A method and system for intelligent control and management of water and air flushing pits suitable for tree planting equipment

Abstract

The present invention relates to the field of smart agriculture and forestry technology, and discloses a method and system for intelligently controlling and managing water-gas flushing for tree planting equipment. The method comprises the following steps: initialization and configuration; preliminary execution and information transmission by a lower computer execution terminal; data preprocessing; neural network decision-making on water-gas output pressure and combination ratio; receiving and executing the decision result; and uploading operation location information. The method and system for intelligently controlling and managing water-gas flushing for tree planting equipment provided by the present invention utilizes a combination of water-gas dual-fluid power to improve a pure water flushing solution, utilizes pneumatic power instead of water flow power, effectively reduces flushing water consumption, makes reasonable decisions on flushing water and gas usage, provides an intelligent solution for flushing with a combination of water and gas, utilizes high-performance computing equipment to replace manual decision-making, and improves the intelligence level of the tree planting equipment.

Description

Technical Field

[0001] The present invention relates to the field of smart agriculture and forestry technology, and specifically to an intelligent control and management method and system for water and gas flushing pits suitable for tree planting equipment. Background Art

[0002] With the prevention and control of sand and desertification, strengthening desert ecological protection, and promoting desertification control, it is necessary to promote the protection and restoration of major ecosystems, accelerate cross-regional ecological protection, and promote ecological environment co-governance and ecological restoration projects. However, in harsh desert environments, the lack of water resources has always been one of the major difficulties in desert control. In order to achieve the goal of sand control and desertification control, realize intelligent and rational sand control, and solve the problem of large water resource consumption in the construction of pits by flushing, there is an application bottleneck in water-scarce areas.

[0003] Among existing technologies, the technical solution described in the Chinese invention patent application document "An Air-Impacted Salix Psammophila Planting Device for Planting in Sandy Land" (Grant No. CN 209710825U) primarily addresses the dust problem of existing air-impacted tree planting devices. This problem is addressed by utilizing a dripping water device, improving the dust generation defect and increasing air-impact efficiency. Furthermore, the technical solution described in the Chinese invention patent application document "A Water-Impacted Salix Psammophila Automatic Planting Vehicle and Planting Method Thereof" (Publication No. CN 112005661A) combines water-impacted pit-making with tree-planting machinery, proposing a design solution with related structural innovations.

[0004] However, the above invention technical solutions related to the flushing technology basically remain at the institutional innovation level for the application and deployment of the flushing technology. They do not discuss how to apply the appropriate amount of water in specific situations during the flushing process, and do not propose more advanced and water-saving solutions for flow control. For this reason, the present invention proposes an intelligent control and management method for water and gas flushing suitable for tree planting equipment. Summary of the Invention

[0005] (1) Technical problems solved

[0006] In response to the shortcomings of the existing technology, the present invention provides a water-gas flushing intelligent control and management method and system suitable for tree planting equipment, applies water-gas combination technology to tree planting equipment to improve pure water flushing, provides a water-gas control deployment plan, a flushing control algorithm and its implementation steps, and uses edge computing equipment to realize operation decision-making and operation information uploading, so as to achieve the invention purpose of using the Internet of Things and artificial intelligence technologies to comprehensively broaden the functions of tree planting equipment, realize the intelligence of tree planting equipment and promote the development of smart agriculture and forestry.

[0007] (2) Technical solution

[0008] To achieve the above-mentioned purpose, the present invention provides the following technical solutions:

[0009] A method for intelligently controlling and managing water-gas flushing pits for tree planting equipment includes the following steps:

[0010] Step 1: Initialize the settings through the water and air flushing intelligent control and management system suitable for tree planting equipment;

[0011] Step 2: Initial execution and information transmission on the lower computer execution end;

[0012] Step 3: Use the alternating least squares method to handle missing values in the input samples. Use the OpenCV computer vision library to perform image processing on the input labels. Use the parsed pixel range values as the pit size data label to obtain a pit impact image. Finally, normalize the data preprocessed in the first two steps and package it into the network to prepare for the current round of decision making.

[0013] Step 4: Input the processed data into the neural network and load the weight file. The system can then determine the output water vapor combined pressure output value and the combined ratio. The output result will be transmitted back to the lower computer processor via the system communication link for execution;

[0014] Step 5: Receive and execute decision results;

[0015] Step 6: After a single work process is completed, the current work location data obtained by GPS positioning will be uploaded to the forestry information department's database or information platform via communication equipment. The completed pits will be numbered and used for information management of subsequent tree planting.

[0016] Preferably, the initialization setting in the first step is: the user first selects the manual working mode or the automatic working mode. After powering on, the sensor is always in the working and data transmission state, and the transmitted data is stored in the corresponding channel of the lower computer main processor via the data line and waits to be read.

[0017] Preferably, the water-gas flushing pit intelligent control and management system for tree planting equipment in the first step includes a sensor and user input end, a lower computer and execution end, and an intelligent computing end, wherein the sensor and user input end includes a soil moisture sensor, a pressure sensor, an optical camera, and a GPS positioning module;

[0018] The lower computer execution end consists of a processor, a water pump with electronically controlled flow, an air pump, and a pressure control valve;

[0019] The intelligent computing end consists of a main processor, high-performance edge computing and its acceleration devices, as well as storage devices.

[0020] Preferably, the specific steps in the second step are: first determine the user selection mode. If the user selects the manual working mode, the processor directly controls the water pump and the air pump to work according to the water and gas output parameters set by the user. Regardless of the manual working mode or the automatic working mode, the lower computer processor will read the data of the sensor signal line or the data buffer. After parsing, the data can be distinguished by writing the necessary prefix and end data frames. The intelligent computing end determines the type of data by identifying the communication frame codes before and after a single character string. Which frame code to use as a distinguishing mark can be decided by the specific technical implementer according to needs.

[0021] Preferably, the input samples in the third step are soil moisture and soil hardness;

[0022] The input label is the crater result image;

[0023] The image processing process includes image denoising, grayscale conversion, Canny edge detection and obtaining pixel range.

[0024] Preferably, the weight file of the neural network in the fourth step should be kept updated in real time, that is, the neural network should maintain training activity, and adjust the weights by continuously absorbing the pit impact effect pictures to form a benign closed-loop model.

[0025] Preferably, the specific content of the fifth step is: after receiving the decision result returned by the intelligent computing end, the lower computer execution end converts the relevant percentage data into PWM corresponding output, thereby controlling the water pump, air pump and pressure control valve to operate at appropriate power, thereby outputting appropriate water and air output and water-gas combination ratio.

[0026] (3) Beneficial effects

[0027] Compared with the prior art, the intelligent control and management method and system for water-gas flushing pits for tree planting equipment provided by the present invention has the following beneficial effects:

[0028] 1. This intelligent control and management method and system for water-gas flushing pits suitable for tree planting equipment adopts a combination of water and gas dual fluid power to improve the pure water flushing solution, uses pneumatic power instead of water flow power, effectively reduces the water used for flushing pits, effectively reduces the water consumption for flushing pits, makes reasonable decisions on the water and gas usage for flushing pits, provides an intelligent solution for flushing pits with a combination of water and gas, uses high-performance computing equipment to replace manual decision-making, and improves the intelligence level of tree planting equipment.

[0029] 2. This intelligent water-gas flushing control and management method and system for tree planting equipment allows the flushing decision results to be fed back into the neural network for reinforcement training based on actual conditions, continuously iterating and updating the weight file. This makes the algorithm's decisions more reliable and the decision model possesses stronger learning and updating capabilities and robustness.

[0030] 3. This intelligent water-air flushing control and management method and system is applicable to tree planting equipment, such as embedded intelligent water-air spray gun flushing devices, intelligent tree planting vehicles, and intelligent tree planting machines. This is because this technical solution fully considers the space for application expansion, designs a control scheme that supports user needs, fully taps the functional potential of the equipment, has multiple application scenarios, is highly scalable, and is applicable to a variety of work equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic diagram of the deployment and functions of the sensing and user input terminal system according to an embodiment of the present invention;

[0032] Figure 2 This is a diagram of the system deployment and function of the lower-level execution end of an embodiment of the present invention;

[0033] Figure 3 This is a diagram of the deployment and functionality of the intelligent computing terminal system according to an embodiment of the present invention;

[0034] Figure 4 This is a schematic diagram of the lower computer driving workflow of an embodiment of the present invention;

[0035] Figure 5 Schematic diagram of the workflow of the intelligent computing unit according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0037] Example

[0038] See also Figure 1-5 The embodiment of the present invention provides an intelligent control and management method for water-gas flushing pits for tree planting equipment, comprising the following steps:

[0039] Initialization: After selecting the operating mode, the sensor powers on and begins testing. Users can choose between manual and automatic operating modes. After powering on, the sensor is always in operation and transmitting data. Transmitted data is stored via the data line in the corresponding channel of the host processor, awaiting readout.

[0040] Step 1: The lower-computer execution end performs preliminary execution and information transmission, first determining the user-selected mode. If the user selects manual operation mode, the processor directly controls the water and air pumps according to the user-set water and gas output parameters. Regardless of whether the operation mode is manual or automatic, the lower-computer processor reads the data from the sensor signal line or data buffer. After parsing, it distinguishes the data by writing the necessary preamble and end data frames. The upper computer (i.e., the intelligent computing end) determines the data type by identifying the communication frame codes before and after a single string. The specific frame code used as a distinguishing mark can be determined by the specific technical implementer based on their needs.

[0041] Step 2: Data preprocessing. Environmental data is divided into three categories: soil moisture, soil hardness, and crater result images. Soil moisture and soil hardness are usually processed as input samples. For input samples, the Alternating Least Squares method can be used to process missing values in the collected data to obtain complete data. The crater result image is usually used as the input label. For the input label, the Opencv computer vision library is used for image processing: image denoising->grayscale->Canny edge detection->obtain pixel range, and the resolved pixel range value is used as the crater size data label. Finally, the data preprocessed in the first two steps is normalized and packaged and loaded into the network to prepare for this round of decision-making.

[0042] Step 3: The neural network determines the water vapor output pressure and combination ratio. The processed data is fed into the neural network and loaded into the weight file. The system then determines the output water vapor combined pressure and combination ratio. The output is transmitted back to the lower-level processor via the system communication link for execution.

[0043] It is worth noting that the weight file of the neural network should be kept updated in real time, that is, the neural network should maintain training activity, and adjust the weights based on the continuously absorbed results of the previous round of execution of the system (pictures of the flushing effect), forming a benign closed-loop model to continuously improve the accuracy of decision-making and make water and vapor control more reasonable.

[0044] Step 4: Receive and execute the decision results. After receiving the decision results (output value and combination ratio) returned by the intelligent computing end, the lower computer execution end converts the relevant percentage data into PWM corresponding output, thereby controlling the water pump, air pump and pressure control valve to operate at appropriate power, thereby outputting the appropriate water and air output and water-air combination ratio.

[0045] Step 5: Upload the work location information. After a single work process is completed, the current work location data obtained by GPS positioning will be uploaded to the forestry information department's database or information platform via communication equipment. The completed pits will be numbered and used for subsequent tree planting information management.

[0046] The embodiment of the present invention provides an intelligent water and gas flushing control and management system suitable for tree planting equipment, including a sensor and user input end, a lower computer and execution end, and an intelligent computing end. The sensor and user input end includes a soil moisture sensor, a pressure sensor, an optical camera, and a GPS positioning module.

[0047] The execution end of the lower computer is mainly composed of a processor, a water pump with electronically controllable flow, an air pump and a pressure control valve.

[0048] The intelligent computing end consists of a main processor, high-performance edge computing and its acceleration devices, as well as storage devices.

[0049] The sensing and user input terminal primarily incorporates a variety of sensors for multi-dimensional information collection and real-time feedback, including soil moisture sensors, pressure sensors, optical cameras, and GPS positioning modules. The corresponding return signals are: soil moisture, soil hardness (measured by the magnitude of the electrical signal), the operation image (a photo of the flushed pit), and the coordinates of the current operation position. A manual operation setting module should also be provided for the user. This module should include the necessary buttons and display screens to ensure that users can customize the water and gas output parameters according to their actual work needs, or preferably directly perform water and gas flushing operations.

[0050] It should be noted that for manual operation mode, the solution of the present invention is generally applicable to integration into handheld operation equipment, such as an embedded intelligent water-air spray gun pit flushing device. For automatic operation mode, the solution of the present invention is generally applicable to integration into a tree planting vehicle or tree planting machine. The subsequent description of the mode switching function is based on the above assumptions.

[0051] The lower computer execution end is mainly responsible for module control to achieve water flow and air flow output, and uploads the information sent back by the front-stage sensor to the intelligent computing end. The lower computer execution end is mainly composed of a processor, a water pump with electrically controlled flow, an air pump and a pressure control valve. The processor can be a processor with an ARM-Cortex4 series core, such as STM32H7, etc., and it is particularly important to note that it should have PWM output capabilities. The water pump, air pump and pressure control valve should be able to receive electrical signal analog or PWM signal to control the operating power. The water outlet pipe and the air outlet pipe are selected by the technical implementer according to the actual project requirements. Since the present invention is a method invention, the specific mechanism details will not be elaborated too much. However, it should be noted that the sensor equipment should be placed in a position where it can reach the sand and has relatively good water insulation protection.

[0052] According to the actual engineering environment and communication speed requirements, the lower computer execution end and the intelligent computing end should establish reasonable communication solutions and communication links, such as CAN bus communication, 2.4G WIFI transparent transmission, etc.

[0053] The intelligent computing end primarily performs large-scale computing tasks. It consists of a main processor, high-performance edge computing and its acceleration devices, and storage devices. For hardware deployment, the high-performance edge computing device can use the NVIDIA JETSON series parallel computing unit, and the acceleration device can use FPGA. Software deployment uses an algorithm framework primarily based on the BP neural network. The BP neural network is primarily responsible for water-vapor flushing decisions. The input parameters are soil moisture and soil hardness (represented by the size of the electrical signal), and the output parameters are water flow pressure, air flow pressure, and the water-vapor combination ratio. The storage device stores the following information: the neural network training set, the weight file generated by the training, and the real-time prediction data sample set. The intelligent computing end should be deployed with Ethernet and connected to the forestry department's operation information system to ensure that operation data can be uploaded in real time.

[0054] The above-mentioned embodiments of the present invention provide an intelligent control and management method and system for water-gas flushing suitable for tree planting equipment. The method adopts a combination of water and gas dual fluid power to improve the pure water flushing solution, uses pneumatic power instead of water flow power, effectively reduces the water used for flushing, effectively reduces the water consumption for flushing, makes reasonable decisions on the water and gas usage for flushing, provides an intelligent solution for flushing with a combination of water and gas, uses high-performance computing equipment to replace manual decision-making, and improves the intelligence level of tree planting equipment.

[0055] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent control and management of water and air flushing pits for tree planting equipment, characterized in that: The following steps are involved: Step 1: Initialize the settings through the water and air flushing intelligent control and management system suitable for tree planting equipment; Step 2: Initial execution and information transmission on the lower computer execution end; Step 3: Use the alternating least squares method to handle missing values in the input samples. Use the OpenCV computer vision library to perform image processing on the input labels. Use the parsed pixel range values as the pit size data label to obtain a pit impact image. Finally, normalize the data preprocessed in the first two steps and package it into the network to prepare for the current round of decision making. Step 4: Input the processed data into the neural network and load the weight file. The system will then determine the output value of the combined water-vapor pressure and the combined ratio. The output result will be transmitted back to the lower-level processor via the system communication link for execution. Step 5: Receive and execute decision results; Step 6: After a single workflow is completed, the current operation location data obtained by GPS positioning will be uploaded to the database or information platform of the forestry information department via communication equipment; the completed pits will be numbered for subsequent information management of tree planting.

2. The intelligent control and management method for water-gas flushing pits for tree planting equipment according to claim 1 is characterized in that: The initialization settings in the first step are: the user first selects manual working mode or automatic working mode. After power-on, the sensor is always in working and data transmission state. The transmitted data will be stored in the corresponding channel of the lower computer main processor via the data line and wait for reading.

3. The intelligent control and management method for water-gas flushing pits for tree planting equipment according to claim 1 is characterized by: The first step is the intelligent water and air flushing control and management system for tree planting equipment, which includes a sensor and user input terminal, a lower computer and execution terminal, and an intelligent computing terminal. The sensor and user input terminal includes a soil moisture sensor, a pressure sensor, an optical camera, and a GPS positioning module. The lower computer execution end consists of a processor, a water pump with electronically controlled flow, an air pump, and a pressure control valve; The intelligent computing end consists of a main processor, high-performance edge computing and its acceleration devices, as well as storage devices.

4. The intelligent control and management method for water-gas flushing pits for tree planting equipment according to claim 1 is characterized in that: The specific steps in the second step are: first determine the user selection mode. If the user selects the manual working mode, the processor directly controls the water pump and air pump to work according to the water and gas output parameters set by the user. Regardless of the manual working mode or the automatic working mode, the lower computer processor will read the data of the sensor signal line or the data buffer. After parsing, the data can be distinguished by the pre- and end data frames written. The intelligent computing end determines the data type by identifying the communication frame codes before and after a single character string. Which frame code to use as a distinguishing mark can be decided by the specific technical implementer according to needs.

5. The intelligent control and management method for water-gas flushing pits for tree planting equipment according to claim 1 is characterized in that: The input samples in the third step are soil moisture and soil hardness; The input label is the crater result image; The image processing process includes image denoising, grayscale conversion, Canny edge detection and obtaining pixel range.

6. The intelligent control and management method for water-gas flushing pits for tree planting equipment according to claim 1 is characterized in that: In the fourth step, the weight file of the neural network should be kept updated in real time, that is, the neural network should maintain training activity, and adjust the weights by continuously absorbing the pit impact effect pictures to form a benign closed-loop model.

7. The intelligent control and management method for water-gas flushing pits for tree planting equipment according to claim 1 is characterized in that: The specific content of the fifth step is: after receiving the decision result returned by the intelligent computing end, the lower computer execution end converts the relevant percentage data into PWM corresponding output, thereby controlling the water pump, air pump and pressure control valve to operate at appropriate power, thereby outputting the appropriate water and air output and the water-air combination ratio.

8. An intelligent water-gas flushing control and management system for tree planting equipment that implements the method according to any one of claims 1 to 7, characterized in that: It includes sensing and user input end, lower computer and execution end, and intelligent computing end. The sensing and user input end includes soil moisture sensor, pressure sensor, optical camera and GPS positioning module.

9. The intelligent water-gas flushing control and management system for tree planting equipment according to claim 8 is characterized in that: The lower computer execution end is composed of a processor, a water pump with electrically controllable flow, an air pump and a pressure control valve.

10. The intelligent water-gas flushing control and management system for tree planting equipment according to claim 8, characterized in that: The intelligent computing end is composed of a main processor, high-performance edge computing and its acceleration equipment, and storage equipment.

Images