Agricultural monitoring system based on deep learning technology

Abstract

The invention relates to an agricultural monitoring system based on a deep learning technology, which is applied to the technical field of agricultural monitoring and comprises a data acquisition center, a data transmission module and a data processing center, the agricultural monitoring system can judge whether dust on crop leaves needs to be cleaned or not based on the technologies of image data acquisition, deep learning, image processing and the like, and can give an alarm and remind when the dust needs to be cleaned, so that related personnel are prompted to clean the dust on the crop leaves in time; therefore, influence on growth, development, yield and quality of crops due to excessive dust on the leaves can be effectively avoided, healthy growth of the crops and agricultural development can be promoted, whether the risk of excessive dust on the leaves of the crops exists or not is judged according to the environmental condition before image data acquisition, and if the risk exists, the image data is acquired. Therefore, the intelligence and the reliability of the system are greatly improved.

Description

Technical Field

[0001] The present invention relates to an agricultural monitoring system, and in particular to an agricultural monitoring system based on deep learning technology applied in the field of agricultural monitoring technology. Background Art

[0002] As a key link in the agricultural production management process, agricultural monitoring is closely linked to all aspects of agriculture and plays an indispensable role in the health, stability and sustainable development of agriculture. Traditional agricultural monitoring mainly relies on manual inspections, which is not only time-consuming and labor-intensive, but also difficult to achieve real-time, comprehensive and accurate monitoring. With the advancement of science and technology, agriculture is gradually developing towards intelligence and automation, and intelligent and automated monitoring is also beginning to gradually replace manual inspections.

[0003] The invention patent with publication number CN117972337A discloses a method for monitoring and predicting agricultural meteorological disasters based on multimodal deep learning. The invention can integrate various types of data, including meteorological data, satellite images, soil information, etc., and comprehensively consider the impact of various factors on agricultural meteorological disasters, thereby improving the accuracy of prediction. By utilizing the advantages of deep learning technology, it can learn the associations and features between different modal data. When processing new data, only fine-tuning is required to obtain the prediction results, which is conducive to accelerating the training process.

[0004] The invention patent with publication number CN117974348B discloses a smart agricultural Internet of Things monitoring system, which solves the problem that it is difficult to effectively monitor and manage the influencing factors according to the preferences of the crops due to the unknown response of the crops to the influencing factors of their growth.

[0005] The air in the farmland environment inevitably contains dust, especially in the farmland on both sides of the road. Dust adheres to the leaves of crops, which will affect the photosynthesis and transpiration of crops, and thus affect the growth, development, yield and quality of crops. Although the agricultural monitoring system in the existing technology can realize intelligent and automated agricultural monitoring to a certain extent, it generally does not have the function of monitoring the amount of dust on the leaves of crops, so it is impossible to remind or alarm when there is too much dust on the leaves, which can easily lead to the growth, development, yield and other aspects of crops being affected by the excessive amount of dust on the leaves. Therefore, we propose an agricultural monitoring system based on deep learning technology. Summary of the invention

[0006] In view of the above-mentioned prior art, the technical problem to be solved by the present invention is that the agricultural monitoring system in the prior art is unable to remind or alarm when there is too much dust on the leaves, which can easily lead to the growth, development and yield of crops being affected by the excessive amount of dust on the leaves.

[0007] To solve the above problems, the present invention provides an agricultural monitoring system based on deep learning technology, including a data acquisition center, a data transmission module, and a data processing center. The data acquisition center includes a data acquisition module, the data acquisition module includes an image acquisition unit, and the data processing center includes a data storage module, an analysis and alarm module, an intelligent alarm module, and a model training module based on deep learning technology;

[0008] The image acquisition unit is used to collect image data of crop leaves;

[0009] The data transmission module is used to transmit the image data collected by the image acquisition unit to the data processing center;

[0010] The data storage module is used to store the data collected by the data collection center;

[0011] The model training module is used to build and train deep learning models;

[0012] The intelligent alarm module is used for alarm;

[0013] The analysis and control alarm module has an image processing function, which is used to call the trained deep learning model to analyze the image data to determine whether the dust on the crop leaves needs to be cleaned. Each time the crop leaf image data is collected, the analysis and control alarm module will make a judgment. When the judgment result is yes, the analysis and control alarm module will control the intelligent alarm module to alarm the relevant personnel.

[0014] In the above-mentioned agricultural monitoring system based on deep learning technology, it is possible to judge whether the dust on the leaves of crops needs to be cleaned based on image data acquisition, deep learning, image processing and other technologies, and to issue alarms and reminders when the dust needs to be cleaned.

[0015] As a further improvement of the present application, the data acquisition center further includes an acquisition control module, the data acquisition module further includes an environmental monitoring unit, and the acquisition control module includes an acquisition setting unit, an analysis and control unit, and a timing feedback module;

[0016] The data acquisition module collects image data of crop leaves through a mapping drone. The mapping drone consists of a drone body and a camera mounted on the drone body. The environmental monitoring unit is used to collect environmental data around the farmland in real time. The environmental data includes dust concentration data and meteorological data, and the data collected by the environmental monitoring unit will be transmitted to the analysis and control unit in real time. The collection setting unit is used to set the collection path and collection points of the mapping drone. The analysis and control unit is used to control the mapping drone to perform image data collection. The timing feedback module is used to perform timing operations under the control of the analysis and control unit. There are two modes in which the analysis and control unit controls the mapping drone to perform image data collection, namely, the resident judgment mode and the close injection weekly collection mode.

[0017] As a further improvement of the present application, the collection setting unit is also used to set a concentration threshold and a time threshold, the time threshold includes a rain interval time threshold and a concentration exceeding time threshold, and the analysis and control unit is also used to analyze the environmental data collected by the environmental monitoring unit according to the concentration threshold and the time threshold to determine whether image data collection is needed. When the judgment result is yes, the analysis and control unit will control the mapping drone to perform image data collection according to the set collection path and collection points.

[0018] As a further improvement of the present application, in the resident judgment and sampling mode, after each rainfall ends, the analysis and control unit will control the timing feedback module to perform rain interval timing. When the dust concentration exceeds the concentration threshold, the analysis and control unit will control the timing feedback module to perform concentration excess timing. When the duration of the rain interval timing reaches the rain interval duration threshold or the duration of the concentration excess timing reaches the concentration excess duration threshold, the analysis and control unit will determine that image data collection is needed, and control the timing feedback module to clear the rain interval timing and the concentration excess timing.

[0019] As a further improvement of the present application, the concentration excess time threshold is smaller than the rain interval time threshold. When rain occurs during the rain interval timing and the concentration excess timing, the analysis and control unit will control the timing feedback module to clear the rain interval timing and the concentration excess timing. During the concentration excess timing, when the dust concentration is lower than the concentration threshold, the analysis and control unit will control the timing feedback module to suspend the concentration excess timing. When the dust concentration is higher than the concentration threshold again, the analysis and control unit will control the timing feedback module to continue the concentration excess timing.

[0020] As a further improvement of the present application, the collection setting unit is also used to set the collection cycle. In the permanent judgment and collection mode, after the image data of the crop leaves are collected, when the judgment result of the analysis and control module is no, the analysis and control unit will switch the mode from the permanent judgment and collection mode to the intensive weekly collection mode. In the intensive weekly collection mode, the analysis and control unit will regularly control the collection drone to perform image data collection according to the collection cycle. When the judgment result of the analysis and control module is yes or rainy weather occurs, the analysis and control module will switch the mode back to the permanent judgment and collection mode.

[0021] As another improvement of the present application, the mapping UAV also includes a dust-cleaning water spray assembly mounted on the UAV body. The dust-cleaning water spray assembly is used to clean dust on crop leaves by spraying water, and the dust-cleaning water spray assembly is configured with a quantitative water spraying function. The analysis and control unit is also used to control the dust-cleaning water spray assembly, and the collection setting unit is also used to set the single water spraying amount.

[0022] As another improvement and supplement to the present application, the analysis and control module is also used to determine whether the cleaning effect of the dust cleaning water spray component on crop leaves meets the standard, and to generate a dust cleaning water consumption recommendation based on the single water spray volume and the number of water sprays required to meet the cleaning effect. The intelligent alarm module is also used to send the dust cleaning water consumption recommendation to relevant personnel under the control of the analysis and control module.

[0023] In summary, the present application, through the joint setting of the data collection center and the data processing center, enables the monitoring system in the present application to determine whether the dust on the leaves of crops needs to be cleaned based on image data collection, deep learning, image processing and other technologies, and can issue alarms and reminders when the dust needs to be cleaned, prompting relevant personnel to clean the dust on the leaves of crops in time, thereby effectively avoiding the growth, development, yield and quality of crops affected by excessive dust on the leaves, thereby promoting the healthy growth of crops and the development of agriculture, and before image data collection, it will be judged according to the environmental conditions whether there is a risk of excessive dust on the leaves of crops. Only when there is a risk will image data collection be performed, which greatly improves the intelligence and reliability of the system; and through the joint setting of the dust cleaning water spray component, the analysis and control unit, the analysis and control alarm module, etc., when the judgment result is that the dust on the leaves needs to be cleaned, the monitoring system in the present application will also generate a dust cleaning water consumption recommendation based on the simulated water spray cleaning experiment, thereby providing decision support and basis for dust cleaning, thereby reducing the waste of water resources while ensuring the cleaning effect, and further improving the intelligence and reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a system structure block diagram of an agricultural monitoring system based on deep learning technology in the first embodiment of the present application;

[0025] Figure 2 This is a system structure block diagram of the data acquisition center in the first embodiment of the present application;

[0026] Figure 3 This is a workflow diagram of an agricultural monitoring system based on deep learning technology in the first embodiment of the present application;

[0027] Figure 4 This is a workflow diagram of the resident judgment mode in the first embodiment of the present application;

[0028] Figure 5 This is a system structure block diagram of the data acquisition center in the second implementation mode of this application;

[0029] Figure 6 This is a workflow diagram for simulating a water jet cleaning experiment in the second embodiment of the present application. DETAILED DESCRIPTION

[0030] Two implementation modes of the present application are described in detail below with reference to the accompanying drawings.

[0031] The first implementation method:

[0032] Figure 1-Figure 4 An agricultural monitoring system based on deep learning technology is shown, including a data acquisition center, a data transmission module, and a data processing center. The data acquisition center includes a data acquisition module, the data acquisition module includes an image acquisition unit, and the data processing center includes a data storage module, an analysis and alarm module, an intelligent alarm module, and a model training module based on deep learning technology;

[0033] The image acquisition unit is used to collect image data of crop leaves;

[0034] The data transmission module is used to transmit the image data collected by the image acquisition unit to the data processing center;

[0035] The data storage module is used to store the data collected by the data collection center;

[0036] The model training module is used to build and train deep learning models (for example, when building a deep learning model, a convolutional neural network (CNN) architecture can be used as the basis. In the training stage, leaf images containing different crops, growth stages, and dust conditions are collected. After professional annotation and proportional division of the data set, the model is trained. A suitable initialization method, cross entropy or mean square error loss function, Adam optimizer, and learning rate decay and early stopping mechanism are used. The model is trained until the accuracy requirement is met in the test set, and accurate dust judgment capability is achieved. The trained deep learning model can assist the analysis and control module in completing the work of judging whether the dust on the leaves of crops needs to be cleaned. Since the construction and training of deep learning models are mostly mature existing technologies, they are not elaborated here);

[0037] The intelligent alarm module is used for alarm;

[0038] The analysis and control module has an image processing function, which is used to call the trained deep learning model to analyze the image data to determine whether the dust on the crop leaves needs to be cleaned (the analysis and control module will first perform pre-processing such as format conversion and normalization on the image data, and then input the processed data into the trained deep learning model, and then compare the output of the model with the preset threshold to determine whether the dust needs to be cleaned). Each time the crop leaf image data is collected, the analysis and control module will make a judgment. When the judgment result is yes, the analysis and control module will control the intelligent alarm module to alarm the relevant personnel (the alarm method can be SMS, email, etc.).

[0039] The data collection center also includes a collection control module, which also includes an environmental monitoring unit. The collection control module includes a collection setting unit, an analysis and control unit, and a timing feedback module.

[0040] The data acquisition module collects image data of crop leaves through a mapping drone. The mapping drone consists of a drone body and a camera mounted on the drone body. The environmental monitoring unit is used to collect environmental data around the farmland in real time. The environmental data includes dust concentration data and meteorological data, and the data collected by the environmental monitoring unit will be transmitted to the analysis and control unit in real time. The collection setting unit is used to set the collection path and collection points of the mapping drone. The analysis and control unit is used to control the mapping drone to perform image data collection. The timing feedback module is used to perform timing operations under the control of the analysis and control unit. There are two modes in which the analysis and control unit controls the mapping drone to perform image data collection, namely, the resident judgment mode and the close injection weekly collection mode.

[0041] The acquisition setting unit is also used to set the concentration threshold and the duration threshold. The duration threshold includes the rain interval duration threshold and the concentration exceeding duration threshold. The analysis and control unit is also used to analyze the environmental data collected by the environmental monitoring unit according to the concentration threshold and the duration threshold to determine whether image data acquisition is required. When the judgment result is yes, the analysis and control unit will control the mapping drone to perform image data acquisition according to the set acquisition path and acquisition point. The acquisition point is on the acquisition path. When the analysis and control unit controls the mapping drone to perform image data acquisition, it will control the mapping drone to fly according to the set acquisition path. After flying to the acquisition point, The analysis and control unit will control the mapping UAV to hover at the collection point and control the camera to take pictures of the crop leaves, thereby collecting image data of the crop leaves. The number of collection points can be one or more, and the specific number can be set by technical personnel in this field as appropriate. When the number of collection points is multiple, after the photo operation is completed at the first collection point, the analysis and control unit will control the mapping UAV to fly to the next collection point until the photo operation is completed at the last collection point. In this way, it is considered that one crop leaf image data collection is completed. After the collection is completed, the analysis and control unit will control the mapping UAV to return.

[0042] In the resident judgment mode, after each rainfall ends (the meteorological data contains rainfall information, so the analysis and control unit can accurately understand the information about rainfall), the analysis and control unit will control the timing feedback module to perform rain interval timing. When the dust concentration exceeds the concentration threshold, the analysis and control unit will control the timing feedback module to perform concentration excess timing. When the duration of the rain interval timing reaches the rain interval time threshold or the duration of the concentration excess timing reaches the concentration excess time threshold, the analysis and control unit will determine that image data acquisition is required, and control the timing feedback module to clear the rain interval timing and the concentration excess timing. When it rains, rainwater will wash the leaves and clean up the dust on the leaves. If there is no rainfall for a long time, or the dust concentration in the environment is at a high level for a long time, the risk of excessive dust on the leaves of crops will be greatly increased. Therefore, collecting environmental data and setting concentration thresholds are to determine whether there is a risk of excessive dust on the leaves of crops. When the duration of the rain interval timing reaches the rain interval time threshold or the duration of the concentration excess timing reaches the concentration excess time threshold, it means that there is a risk, so the analysis and control unit will determine that image data acquisition is required.

[0043] The concentration excess time threshold is smaller than the rain interval time threshold. When rain occurs during the rain interval timing and the concentration excess timing, the analysis and control unit will control the timing feedback module to clear the rain interval timing and the concentration excess timing. During the concentration excess timing, when the dust concentration is lower than the concentration threshold, the analysis and control unit will control the timing feedback module to suspend the concentration excess timing. When the dust concentration is higher than the concentration threshold again, the analysis and control unit will control the timing feedback module to continue the concentration excess timing.

[0044] The acquisition setting unit is also used to set the acquisition cycle. In the permanent judgment and acquisition mode, after the image data of the crop leaves are collected, when the judgment result of the analysis and control module is no, the analysis and control unit will switch the mode from the permanent judgment and acquisition mode to the intensive weekly acquisition mode. In the intensive weekly acquisition mode, the analysis and control unit will regularly control the acquisition drone to perform image data acquisition according to the acquisition cycle. When the judgment result of the analysis and control module is yes or rainy weather occurs, the analysis and control module will switch the mode back to the permanent judgment and acquisition mode.

[0045] The monitoring system in the present application can determine whether the dust on the leaves of crops needs to be cleaned based on image data acquisition, deep learning, image processing and other technologies, and can issue alarms and reminders when the dust needs to be cleaned, prompting relevant personnel to clean the dust on the leaves of crops in time, thereby effectively avoiding the growth, development, yield and quality of crops affected by excessive dust on the leaves, thereby promoting the healthy growth of crops and the development of agriculture. Before collecting image data, it will determine whether there is a risk of excessive dust on the leaves of crops based on environmental conditions. Image data will only be collected when there is a risk, which greatly improves the intelligence and reliability of the system.

[0046] The second implementation method:

[0047] Figure 5-Figure 6 An agricultural monitoring system based on deep learning technology is shown. Different from the first embodiment, the mapping drone also includes a dust-cleaning water spray component mounted on the drone body. The dust-cleaning water spray component is used to clean dust on crop leaves by spraying water, and the dust-cleaning water spray component is configured with a quantitative water spray function. The analysis and control unit is also used to control the dust-cleaning water spray component. The collection setting unit is also used to set a single water spray volume. The analysis and control module is also used to determine whether the cleaning effect of the dust-cleaning water spray component on crop leaves is up to standard, and to generate a dust cleaning water consumption recommendation based on the single water spray volume and the number of water sprays required to achieve the cleaning effect. The intelligent alarm module is also used to send the dust cleaning water consumption recommendation to relevant personnel under the control of the analysis and control module (the sending method can be SMS, email, etc.).

[0048] Normally, water spray cleaning is one of the common methods for cleaning dust on crop leaves, but controlling the amount of water used is a difficult point. Too much water can easily cause a waste of water resources, and too little water can easily lead to unsatisfactory cleaning effects. In order to solve this problem, in this embodiment, when the judgment result of the analysis and control alarm module is yes, the system will conduct a simulated water spray cleaning experiment. The specific steps of the experiment are as follows: the analysis and control unit controls the mapping drone to go to one of the collection points. After arriving at the collection point, the analysis and control unit will control the mapping drone to hover at the collection point, and control the dust cleaning water spray component to spray water on the crop leaves for the first time according to the single water spraying amount, so as to clean the dust on the crop leaves for the first time. After the water spraying is completed, the analysis and control unit will control the camera to take pictures of the crop leaves. Thus, the image data after the first cleaning is collected, and then the analysis and control module will call the trained deep learning model to analyze the image data after the first cleaning to determine whether the cleaning effect meets the standard. When the judgment result is no (that is, when the cleaning effect does not meet the standard), the analysis and control unit will control the dust cleaning water spray component to spray water on the crop leaves for the second time, and so on, until the judgment result of the analysis and control module is yes (that is, until the cleaning effect meets the standard). When the judgment result of the analysis and control module is yes (that is, when the cleaning effect meets the standard), the analysis and control unit will control the image collection drone to return, and the analysis and control module will generate dust cleaning water consumption recommendations based on the number of water sprays and the single water spray volume, and control the intelligent alarm module to send the dust cleaning water consumption recommendations to relevant personnel;

[0049] Therefore, through the joint setting of the dust cleaning water spray component, the analysis and control unit, the analysis and control alarm module, etc., when the judgment result is that the dust on the blades needs to be cleaned, the monitoring system in this application will also generate dust cleaning water consumption recommendations based on the simulated water spray cleaning experiment, thereby providing decision-making support and basis for dust cleaning, and thus reducing the waste of water resources while ensuring the cleaning effect, further improving the intelligence and reliability of the system.

[0050] The dust-cleaning water spray component adopts the existing technology. Those skilled in the art can select a suitable device with a quantitative water spraying function from the existing technology as the dust-cleaning water spray component in this application. For example, the dust-cleaning water spray component can be a device for spraying pesticides on a plant protection drone. This is a well-known technology for those skilled in the art and will not be elaborated here.

[0051] In view of current practical needs, the above-mentioned implementation mode adopted in this application is not limited to the scope of protection. Various changes made within the knowledge scope of technical personnel in this field without departing from the concept of this application still fall within the scope of protection of the present invention.

Claims

1. An agricultural monitoring system based on deep learning technology, characterized in that: It includes a data acquisition center, a data transmission module, and a data processing center. The data acquisition center includes a data acquisition module, the data acquisition module includes an image acquisition unit, and the data processing center includes a data storage module, an analysis and alarm control module, an intelligent alarm module, and a model training module based on deep learning technology; The image acquisition unit is used to collect image data of crop leaves; The data transmission module is used to transmit the image data collected by the image acquisition unit to the data processing center; The data storage module is used to store the data collected by the data collection center; The model training module is used to build and train a deep learning model; The intelligent alarm module is used for alarm; The analysis and control alarm module has an image processing function, which is used to call the trained deep learning model to analyze the image data to determine whether the dust on the crop leaves needs to be cleaned. Each time the crop leaf image data is collected, the analysis and control alarm module will make a judgment. When the judgment result is yes, the analysis and control alarm module will control the intelligent alarm module to alarm the relevant personnel.

2. The agricultural monitoring system based on deep learning technology according to claim 1, characterized in that: The data acquisition center also includes an acquisition control module, the data acquisition module also includes an environmental monitoring unit, and the acquisition control module includes an acquisition setting unit, an analysis and control unit, and a timing feedback module; The data acquisition module collects image data of crop leaves through a mapping drone. The mapping drone consists of a drone body and a camera mounted on the drone body. The environmental monitoring unit is used to collect environmental data around the farmland in real time. The environmental data includes dust concentration data and meteorological data, and the data collected by the environmental monitoring unit will be transmitted to the analysis and control unit in real time. The acquisition setting unit is used to set the acquisition path and acquisition points of the mapping drone. The analysis and control unit is used to control the mapping drone to perform image data acquisition. The timing feedback module is used to perform timing operations under the control of the analysis and control unit. There are two modes in which the analysis and control unit controls the mapping drone to perform image data acquisition, namely, a permanent judgment mode and a close injection weekly acquisition mode.

3. The agricultural monitoring system based on deep learning technology according to claim 2 is characterized in that: The acquisition setting unit is also used to set a concentration threshold and a duration threshold, and the duration threshold includes a rain interval duration threshold and a concentration exceeding duration threshold. The analysis and control unit is also used to analyze the environmental data collected by the environmental monitoring unit according to the concentration threshold and the duration threshold to determine whether image data acquisition is needed. When the judgment result is yes, the analysis and control unit will control the mapping drone to perform image data acquisition according to the set acquisition path and acquisition points.

4. The agricultural monitoring system based on deep learning technology according to claim 3 is characterized in that: In the resident judgment and sampling mode, after each rainfall ends, the analysis and control unit will control the timing feedback module to perform rain interval timing. When the dust concentration exceeds the concentration threshold, the analysis and control unit will control the timing feedback module to perform concentration excess timing. When the duration of the rain interval timing reaches the rain interval duration threshold or the duration of the concentration excess timing reaches the concentration excess duration threshold, the analysis and control unit will determine that image data collection is needed and control the timing feedback module to clear the rain interval timing and concentration excess timing.

5. The agricultural monitoring system based on deep learning technology according to claim 4 is characterized in that: The concentration excess time threshold is smaller than the rain interval time threshold. When rain occurs during the rain interval timing and the concentration excess timing, the analysis and control unit will control the timing feedback module to clear the rain interval timing and the concentration excess timing. During the concentration excess timing, when the dust concentration is lower than the concentration threshold, the analysis and control unit will control the timing feedback module to suspend the concentration excess timing. When the dust concentration is higher than the concentration threshold again, the analysis and control unit will control the timing feedback module to continue the concentration excess timing.

6. The agricultural monitoring system based on deep learning technology according to claim 5, characterized in that: The acquisition setting unit is also used to set the acquisition cycle. In the permanent judgment and acquisition mode, after the image data of the crop leaves are collected, when the judgment result of the analysis and control module is no, the analysis and control unit will switch the mode from the permanent judgment and acquisition mode to the intensive weekly acquisition mode. In the intensive weekly acquisition mode, the analysis and control unit will regularly control the acquisition drone to perform image data acquisition according to the acquisition cycle. When the judgment result of the analysis and control module is yes or rainy weather occurs, the analysis and control module will switch the mode back to the permanent judgment and acquisition mode.

7. The agricultural monitoring system based on deep learning technology according to claim 6, characterized in that: The mapping drone also includes a dust-cleaning water spray component mounted on the drone body, and the dust-cleaning water spray component is used to clean dust on crop leaves by spraying water, and the dust-cleaning water spray component is configured with a quantitative water spray function. The analysis and control unit is also used to control the dust-cleaning water spray component, and the collection setting unit is also used to set a single water spray volume.

8. The agricultural monitoring system based on deep learning technology according to claim 7, characterized in that: The analysis and control module is also used to determine whether the cleaning effect of the dust cleaning water spray component on crop leaves meets the standard, and to generate a dust cleaning water consumption recommendation based on the single water spray volume and the number of water sprays required to meet the cleaning effect. The intelligent alarm module is also used to send the dust cleaning water consumption recommendation to relevant personnel under the control of the analysis and control module.

Images