Drip irrigation system water pressure regulation and distributed wireless sensor network system based on Q-learning learning method

Abstract

The invention discloses a drip irrigation system water pressure regulation and distributed wireless sensor network system based on a Q-learning learning method, and solves the problem that the irrigation demand of crops is difficult to guarantee due to non-uniform water pressure distribution of an existing drip irrigation pipeline in an uneven terrain. The system judges whether the water pressures of the water outlets are the same or not, adjusts the water outlets with different water pressures, and overall plans the water pressure of the whole irrigation network according to the wireless sensor network. The drip irrigation system water pressure regulation and distributed wireless sensor network system based on the Q-learning method can be used for agricultural pipeline irrigation.

Description

Technical Field

[0001] The present application relates to agricultural automation control, wireless sensor networks, irrigation system management, and machine learning applications, and specifically to a drip irrigation system water pressure regulation and a distributed wireless sensor network system based on a Q-learning learning method. Background Art

[0002] A drip irrigation system water pressure regulation and distributed wireless sensor network system based on Q-learning learning method aims to optimize the water pressure management of the drip irrigation system through intelligent algorithms and advanced sensor technology. The system uses wireless sensor networks to monitor key parameters such as water pressure and soil moisture at each drip irrigation point in real time, and dynamically adjusts the water pressure through the Q-learning algorithm to ensure that crops receive uniform and appropriate water supply, thereby improving water resource utilization efficiency and crop yield. However, under uneven terrain conditions, there will be differences in water pressure at the outlet of the drip irrigation pipe, resulting in crops at different heights and locations unable to obtain consistent water supply, which in turn affects crop growth and yield. This has always been one of the main challenges faced by the system. Summary of the invention

[0003] In view of this, the embodiments of the present disclosure provide a drip irrigation system water pressure regulation and a distributed wireless sensor network system based on the Q-learning learning method, which at least partially solves the problems existing in the prior art.

[0004] A water pressure regulation and distributed wireless sensor network system for a drip irrigation system based on a Q-learning learning method, the system comprising: a data acquisition module, used to install a pressure sensor at each water outlet of a drip irrigation pipeline, collect water pressure data, and record terrain information at the location of the water outlet; a Q-learning learning module, used to receive the water pressure data and terrain information collected by the data acquisition module, learn based on the Q-learning algorithm, establish a correlation model between the water pressure and the terrain at the water outlet, and output a regulation instruction based on the model; a distributed wireless sensor network module, used to construct a distributed wireless sensor network according to the water pressure value generated by the Q-learning learning module, each sensor node dynamically adjusts the network topology according to its own water pressure value and the communication status of the adjacent nodes, and realizes adaptive network formation and node management; a control module, used to receive the water pressure regulation instruction output by the Q-learning learning module, and control the flow regulating device of the drip irrigation pipeline to ensure that each water outlet reaches a preset water pressure balance.

[0005] Preferably, the Q-learning module includes a feedback loop for updating the Q table according to the deviation between the actual water pressure value of each water outlet and the target water pressure value to optimize the water pressure regulation effect. The specific steps of the Q-learning module including a feedback loop are as follows: S1: Establish the state representation of the water outlet of the drip irrigation pipe.

[0006] S2: Action set design: Based on the slope elevation angle θ and valve opening ψ, design n typical valve opening data sets {ψ-1, ψ-2…ψ-n} for m typical elevation angles {θ-1, θ-2,…θm}; S3: Reward function design: Get the maximum cumulative reward from the starting point to the target point through environmental feedback and establish a reward function; S4: Action selection strategy design; first determine the size of the terrain elevation angle θ, the selection range of θ is the set m typical elevation angles {θ-1, θ-2, ... θm}; according to the size of the elevation angle, select the corresponding n typical gate opening data sets ai = {ψ-(i1), ψ-(i2), ... ψ-(in)}; then set a greedy value ε, and generate a random value p∈(0,1) each time an action is selected, which is used to determine whether the next action is a random selection or a strategy selection, to avoid the algorithm falling into a local optimum; when p is less than ε, randomly select the action value; when p is greater than ε, select the action with the largest Q value in the action set as the action selected this time; S5: Design of the Q value table initialization method; use the difference between the outlet water pressure and the target water pressure value to initialize the Q value, that is, the closer to the target value, the smaller the Q value; S6: According to the action selection strategy, select action a from the action selection set ai corresponding to the terrain elevation angle; S7: Obtain the immediate reward value Reward and the new valve opening S' after executing action a according to the reward function; S8: Update the Q value corresponding to state S and action a according to the following Q value update function; Q(S,a)←(1-α)Q(S,a)+α[r+γmaxQ(S',a)] S9: Determine whether the water pressure reaches the target value. If not, continue to select the next action according to the action selection strategy, otherwise end this round of learning.

[0007] Preferably, the distributed wireless sensor network module forms an adaptive connection relationship between nodes based on the water pressure value and terrain information of each sensor node, so as to realize dynamic topology adjustment and improve the scalability and fault tolerance of the network.

[0008] Preferably, the data acquisition module includes a 5G module for recording the geographical location of each water outlet, thereby providing accurate terrain data for analysis by the Q-learning module.

[0009] A water pressure regulation method for a drip irrigation system based on a Q-learning learning method comprises the following steps: collecting water pressure data and terrain information of each water outlet through a data acquisition module; using a Q-learning learning module to learn the water pressure data and terrain information, and constructing a correlation model between the water pressure of the water outlet and the terrain; outputting a water pressure regulation instruction based on the model, and sending it to a control module to adjust the water pressure of each water outlet; updating the Q table of the Q-learning model according to the real-time water pressure feedback information of each water outlet, and gradually optimizing the water pressure regulation effect.

[0010] Preferably, through the distributed wireless sensor network module, a distributed wireless sensor network based on the water pressure value of each water outlet is constructed, so that the network topology can be adaptively adjusted according to the change of water pressure data.

[0011] Preferably, when constructing a decentralized wireless sensor network, the sensor nodes reorganize the network according to their own water pressure values ​​and the connection status of adjacent nodes to adapt to network coverage requirements under different terrain conditions.

[0012] The disclosed embodiment provides a drip irrigation system water pressure regulation and distributed wireless sensor network system based on the Q-learning learning method, the system comprising: a data acquisition module, used to install a pressure sensor at each water outlet of the drip irrigation pipe, collect water pressure data, and record the terrain information of the location of the water outlet; a Q-learning learning module, used to receive the water pressure data and terrain information collected by the data acquisition module, learn based on the Q-learning algorithm, establish a correlation model between the water pressure and the terrain at the water outlet, and output a regulation instruction based on the model; a distributed wireless sensor network module, used to build a distributed wireless sensor network according to the water pressure value generated by the Q-learning learning module, each sensor node dynamically adjusts the network topology according to its own water pressure value and the communication status of the adjacent nodes, and realizes adaptive network formation and node management; a control module, used to receive the water pressure regulation instruction output by the Q-learning learning module, and control the flow regulating device of the drip irrigation pipe to ensure that each water outlet reaches a preset water pressure balance.

[0013] According to the system of claim 1, the Q-learning module includes a feedback loop for updating the Q table according to the deviation between the actual water pressure value and the target water pressure value of each water outlet to optimize the water pressure regulation effect. According to the system of claim 1, the distributed wireless sensor network module forms an adaptive connection relationship between nodes based on the water pressure value and terrain information of each sensor node to achieve dynamic topological structure adjustment and improve the scalability and fault tolerance of the network. According to the system of claim 1 or 2, the data acquisition module includes a GPS module for recording the geographical location of each water outlet, thereby providing accurate terrain data for analysis by the Q-learning module. A method for regulating water pressure of a drip irrigation system based on a Q-learning learning method comprises the following steps: collecting water pressure data and terrain information of each water outlet through a data acquisition module; learning the water pressure data and terrain information using a Q-learning learning module to construct a correlation model between water pressure and terrain of the water outlet; outputting a water pressure regulation instruction based on the model and sending it to a control module to adjust the water pressure of each water outlet; updating the Q table of the Q-learning model according to the real-time water pressure feedback information of each water outlet, and gradually optimizing the water pressure regulation effect. According to the method of claim 5, a distributed wireless sensor network based on the water pressure value of each water outlet as a node is constructed through the distributed wireless sensor network module, so that the network topology can be adaptively adjusted according to the change of water pressure data. According to the method of claim 6, when constructing a distributed wireless sensor network, the sensor node reorganizes the network according to its own water pressure value and the connection status of the adjacent nodes to adapt to the network coverage requirements under different terrain conditions. Through the solution of the embodiment of the present disclosure, it is possible to solve the problem that the different water pressures at the outlet of the drip irrigation pipe under uneven terrain conditions affect the growth of crops. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the exemplary implementation methods of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the embodiments of the present disclosure and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 The present invention is a flow chart of water pressure regulation of drip irrigation system and distributed wireless sensor network system based on Q-learning method.

[0016] Figure 2 It is a flowchart of the Q-learning learning module. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the embodiments of the present disclosure more clear, the embodiments of the present disclosure are further described in detail in combination with the embodiments and drawings. The schematic implementation methods of the embodiments of the present disclosure and their descriptions are only used to explain the embodiments of the present disclosure and are not intended to limit the embodiments of the present disclosure.

[0018] Next, referring to the accompanying drawings, the present invention describes how a drip irrigation system water pressure regulation and distributed wireless sensor network system based on the Q-learning learning method solves the problem of different water pressures at the outlet of the drip irrigation pipe under uneven terrain conditions affecting crop growth through various steps.

[0019] In the first step, the data acquisition module installs pressure sensors at each outlet of the drip irrigation pipe to collect water pressure data and record the terrain information of the outlet location. In this step, each outlet is equipped with a pressure sensor to monitor the water pressure. At the same time, the 5G module records the geographical location of each outlet and provides accurate terrain data. These data are crucial for subsequent learning and adjustment. For example, if a water outlet is located at a higher place, its water pressure is usually lower than that of other lower outlets. This difference in terrain will cause crops in different locations to receive different amounts of irrigation water, which in turn affects crop growth. Specifically, suppose a drip irrigation pipe is laid on a hill, where outlet A is located at the top of the hill and outlet B is located at the foot of the hill. At this time, the data acquisition module will record the low water pressure value displayed by the pressure sensor at A and the high water pressure value displayed by the pressure sensor at B, and transmit them to the subsequent learning module in combination with the geographical location information of the two.

[0020] In the second step, the Q-learning module is used to learn the received water pressure data and terrain information to build a correlation model between the water pressure at the outlet and the terrain. In this process, the Q-learning algorithm finds out the changing rules of the water pressure at the outlet under different terrain conditions by learning a large amount of historical data, and then establishes a correlation model. This model uses water pressure as the dependent variable and terrain parameters (such as altitude and slope) as independent variables to predict the expected water pressure value under different terrain conditions. For example, based on historical data, the Q-learning module may find that the water pressure drops by 0.1 MPa for every 1 meter increase in the height of the outlet. Specifically, when the system receives new terrain data, such as the altitude of a water outlet of 50 meters, the Q-learning model will predict the expected water pressure value at this location to be 0.9 MPa (assuming that the water pressure value at the reference location is 1.4 MPa), and compare it with the actual measured water pressure value to obtain the deviation.

[0021] The third step is to output a water pressure adjustment instruction based on the model and send it to the control module to adjust the water pressure of each water outlet. In this step, after the control module receives the adjustment instruction output by the Q-learning learning module, it adjusts the water pressure of each water outlet by controlling the flow regulating device in the drip irrigation pipe. This step ensures that even if the terrain is uneven, the water pressure of all water outlets can be maintained within the preset target value range, thereby ensuring the consistency of irrigation. For example, if the Q-learning model predicts that the water pressure of a water outlet at a high ground is 0.8 MPa, and the target value is 1.0 MPa, the system will issue an adjustment instruction to increase the water pressure of the water outlet. Specifically, the control module may increase the water pressure at the high-pressure position to the target value by adjusting the valve opening, while reducing the water pressure at the low position to maintain the balance of the entire system.

[0022] The fourth step is to update the Q-table of the Q-learning model based on the real-time water pressure feedback information of each water outlet, and gradually optimize the water pressure regulation effect. The Q-learning module continuously receives the real-time water pressure data of each water outlet, compares it with the target value, and updates the Q value in the Q-table according to the deviation. The higher the Q value, the greater the probability of taking a certain action, and eventually the system gradually learns the optimal water pressure regulation strategy under different terrain conditions. For example, if the water pressure at a water outlet on a high ground is still insufficient after adjustment, the system will continue to increase the regulation intensity at that location until the target value is reached. Specifically, the Q-learning module will record the effect after each adjustment, and through iterative learning, gradually optimize the water pressure regulation plan to ensure the long-term stability of the entire system.

[0023] The fifth step is to construct a distributed wireless sensor network based on the water pressure value of each water outlet as the node through the distributed wireless sensor network module, so that the network topology can be adaptively adjusted according to the change of water pressure data. This module is not only responsible for transmitting data, but also realizes the dynamic reorganization of the network. When the water pressure of a node changes, the network topology will be adjusted accordingly to ensure the efficient transmission of signals and the stable operation of the system. For example, in high-altitude areas, if the water pressure of a node drops significantly, the surrounding nodes will dynamically adjust the network connection to form a more stable transmission path. Specifically, assuming that a node A is located on a steeper hillside and there are multiple nodes around it, if the water pressure value of A is abnormally low, other nodes in the network will adjust the route to bypass the faulty node A to maintain the connectivity of the entire network.

[0024] Step 6: When building a distributed wireless sensor network, the sensor nodes reorganize the network according to their own water pressure values ​​and the connection status of adjacent nodes to adapt to the network coverage requirements under different terrain conditions. This adaptive adjustment not only improves the flexibility of the network, but also enhances its fault tolerance. Even if some nodes fail, other nodes can quickly fill the gap. The water pressure regulation and distributed wireless sensor network system of a drip irrigation system based on a Q-learning method of the present invention includes the following main steps, aiming to solve the technical problem that the different water pressures at the outlet of the drip irrigation pipe affect the growth of crops under the condition of uneven terrain: Data collection: Install pressure sensors at each outlet of the drip irrigation pipe to collect water pressure data at each outlet in real time. At the same time, the data collection module records the geographical location of each outlet through the built-in 5G module to obtain accurate terrain information. These data will be used for subsequent analysis and adjustment processes.

[0025] Q-learning: After receiving the water pressure data and terrain information transmitted by the data acquisition module, the Q-learning module learns based on the Q-learning algorithm. This module establishes a correlation model between the water pressure at the outlet and the terrain, and analyzes the ideal water pressure distribution at each outlet under different terrain conditions. Through continuous learning and optimization, the model can gradually improve the accuracy of water pressure regulation.

[0026] Network construction: The distributed wireless sensor network module uses the water pressure value of each water outlet as node information to build a dynamic wireless sensor network. Each node (i.e., water outlet) in the network can dynamically adjust the network topology according to its own water pressure value and the communication status of adjacent nodes. In this way, even if some nodes lose contact due to terrain changes, the network can still maintain stability and adaptive capabilities.

[0027] Water pressure regulation: The control module receives the water pressure regulation command output by the Q-learning module and controls the flow regulation device of the drip irrigation pipe. Specifically, the control module ensures that the water pressure of each outlet reaches the preset balance state by adjusting the flow valve or other regulating equipment. Through this precise control, each outlet can output stable water pressure under different terrain conditions to meet the needs of crop growth.

[0028] Feedback optimization: In order to further improve the accuracy and effect of water pressure regulation, the system also includes a feedback loop. In actual applications, the real-time water pressure feedback information of each outlet will be continuously transmitted back to the Q-learning learning module. By comparing the deviation between the actual water pressure value and the target water pressure value, the Q-learning model continuously updates the Q table and gradually optimizes the water pressure regulation strategy.

[0029] Through the above steps, the invention can achieve precise water pressure control in complex terrain environments, ensuring that the drip irrigation system provides uniform and appropriate amounts of water to crops, effectively solving the problem of inconsistent water pressure caused by uneven terrain, and improving irrigation efficiency and crop growth quality.

[0030] The above is a preferred embodiment of the present application. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A drip irrigation system water pressure regulation and distributed wireless sensor network system based on Q-learning learning method. The system first determines whether the water pressure of each water outlet is the same through the data acquisition module. If the water pressure of each water outlet is the same, it determines whether the water pressure value meets the irrigation demand; if the water pressure of each water outlet is different, the data acquisition module feeds back the outlet with different water pressure values, and then the Q-learning learning module adjusts the outlet valve. Finally, the distributed wireless sensor network module adaptively adjusts the entire pipeline outlet; A drip irrigation system water pressure regulation and distributed wireless sensor network system based on Q-learning method, characterized in that: The system includes: a data acquisition module, which is used to install pressure sensors at each water outlet of the drip irrigation pipeline, collect water pressure data, and record the terrain information of the location of the water outlet; The Q-learning module is used to receive the water pressure data and terrain information collected by the data acquisition module, learn based on the Q-learning algorithm, establish a correlation model between the water pressure at the water outlet and the terrain, and output adjustment instructions based on the model; the distributed wireless sensor network module is used to build a distributed wireless sensor network according to the water pressure value generated by the Q-learning module. Each sensor node dynamically adjusts the network topology according to its own water pressure value and the communication status of the adjacent nodes to achieve adaptive network formation and node management; the control module is used to receive the water pressure adjustment instruction output by the Q-learning module and control the flow adjustment device of the drip irrigation pipe to ensure that each water outlet reaches the preset water pressure balance.

2. The system according to claim 1, characterized in that The Q-learning module includes a feedback loop for updating the Q table according to the deviation between the actual water pressure value and the target water pressure value of each water outlet to optimize the water pressure regulation effect; The Q-learning module contains a feedback loop with the following steps: S1: Establish the state representation of the water outlet of the drip irrigation pipeline; S2: Action set design: Based on the slope elevation angle θ and valve opening ψ, design n typical valve opening data sets {ψ-1, ψ-2…ψ-n} for m typical elevation angles {θ-1, θ-2,…θm}; S3: Reward function design: Get the maximum cumulative reward from the starting point to the target point through environmental feedback and establish a reward function; S4: Action selection strategy design; first determine the size of the terrain elevation angle θ, the selection range of θ is the set m typical elevation angles {θ-1, θ-2, ... θm}; according to the size of the elevation angle, select the corresponding n typical gate opening data sets ai = {ψ-(i1), ψ-(i2), ... ψ-(in)}; then set a greedy value ε, and generate a random value p∈(0,1) each time an action is selected, which is used to determine whether the next action is a random selection or a strategy selection, to avoid the algorithm falling into a local optimum; when p is less than ε, randomly select the action value; when p is greater than ε, select the action with the largest Q value in the action set as the action selected this time; S5: Design of the Q value table initialization method; use the difference between the outlet water pressure and the target water pressure value to initialize the Q value, that is, the closer to the target value, the smaller the Q value; S6: According to the action selection strategy, select action a from the action selection set ai corresponding to the terrain elevation angle; S7: Obtain the immediate reward value Reward and the new valve opening S' after executing action a according to the reward function; S8: Update the Q value corresponding to state S and action a according to the following Q value update function; Q(S,a)←(1-α)Q(S,a)+α[r+γmaxQ(S',a)] S9: Determine whether the water pressure reaches the target value. If not, continue to select the next action according to the action selection strategy, otherwise end this round of learning.

3. The system according to claim 1, characterized in that The distributed wireless sensor network module forms an adaptive connection relationship between nodes based on the water pressure value and terrain information of each sensor node, so as to realize dynamic topological structure adjustment and improve the scalability and fault tolerance of the network.

4. The system according to claim 1 or 2, characterized in that: The data acquisition module includes a 5G module, which is used to record the geographical location of each water outlet, thereby providing accurate terrain data for analysis by the Q-learning module.

5. A water pressure regulation method for a drip irrigation system based on Q-learning method, characterized in that: The following steps are involved: Collect water pressure data and terrain information of each water outlet through the data acquisition module; The water pressure data and terrain information are learned using a Q-learning module to construct a correlation model between the water pressure at the water outlet and the terrain; a water pressure adjustment instruction is output based on the model and sent to a control module to adjust the water pressure at each water outlet; the Q table of the Q-learning model is updated according to the real-time water pressure feedback information of each water outlet to gradually optimize the water pressure adjustment effect.

6. The method according to claim 5, characterized in that Through the distributed wireless sensor network module, a distributed wireless sensor network based on the water pressure value of each water outlet is constructed, so that the network topology can be adaptively adjusted according to the change of water pressure data.

7. The method according to claim 6, characterized in that When constructing a decentralized wireless sensor network, the sensor nodes reorganize the network according to their own water pressure values ​​and the connection status of adjacent nodes to adapt to the network coverage requirements under different terrain conditions.

Images