device spraying data to the server for early warning if the initial carbon dioxide gas concentration data does not meet the preset standard value.

6. The online automatic monitoring system for carbon dioxide gas concentration in soil according to claim 1, characterized in that: The carbon dioxide supplement module includes the following submodules: a gas release unit, configured to transmit carbon dioxide into the soil through a carbon dioxide release device if the initial carbon dioxide gas concentration data does not meet a preset standard value; A gas value unit, used to determine that when the initial carbon dioxide gas concentration data is 0.2%-0.3%, a carbon dioxide gas with a concentration of 3% is released through the carbon dioxide release device, and 300 ml is released per square meter; when the initial carbon dioxide gas concentration data is less than 0.2%, a carbon dioxide gas with a concentration of 5% is released through the carbon dioxide release device, and 400 ml is released per square meter; The release arrangement unit is used to determine that the carbon dioxide release device is inserted into the soil 6cm, and the arrangement density of the carbon dioxide release device is 1 / .

7. The online automatic monitoring system for carbon dioxide gas concentration in soil according to claim 1, characterized in that: The gas data monitoring module includes the following units: The gas data acquisition unit is used to monitor the carbon dioxide gas data in the soil in real time. The carbon dioxide gas data in the soil is acquired every 5 minutes through the carbon dioxide sensor to obtain the target carbon dioxide gas concentration data; a gas data transmission unit, configured to record the carbon dioxide concentration release data corresponding to the target carbon dioxide gas concentration data, and input the target carbon dioxide gas concentration data and the carbon dioxide concentration release data into a database for storage; The gas release control unit is used to stop using the carbon dioxide release device to transmit carbon dioxide into the soil when the target carbon dioxide gas concentration data meets the preset standard value.

8. Implementing the method for online automatic monitoring of carbon dioxide gas concentration in soil as claimed in claim 1, characterized in that: The method for online automatic monitoring of carbon dioxide gas concentration comprises the following steps: Acquire historical carbon dioxide gas data in the soil through a carbon dioxide gas sensor, and perform data preprocessing on the historical carbon dioxide gas data to obtain training carbon dioxide gas data; Establishing a CNN convolutional neural network model, inputting the training carbon dioxide gas data into the CNN convolutional neural network model for training to obtain carbon dioxide gas concentration data; Determining whether the carbon dioxide gas concentration data is greater than a warning threshold; if the carbon dioxide gas concentration data is greater than the warning threshold, spraying calcium hydroxide through a carbon dioxide neutralization device and collecting real-time carbon dioxide gas data in the soil in real time; Inputting the real-time carbon dioxide gas data into the CNN convolutional neural network model for recognition to obtain initial carbon dioxide gas concentration data; determining whether the initial carbon dioxide gas concentration data meets a preset standard value, and if the initial carbon dioxide gas concentration data meets the preset standard value, transmitting the initial carbon dioxide gas concentration data to a display terminal; If the initial carbon dioxide gas concentration data does not meet the preset standard value, carbon dioxide is transmitted to the soil through the carbon dioxide release device; The carbon dioxide gas data in the soil is monitored in real time to obtain target carbon dioxide gas concentration data; when the target carbon dioxide gas concentration data meets the preset standard value, the carbon dioxide release device is stopped from transmitting carbon dioxide into the soil.

9. The method for online automatic monitoring of carbon dioxide gas concentration in soil according to claim 8, characterized in that: The method of establishing a CNN convolutional neural network model and inputting the training carbon dioxide gas data into the CNN convolutional neural network model for training to obtain carbon dioxide gas concentration data includes: Establishing a CNN convolutional neural network model and optimizing the CNN convolutional neural network model using a back propagation algorithm; Inputting the training carbon dioxide gas data into the CNN convolutional neural network model for training and keeping the model weights unchanged; Calculate the predicted value of the CNN convolutional neural network model using the back propagation algorithm, and calculate the error of the output layer by comparing the predicted value with the expected output; The error is propagated layer by layer in reverse order to recursively calculate the local gradient of each neuron in the CNN convolutional neural network model, calculate the change in each weight, update the weight of the CNN convolutional neural network model, and obtain carbon dioxide gas concentration data.

10. The method for online automatic monitoring of carbon dioxide gas concentration in soil according to claim 8, characterized in that: The determining whether the carbon dioxide gas concentration data is greater than a warning threshold, and if the carbon dioxide gas concentration data is greater than the warning threshold, spraying calcium hydroxide through a carbon dioxide neutralization device and collecting real-time carbon dioxide gas data in the soil in real time, includes: Determine whether the carbon dioxide gas concentration data is greater than a warning threshold, the warning threshold being 3%; If the carbon dioxide gas concentration data is greater than the warning threshold, calcium hydroxide is sprayed through the carbon dioxide neutralization device. When the carbon dioxide gas concentration data is 3%-5%, the spraying concentration of calcium hydroxide is 6%, and 100 ml is sprayed per square meter; When the carbon dioxide gas concentration is greater than 5%, the spray concentration of calcium hydroxide is 8%, spraying 200ml per square meter; The carbon dioxide neutralization device at least includes a calcium hydroxide spraying device, a calcium oxide powder spraying device, and a sodium hydroxide spraying device.

Images