Variable irrigation control method for large sprinkler

Abstract

The application discloses a variable irrigation control method for a large-scale sprinkling irrigation machine, which comprises the following steps: adjusting the water irrigation amount of each sprinkler head by installing a variable irrigation controller on each sprinkler head of the sprinkling irrigation machine, forming a network by using wireless communication between the variable irrigation controllers, and realizing real-time adjustment of the valve opening degree by using a valve opening degree adjustment algorithm built in the variable irrigation controller, so that the water discharge amount of the sprinkler head is quickly and stably adjusted to the target water discharge amount required by an irrigation prescription map. The variable irrigation controller is used for controlling the valve to adjust the flow of each sprinkler head of the sprinkling irrigation machine, and a cooperative control network is formed between each valve, an artificial intelligence algorithm is used for calculating the opening degree of each valve, so that the variable irrigation of the sprinkling irrigation machine is realized by independently adjusting the valve.

Description

Technical Field

[0001] This invention relates to the field of agricultural irrigation technology, and in particular to a variable irrigation control method for a large sprinkler irrigation machine. Background Technology

[0002] Irrigation is the most important part of agricultural production. Large sprinkler irrigation machines, as a modern irrigation technology carrier with a high degree of integration of water-saving and efficient irrigation technology equipment in modern agriculture, will play a significant role in the development of intensive and large-scale agricultural production and the construction of ecological farmland in my country due to their advantages of good irrigation quality, wide applicability and high degree of automation.

[0003] There are two common types of large sprinkler systems: pointer-type and translational-type. The pointer-type sprinkler, also known as a center-pivot sprinkler, has sprinklers mounted on a self-propelled support frame. During operation, the machine rotates around a central point equipped with a water supply system while spraying water. Because its movement resembles the hands of a clock, it is called a pointer-type sprinkler, also known as a center-pivot sprinkler or circular sprinkler. The truss height from the ground is typically 2-3 meters, and the truss span is typically 30-65 meters. A single sprinkler can consist of more than 10 spans. The translational-type sprinkler consists of branch pipes, sprinklers, wheels, and a truss, and moves along a direction perpendicular to the branch pipes. Like the pointer-type sprinkler system, its branch pipes are also supported on a self-propelled truss. However, the truss moves in a straight line, spraying water as it moves. This type of sprinkler system is supplied with water from an open channel perpendicular to the branch pipes, or from a water supply valve on the main pipe via a flexible hose. The irrigation area is rectangular. The structure of the trellis and tower vehicle is similar to that of the pointer-type sprinkler, and the longest translational sprinkler can reach 1000 meters.

[0004] However, due to the long span of the sprinkler irrigation machine and the undulating terrain, uneven irrigation occurs, leading to problems such as decreased crop quality and water waste, thus hindering the sprinkler irrigation machine from fully realizing its role in improving agricultural production efficiency. Researchers have attempted to use automatic control technology to control the spray volume of the sprinkler irrigation machine in different areas based on crop growth and terrain undulations, thereby improving irrigation uniformity. Variable Refrigerant Irrigation (VRI) technology, which originated in the early 20th century, has become an important research direction for large-scale sprinkler irrigation machines. To date, foreign research on the application of VRI technology in precision irrigation has made significant progress. Based on the control strategies for achieving VRI, there are three types: SPEED-VRI, ZONE-VRI, and IS-VRI. Early VRI technology mainly achieved variable control of different areas by controlling the travel speed of the sprinkler irrigation machine. Although this method achieved adjustment of irrigation volume in the direction of travel, the large span of the sprinkler irrigation machine resulted in the same irrigation volume in the direction of the sprinkler irrigation machine, limiting further adjustment of irrigation uniformity. The subsequent ZONE-VRI technology divides the sprinkler heads on the sprinkler into several groups, with the flow rate of each group uniformly adjusted, successfully solving the problem of uneven irrigation along the sprinkler direction. However, the precision of this method depends on the grouping; variable control cannot be achieved within a single group. Currently, the IS-VRI (Individual Sprinkler VRI) method, which offers more flexible control and higher irrigation uniformity, is gradually becoming the main direction for variable control of sprinkler machines. This method treats individual sprinklers as variable control objects, allowing for adjustable flow rates for each sprinkler. It can divide the plot into smaller management units, maximizing irrigation uniformity. Chinese patent CN116384195A (publication number CN116384195A) provides a method for calculating the water demand prescription for each area within a sprinkler irrigation area. It can calculate the water volume required at different locations within the sprinkler irrigation area based on factors such as terrain. The calculation results are as follows... Figure 1 Chinese patent CN116649189A (Publication No. CN116649189A) discloses a method for adjusting sprinkler head flow based on PWM control. Figure 2 This enables the flow regulation of the sprinkler head. Meanwhile, the Chinese patent for a single-valve pulse width modulation controller for large variable sprinkler irrigation machines (publication number CN110161912A) provides a design method for such a controller.

[0005] Existing methods do not consider the impact on the actual flow rate between sprinklers when they are opened simultaneously on the sprinkler pipeline due to the limitation of water pump pressure. The existing patents use the PWM method to control the flow rate of each sprinkler head by controlling the opening time of each sprinkler head. It is difficult to predict and control the situation when sprinklers are opened simultaneously. When considering the pressure changes caused by the change in pipeline flow after the sprinkler heads are opened, the control accuracy of such methods will decrease significantly, and it is difficult to achieve the expected control results in practical applications.

[0006] Therefore, how to provide a variable irrigation method suitable for large-scale translational sprinkler irrigation machines has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0007] In view of this, the purpose of this invention is to provide a variable irrigation control method for a large sprinkler irrigation machine, which uses a variable irrigation controller to control valves to adjust the flow rate of each sprinkler head of the sprinkler irrigation machine. At the same time, each valve forms a collaborative control network, and artificial intelligence algorithms are used to calculate the opening degree of each valve, thereby realizing variable irrigation of the sprinkler irrigation machine with independent valve adjustment.

[0008] The present invention solves the technical problem by adopting the following technical solution:

[0009] A variable irrigation control method for a large sprinkler irrigation machine includes: adjusting the water output of each sprinkler head by installing a variable irrigation controller; forming a network of wireless communication between the variable irrigation controllers; and adjusting the valve opening in real time by using a built-in valve opening adjustment algorithm in the variable irrigation controller, so that the water output of the sprinkler head can be quickly stabilized at the target water output required by the irrigation prescription map.

[0010] Furthermore, the variable irrigation controller (1) is installed on each of the sprinkler (3) branches on the truss (2); the variable irrigation controller (1) achieves Internet access by connecting to the irrigation gateway (4) installed on the main tower vehicle (5).

[0011] Furthermore, the variable irrigation controller mainly consists of a ball valve for controlling the switching of the sprinkler branch pipe, an adjustable ball valve control assembly, a pressure sensor, and a control circuit. The ball valve, which controls the branch pipe switching, connects the truss and the sprinkler branch pipe; opening the ball valve allows water to be delivered to the sprinkler head. The ball valve's switching is controlled by the adjustable ball valve control assembly. The pressure sensor is located inside the sprinkler branch pipe and is used to measure the water pressure in the sprinkler branch pipe in real time. The control circuit is connected to the ball valve control assembly and the pressure sensor for communication.

[0012] Furthermore, the ball valve control assembly mainly consists of a motor, a reduction gear, and a position sensor; wherein, the output end of the motor is connected to the ball valve via the reduction gear, and the opening degree of the ball valve is adjusted by the rotation of the motor; the position sensor is used to provide feedback on the opening degree of the ball valve.

[0013] Furthermore, the control circuit uses an ARM M4 series chip as the core MCU, embedding a real-time operating system and lightweight AI components.

[0014] Furthermore, each core MCU has a valve opening calculation model. The input to the valve opening calculation model is the target flow rate of each other nozzle and the target flow rate of this valve. The model calculates and outputs the opening of this valve.

[0015] Furthermore, in the valve opening calculation model, the forgotten gate control information is removed, the input gate control new information is added, and the output gate determines the next state output.

[0016] Furthermore, the irrigation gateway is a gateway device with two communication functions, supporting Internet access and local area network wireless communication; it connects to various variable irrigation controllers through local area network wireless communication, distributes irrigation volume targets for each plot, and simultaneously acquires the current pressure data of each sprinkler head; it receives irrigation prescription maps from the network platform through Internet access, and the information in the irrigation prescription map is the irrigation volume for each plot to be irrigated.

[0017] The variable irrigation control method for a large sprinkler irrigation machine disclosed in this invention has the following beneficial effects:

[0018] (1) The variable irrigation control method for large sprinkler irrigation machines of the present invention takes into account the mutual influence between the flow rates of all sprinkler heads on the sprinkler irrigation machine, and constructs a decision network using wireless communication and edge computing methods, which can quickly stabilize the flow rates of each sprinkler head on the entire sprinkler irrigation machine at the target value.

[0019] (2) The present invention adopts an independent artificial intelligence control scheme based on edge computing, which eliminates the limitation on the number of devices. That is, when the number of sprinklers increases or decreases, only the corresponding independent variable irrigation controller needs to be added, without the need to rebuild the overall control method and control model. Attached Figure Description

[0020] Figure 1 A schematic diagram showing the calculation results of irrigation volume in a traditional sprinkler irrigation area;

[0021] Figure 2 This is a flowchart of a traditional nozzle flow regulation method based on PWM control.

[0022] Figure 3 This is a schematic diagram of the device installation structure of the present invention;

[0023] Figure 4 This is a schematic diagram of the variable irrigation controller of the present invention;

[0024] Figure 5 This is a schematic diagram of the intelligent calculation model for valve opening degree of the present invention.

[0025] In the figure:

[0026] 1-Variable irrigation controller; 2- Truss; 3- Sprinkler head; 4- Irrigation gateway; 5- Main tower vehicle. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] refer to Figure 3 The present invention discloses a variable irrigation control method for a large sprinkler irrigation machine, comprising: adjusting the irrigation volume of each sprinkler head by installing a variable irrigation controller; forming a network of wireless communication between the variable irrigation controllers; and adjusting the valve opening in real time by using a built-in valve opening adjustment algorithm in the variable irrigation controller, so that the water output of the sprinkler head can be quickly stabilized at the target water output required by the irrigation prescription map.

[0029] The technical solution is further optimized by installing the variable irrigation controller 1 on each of the sprinkler head 3 branch pipes on the truss 2; the variable irrigation controller 1 achieves Internet access by connecting to the irrigation gateway 4 installed on the main tower trolley 5.

[0030] Further optimize the technical solution, refer to Figure 4 The variable irrigation controller mainly consists of a ball valve for controlling the opening and closing of the sprinkler branch pipe, an adjustable ball valve control assembly, a pressure sensor, and a control circuit. The ball valve, which controls the branch pipe opening and closing, connects the truss and the sprinkler branch pipe; opening the ball valve allows water to be delivered to the sprinkler head. The ball valve's opening and closing is controlled by the adjustable ball valve control assembly. The pressure sensor is located inside the sprinkler branch pipe and is used to measure the water pressure in the sprinkler branch pipe in real time. The control circuit is connected to the ball valve control assembly and the pressure sensor for communication.

[0031] To further optimize the technical solution, the ball valve control component mainly consists of a motor, a reduction gear, and a position sensor. The output end of the motor is connected to the ball valve via the reduction gear, and the opening degree of the ball valve is adjusted by the rotation of the motor. The position sensor is used to provide feedback on the opening degree of the ball valve, and the degree of opening of the ball valve can be precisely controlled by the position sensor.

[0032] Further optimizing the technical solution, the control circuit in the variable irrigation controller is the core unit of the entire variable irrigation controller. It adopts a simplified artificial intelligence embedded framework design, using an ARM M4 series chip as the core MCU, embedding a real-time operating system and simplified AI components. It can perform edge computing, realizing edge computing of artificial intelligence algorithms. It simultaneously has data acquisition, control, communication, and intelligent computing functions. The data acquisition function mainly collects data from position and pressure sensors. The control function adjusts the opening of the ball valve by controlling the motor rotation. The valve opening and pressure acquired in the acquisition function can be used as feedback for the control function, thereby achieving precise control of the ball valve opening. The communication function realizes data exchange between controllers through a local area wireless network. Each variable irrigation controller can acquire data such as the current target flow rate and the current pressure of each sprinkler head calculated on the irrigation distribution map. The intelligent computing function calculates the control opening of the ball valve using an intelligent valve opening calculation model based on the acquired target flow rate and the current pressure of each sprinkler head.

[0033] Further optimizing the technical solution, the irrigation gateway is a gateway device with two communication functions, supporting internet access and local area network (LAN) wireless communication. It connects to various variable irrigation controllers via LAN wireless communication, distributing irrigation targets to each plot and simultaneously acquiring current pressure data from each sprinkler head. Through internet access, it receives irrigation prescription maps from a network platform, with information on the irrigation prescription map indicating the irrigation volume for each plot to be irrigated. It also features sprinkler machine speed adjustment; when the current operating speed cannot meet the irrigation demand, it can adjust the sprinkler machine's forward speed to ensure the required irrigation volume is met.

[0034] The variable irrigation control method for large sprinkler irrigation machines of this invention adopts a distributed edge computing approach for the opening of each valve. Each valve independently calculates its opening according to the irrigation prescription map's water volume requirements, ultimately achieving the overall water volume requirement of the sprinkler irrigation machine. First, it is necessary to establish the relationship between pressure and flow rate, converting the flow rate of the sprinkler head into the pressure of the sprinkler branch pipe. Then, the branch pipe pressure is adjusted by regulating the valve opening to achieve the regulation of the water volume. Here, the flow rate is the product of the flow velocity and the irrigation time, and the irrigation time is the length of the plot divided by the traveling speed of the sprinkler irrigation machine. The relationship between different sprinklers and flow velocities follows different models, and the relationship between pressure and flow velocity can be expressed by formula (1). The coefficients a and b can be obtained experimentally according to the different sprinklers used.

[0035] F = aPb (1)

[0036] Where F is the flow velocity and P is the pressure of the nozzle branch pipe.

[0037] Because there is a strong correlation between the branch pipe pressure and valve opening on the same sprinkler machine—that is, under the same valve opening, the actual pressure of the sprinkler branch pipe is affected by the valve openings of other sprinkler branch pipes on the sprinkler machine—it is difficult to obtain the actual required flow rate using traditional methods for calculating valve opening. This invention provides an artificial intelligence calculation model based on a Bidirectional Long Short Memory (BiLSTM) network, namely a valve opening calculation model. Each core MCU has a valve opening calculation model. The input of the valve opening calculation model is the target flow rate of each other sprinkler head and the target flow rate of the current valve. The model calculates and outputs the opening of the current valve. Since a large number of sprinkler heads are simultaneously controlled on a single sprinkler machine, using all the flow rates of these sprinkler heads as input to the valve opening calculation model would result in a very large model size. Using the valve opening calculation model, some sprinkler heads with little impact on the current valve can be removed, thereby reducing the model size and achieving simplified edge computing. In the valve opening calculation model, the forget gate removes control information, the input gate adds new control information, and the output gate determines the next state output. This gating mechanism allows BiLSTM to selectively retain or ignore information, thereby improving its learning ability. Its structure is as follows: Figure 5 As shown.

[0038] The forget gate is responsible for deciding which information should be discarded, removing redundant nozzle flow information. t The calculation formula is shown in formula (2):

[0039] f t =σ(W f ·[h t-1 ,x t ]+b f (2)

[0040] Where σ is the sigmoid function, W f h is the forget gate weight matrix. t-1 x is the previous input to the cell. t For the current cell's input, b f This is the bias term for the forget gate.

[0041] The input gate determines which nozzle flow information is added to the unit state, as shown in formulas (3) and (4):

[0042] i t =σ(W i ·[h t-1 ,x t ]+b i(3)

[0043]

[0044] Where σ is the sigmoid function, W i Let h be the weight matrix of the input gate. t-1 x is the previous input to the cell. t For the current cell's input, b i W is the bias term for the input gate. C To calculate the weight matrix of cell states, b C This is a bias term used to calculate the cell state.

[0045] Current cell state C t It is the previous unit state C t-1 Multiply by the forget gate output, and add the new candidate value. Multiply by the input gate output i t As shown in formula (5):

[0046]

[0047] The output gate controls the information flow from the cell state to the hidden state, and the calculation formulas are shown in formulas (6) and (7):

[0048] o t =σ(W o ·[h t-1 ,x t ]+b o (6)

[0049] h t =o t *tanh(C t (7)

[0050] Among them, o t For the current output, W o Let b be the weight matrix of the output gate. o This is the bias term for the output gate.

[0051] Due to the flow limitations of the main pipeline of the sprinkler irrigation machine, the irrigation flow target may not be achieved in some cases. Therefore, an external attention mechanism is introduced to adjust the model output by treating the constraints as external attention.

[0052] Ultimately, the valve opening calculation model is deployed on each variable irrigation controller, and each variable irrigation controller performs more precise flow regulation by considering the flow targets and opening states of other valves through edge computing.

[0053] This invention presents a variable irrigation control method for large sprinkler irrigation machines, constructed using distributed edge computing. A variable irrigation controller is installed on each valve of the sprinkler irrigation machine. The controller acquires the flow rate of other valves on the same machine and calculates the valve opening using intelligent edge computing methods. This invention employs a lightweight, simplified artificial intelligence framework design, incorporating a real-time operating system with simplified AI support components, enabling the execution of AI algorithms on the controller. This invention uses a bidirectional long short-term memory network to construct the valve opening calculation model, ensuring calculation accuracy while discarding excessive redundant information, reducing algorithm complexity, and satisfying edge computing requirements on the variable irrigation controller. This invention enables independent adjustment of the flow rate of individual valves, allowing real-time adjustment of irrigation volume according to the irrigation prescription map, thus improving the uniformity of sprinkler irrigation.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A variable irrigation control method for a large sprinkler irrigation machine, characterized in that, include: By installing a variable irrigation controller on each sprinkler head of the sprinkler, the water output of the sprinkler head can be adjusted. The variable irrigation controllers are connected by a wireless communication network. Through the built-in valve opening adjustment algorithm of the variable irrigation controller, the valve opening can be adjusted in real time, so that the water output of the sprinkler head can be quickly stabilized at the target water output required by the irrigation prescription map. The variable irrigation controller mainly consists of a ball valve for controlling the switching of the sprinkler branch pipe, an adjustable ball valve control assembly, a pressure sensor, and a control circuit. The ball valve, controlling the branch pipe switching, connects the truss and the sprinkler branch pipe; opening the ball valve allows water to be delivered to the sprinkler head. The ball valve's switching is controlled by the adjustable ball valve control assembly. The pressure sensor is located inside the sprinkler branch pipe and is used to measure the water pressure in the sprinkler branch pipe in real time. The control circuit is connected to the ball valve control assembly and the pressure sensor for communication. The ball valve control assembly mainly consists of a motor, a reduction gear, and a position sensor. The output of the motor is connected to the ball valve via the reduction gear, and the opening degree of the ball valve is adjusted by the rotation of the motor. The position sensor is used to provide feedback on the opening degree of the ball valve. The control circuit uses an ARM M4 series chip as the core MCU, embeds a real-time operating system and a simplified AI support component, and is able to run artificial intelligence algorithms on the controller. Each core MCU has a valve opening calculation model. The input to the valve opening calculation model is the target flow rate of each other nozzle and the target flow rate of this valve. The model calculates and outputs the opening of this valve. In the valve opening calculation model, the forget gate control information is removed, the input gate control new information is added, and the output gate determines the next state output.

2. The variable irrigation control method for a large sprinkler irrigation machine according to claim 1, characterized in that, The variable irrigation controller (1) is installed on each of the nozzle (3) branches on the truss (2); the variable irrigation controller (1) is connected to the Internet by connecting to the irrigation gateway (4) installed on the main tower vehicle (5).

3. The variable irrigation control method for a large sprinkler irrigation machine according to claim 2, characterized in that, An irrigation gateway is a gateway device with two communication functions, supporting Internet access and local area network (LAN) wireless communication. It connects to various variable irrigation controllers via LAN wireless communication, distributes irrigation volume targets for each plot, and simultaneously acquires the current pressure data of each sprinkler head. It receives irrigation prescription maps from a network platform via Internet access, and the information in the irrigation prescription map is the irrigation volume for each plot to be irrigated.

Images