Automatic water resource optimization scheduling management method based on irrigation area demonstration area

Abstract

The invention belongs to the technical field of irrigation scheduling management, and particularly discloses an automatic water resource optimization scheduling management method based on an irrigation area demonstration area, and the method comprises the steps: obtaining N irrigation areas in a current concerned irrigation area, and setting a scheduling set containing a plurality of water irrigation modes for each irrigation area; constructing a selection optimization algorithm for selecting water irrigation modes one by one for all irrigation areas at each moment; acquiring a local state set of each irrigation area in real time, and inputting all the local state sets into a selection optimization algorithm so as to select water irrigation modes of all the irrigation areas at the same moment; the method has the advantages that the adaptability of the irrigation system is enhanced by acquiring the local state set in real time and applying the selection optimization algorithm, the water resource configuration is optimized, and the irrigation efficiency and the crop growth are ensured. Through intelligent management, waste of water resources is reduced, the cost is reduced, and the economic and environmental sustainability of agricultural production is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of irrigation scheduling management, and in particular to an automated water resource optimization scheduling management method based on an irrigation demonstration area. Background Art

[0002] In traditional agricultural irrigation systems, irrigation areas mainly rely on fixed channel water sources for water supply. These systems are designed with conventional climatic conditions in mind and can meet basic irrigation needs in years with sufficient water. However, these systems show obvious shortcomings in extreme drought conditions. Especially in the case of drought-induced water shortages, water source scheduling relying on fixed channels often cannot meet the needs of all irrigation areas because these systems lack the ability to flexibly adjust water sources and optimize water flow paths. In addition, due to the lack of efficient resource allocation strategies and rapid response mechanisms to drought, traditional irrigation systems are often unable to reasonably allocate limited water resources, resulting in excess water resources in some areas and severe water shortages in other areas.

[0003] Therefore, an automated water resources optimization scheduling management method based on the irrigation demonstration area is proposed to solve the above-mentioned problems. Summary of the invention

[0004] The present invention aims to provide an automated water resource optimization scheduling management method based on an irrigation demonstration area to solve or improve the above-mentioned technical problem that traditional agricultural irrigation systems are often unable to effectively cope with water resource shortages under extreme drought conditions due to the lack of flexible water source scheduling and efficient resource allocation capabilities.

[0005] In view of this, a first aspect of the present invention is to provide an automated water resource optimization scheduling management method based on an irrigation demonstration area.

[0006] The first aspect of the present invention provides an automated water resource optimization scheduling management method based on an irrigation demonstration area, comprising the following steps: obtaining N irrigation areas within the current irrigation area of ​​interest, and establishing a scheduling set containing multiple water irrigation methods for each of the irrigation areas; constructing a selection optimization algorithm for selecting the water irrigation method for all the irrigation areas one by one at each moment; obtaining a local state set for each of the irrigation areas in real time, and inputting all of the local state sets into the selection optimization algorithm to select the water irrigation method for all of the irrigation areas at the same moment; the step of the selection optimization algorithm selecting the water irrigation method for the irrigation area includes: constructing an action space set for each of the irrigation areas through the type of irrigation water allocation and water irrigation method, and constructing an independent local network for each of the irrigation areas; obtaining a state set of the irrigation area according to the local state set, and calculating the action value of all actions of the irrigation area in the action space set through the local network; retaining the action with the highest action value, and adjusting water for the current irrigation area according to the irrigation water allocation and water irrigation method contained in the retained action.

[0007] In any of the above technical solutions, the water irrigation methods of the scheduling set include channel water source call, path optimal call, water source optimal call and water volume optimal call; the channel water source call step includes: obtaining all channel paths in the current irrigation area, opening the currently obtained channel path until the channel water levels of all current channel paths are consistent; the path optimal adjustment step includes: obtaining the optimal path between the water pump and all grid fields in the current irrigation area, and irrigating the grid fields according to the optimal path; the water source optimal call step includes: obtaining the callable irrigation area adjacent to the current irrigation area, and irrigating the current irrigation area through the callable irrigation area; the water volume optimal call step includes: obtaining the required water volume and the callable water volume of all the irrigation areas, and irrigating all the irrigation areas through an external water source according to the difference between the required water volume and the callable water volume.

[0008] In any of the above technical solutions, the step of retaining the action with the highest action value includes: Among them, the is the action with the highest action value; is the action space set, and the Allocate water for irrigation, For water irrigation; is the number of the irrigation area, and is a positive integer from 1 to N; is the state set; is the parameter used to estimate the expected reward of each action; is the value of the action.

[0009] In any of the above technical solutions, the selection optimization algorithm includes the local networks of all the irrigation areas and the global network constructed by mixing the local networks of all the irrigation areas, and has the following situations: Scenario 1: Obtain a comprehensive action value according to the action values ​​of all local networks at each unit time, and update the global network according to the comprehensive action value; In scenario 2, the update of the local network in the next time unit takes into account the influence of the global network on the current local network.

[0010] In any of the above technical solutions, at each unit time, the local network of each irrigation area is updated by an action value, specifically including: Among them, the is the learning rate; is the discount factor; The benefit value of the current action is adopted for the irrigation area; is the state set for the next unit time; is the action for the next unit time; is the maximum action value among all actions in the next state set; is the adjustment coefficient, which is used to control the influence of the global network on the local network update; is the gradient of the global network relative to the local network, which is used to indicate the influence of the global network on the local network; is the comprehensive action value of the global network; is the current action value.

[0011] In any of the above technical solutions, the comprehensive action value is obtained as follows: Among them, the is a weight matrix used to linearly transform the action value of the local network; is a physical state measurement value vector, and includes at least one item in the local state set; is the weight matrix used to adjust The impact on the calculation of the comprehensive action value; A one-hot encoding vector representing the water irrigation method of the local network; is a weight matrix used to adjust the influence of water irrigation mode in the calculation of comprehensive action value; is a bias term used in the first layer of the global network; is an activation function used to introduce nonlinearity to help the global network capture complex relationships; is the weight matrix of the second layer, and is a bias term, which is used together to convert the output of the hidden layer h into a comprehensive action value.

[0012] In any of the above technical solutions, at each unit time, the global network is updated by the comprehensive action value, specifically including: calculating the difference between the comprehensive action value of the current time unit and the predicted comprehensive action value of the next time unit; using the corrected difference value to update the network parameters of the global network through the loss function through the gradient descent method; and calculating the gradient of the loss function relative to each parameter through the back propagation algorithm to reduce the loss.

[0013] In any of the above technical solutions, the loss function includes the following formula: Among them, the is a loss function used to characterize the deviation between the predicted comprehensive action value and the actual obtained value; is a global reward; is the discount factor; is the expected comprehensive action value under all possible actions in the next state set; is the comprehensive action value estimate of the current state set; is the network parameter of the current global network; are the parameters of the target network.

[0014] In any of the above technical solutions, the comprehensive action value is obtained by weighted aggregation of the action values ​​of all local networks, and the aggregate weight of each action value in the next time unit is determined by the corresponding local state set in the current time unit.

[0015] Compared with the prior art, the present invention has the following beneficial effects: By acquiring the local state set of each irrigation area in real time and dynamically selecting the most appropriate irrigation method based on this data using a selection optimization algorithm, the present invention greatly enhances the adaptability and responsiveness of the irrigation system. This enables the system to adjust the irrigation strategy in real time, better cope with different environmental conditions, and optimize the use of water resources.

[0016] By constructing a scheduling set, a variety of irrigation options are provided for each irrigation area, and the optimal water resource allocation is determined in combination with the current local state set. This method can ensure the optimal use of water resources across the entire region, reduce waste, and improve irrigation efficiency.

[0017] By precisely controlling the amount of water and irrigation method for each irrigation area, the present invention helps ensure that crops receive appropriate water supply, promotes healthy growth of crops, and thus improves yield and quality. In addition, by optimizing the use of water resources, it also helps protect the ecological environment and supports sustainable agricultural practices.

[0018] Intelligent irrigation management reduces the possibility of over-irrigation and water waste, thereby reducing water costs and maintenance expenses. At the same time, more precise water management also helps farmers increase crop yields and increase economic benefits.

[0019] Additional aspects and advantages of embodiments according to the present invention will become apparent in the following description or may be learned through practice of embodiments according to the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 The figure is a flow chart of the method steps of the present invention. DETAILED DESCRIPTION

[0021] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0022] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0023] See also Figure 1 , the following describes some embodiments of the present invention based on the automated water resources optimization scheduling management method of the irrigation district demonstration area.

[0024] The embodiment of the first aspect of the present invention proposes an automated water resource optimization scheduling management method based on an irrigation demonstration area. In some embodiments of the present invention, such as Figure 1 As shown, the automated water resources optimization scheduling management method includes: S101, obtaining N irrigation areas in the current focused irrigation area, and establishing a scheduling set including multiple water-using irrigation methods for each irrigation area.

[0025] As can be seen from the above, all irrigation areas within the current irrigation area of ​​interest are identified and obtained. Specifically, this involves a detailed geographical and resource assessment of the irrigation area, including the soil type, crop demand, and historical irrigation data of each irrigation area. These data provide the necessary basic information for subsequent water resource scheduling, ensuring that all water-demanding areas are included in the management scope. Subsequently, a scheduling set containing multiple water-using irrigation methods is established for each irrigation area, which is another core component of the present invention. The irrigation methods included in the scheduling set may cover a variety of methods such as drip irrigation, sprinkler irrigation, and underground infiltration irrigation, each of which has its specific applicable conditions and efficiency. Through such a design, the management system can dynamically select the irrigation method that best suits the current environment and needs based on real-time monitoring of soil moisture, weather forecasts, water resource availability and other parameters. For example, when water resources are tight, the system may give priority to drip irrigation technology with high water utilization efficiency; when water is sufficient, sprinkler irrigation technology with a wider coverage may be used.

[0026] Use information technology and artificial intelligence algorithms to optimize water source scheduling and irrigation routes. The system integrates a sensor network to monitor the water and soil conditions of each irrigation area in real time, and uses satellite images and meteorological data to predict future water needs. Based on this data, the scheduling system can automatically calculate the optimal water resource allocation plan and issue irrigation instructions through the control center to ensure that each area can get the optimal amount of water it needs. In addition, the system has the ability to respond quickly to drought. When the risk of drought is detected to increase, the system can automatically adjust the irrigation strategy, such as increasing the frequency of irrigation, adjusting the irrigation time to night to reduce evaporation, or launching an emergency water source plan. These intelligent response measures have greatly enhanced the irrigation system's ability to adapt to extreme climates and reduced the risk of crop yield reductions caused by poor water resource management.

[0027] Specifically, the water irrigation methods of the scheduling set include channel water source call, optimal path call, optimal water source call and optimal water volume call.

[0028] The steps of channel water source calling include: obtaining all channel paths in the current irrigation area, opening the currently obtained channel paths, until the channel water levels of all current channel paths are consistent.

[0029] The steps of optimal path adjustment include: obtaining the optimal path between the water pump and all grid fields in the current irrigation area, and irrigating the grid fields according to the optimal path.

[0030] The step of optimally calling water sources includes: obtaining a callable irrigation area adjacent to the current irrigation area, and irrigating the current irrigation area through the callable irrigation area.

[0031] The steps of optimal water call include: obtaining the required water volume and available water volume of all irrigation areas, and irrigating all irrigation areas through external water sources according to the difference between the required water volume and the available water volume.

[0032] As can be seen from the above, channel water source call mainly focuses on the effective use of existing channels in the irrigation area. This method includes obtaining all available channel paths in the current irrigation area, ensuring that these paths are effectively opened, and adjusting the water levels of each channel to be consistent to achieve uniform distribution of water sources. Specific steps: First, the system will collect channel path information in all current irrigation areas; then the system will send instructions to relevant control devices to open these channel paths; the system will monitor and adjust the water level in the channel until the water level of all channels is consistent, ensuring that the water flow is smoothly transferred between channels to avoid excess or insufficient water resources in some areas. In a large farmland, multiple irrigation channels run through the entire area. When it is detected that the water level in some areas is lower than the standard water level, the system will automatically open the water supply of the corresponding channel until the water level of all channels reaches the preset equilibrium state.

[0033] The optimal path call determines the optimal water flow path between the water pump and each grid field through an algorithm. This method aims to accurately calculate and implement the most effective water resource transmission path through technical means, and achieve the highest water resource utilization efficiency at the lowest cost. Specific steps: The algorithm analyzes the connection path between the water pump and all grid fields in the current irrigation area, and selects the shortest or lowest cost path; water resources are allocated and irrigation operations are performed according to these optimal paths. In an arid area, in order to maximize the benefits of limited water resources, the system calculates the shortest distance from the main water pump to each irrigation grid field, and implements on-demand irrigation through an automated control system.

[0034] Optimal water resource allocation focuses on cross-regional water resource sharing and scheduling. This method supplements the water resource demand of the current region by identifying and utilizing available water sources in adjacent irrigation areas, and is particularly suitable for water shortages in certain regions. Specific steps: The system evaluates and determines the available water sources in other areas adjacent to the current irrigation area; these additional water sources are introduced into the current area through the control system to alleviate the water shortage problem. When an irrigation area encounters an unexpected drought and there is surplus water in the adjacent area, the system will automatically start the operation of diverting water from the adjacent area to ensure that the water needs of the affected area are met.

[0035] Optimal water allocation implements precise allocation of water resources by comprehensively considering the difference between the required water volume and the available water volume in all irrigation areas. This approach ensures the maximum utilization of water resources through scientific calculations and strategies under the condition of limited resources. Specific steps: The system first obtains the water demand and existing water volume of all irrigation areas; the system determines the necessary water allocation strategy based on the difference between demand and supply, and meets the needs of all areas through external water sources or internal allocation. In the dry season when the total water volume in the entire region is not enough to meet all needs, the system will calculate the gap in each area and give priority to ensuring the water demand of key crop areas, while possibly reducing the water supply to non-key areas.

[0036] S102, constructing a selection optimization algorithm for selecting water irrigation methods for all irrigation areas one by one at each moment.

[0037] From the above, we can know that we can build a selection optimization algorithm, which is specifically used to select the most suitable water irrigation method for all irrigation areas at each moment. The core of the design of this algorithm is to intelligently respond to changing environmental conditions to ensure that at any given moment, each irrigation area can use the most optimized irrigation method to dispatch and manage water resources, so as to achieve the purpose of saving water and increasing efficiency and promoting crop growth. It can analyze the specific needs and external environmental factors of each irrigation area in real time, such as soil moisture, meteorological conditions, water source conditions and crop types, and select the most suitable irrigation method accordingly. This includes but is not limited to drip irrigation, sprinkler irrigation, flood irrigation, etc. Each method has different performance advantages for water efficiency and crop needs under specific conditions.

[0038] S103, obtaining the local state set of each irrigation area in real time, and inputting all the local state sets into the selection optimization algorithm to select the water irrigation mode for all irrigation areas at the same time.

[0039] As mentioned above, real-time data of each irrigation area, such as soil moisture, temperature, crop growth status, water level and other key parameters, are continuously monitored and collected. These data are the basis for understanding the current needs of each area and the key basis for making irrigation decisions. In a wide agricultural production area, sensors are deployed at various key points to collect soil moisture and temperature data, which are sent to the central processing system in real time via wireless networks every few minutes.

[0040] The collected data is summarized and input into the selection optimization algorithm. The algorithm uses this data to evaluate and determine the most appropriate irrigation method for each area. The algorithm takes into account the weights and priorities of various factors when processing the data to ensure the rational allocation and use of water resources. If the soil moisture in a certain area is below a predetermined threshold and the weather forecast shows no future rainfall, the selection optimization algorithm will automatically recommend starting a drip irrigation system to precisely control the water directly to the roots while reducing water waste.

[0041] By acquiring the local state set of each irrigation area in real time and using the selection optimization algorithm for data analysis and decision-making, this process provides strong data support and intelligent decision-making capabilities for irrigation management. This approach enables the irrigation system to dynamically adjust irrigation strategies to adapt to changing environmental conditions and crop needs, thereby maximizing the utilization efficiency of water resources and optimizing the growth environment of crops.

[0042] Specifically, the steps of selecting the water irrigation method for the irrigation area by the optimization algorithm include: The action space set of each irrigation area is constructed by the irrigation water allocation and the type of irrigation method, and an independent local network is constructed for each irrigation area.

[0043] The state set of the irrigation area is obtained according to the local state set, and the action values ​​of all actions of the irrigation area in the action space set are calculated through the local network.

[0044] The action with the highest action value is retained, and water is transferred to the current irrigation area according to the irrigation water allocation and water irrigation method contained in the retained action.

[0045] Based on the above specific description, define and set the range of all possible irrigation actions for each irrigation area, including different irrigation methods (such as drip irrigation, sprinkler irrigation, etc.) and their corresponding water allocation. The construction of the action space set is based on the water efficiency and adaptability of a specific irrigation method, providing all feasible options for subsequent decision-making. For example, for an irrigation area that is often affected by drought, its action space set may include water-efficient drip irrigation and low-water-demand spray irrigation, and each method will be configured with a specific water allocation strategy.

[0046] After obtaining the action space set, the algorithm will calculate the expected effect of each possible action, that is, the action value, based on the current local state set (including data such as soil moisture, weather conditions, and crop growth stage). This step is completed through independent local networks in each irrigation area, which use machine learning techniques to evaluate the performance of various actions under current conditions. In an irrigation area, the local network may analyze the action value of drip irrigation and fixed water supply under the current high temperature and drought conditions to determine which irrigation method can more effectively use water resources and support crop growth.

[0047] Based on the calculation results of the local network, the algorithm will select the irrigation strategy with the highest action value and implement the corresponding water allocation and irrigation method. This ensures that each irrigation decision is data-driven and can achieve the best resource utilization efficiency and crop growth conditions. If the analysis results show that drip irrigation has a higher action value than traditional surface irrigation in persistent hot weather, the system will automatically adjust the irrigation settings, implement the drip irrigation strategy, and irrigate according to the calculated optimal water volume.

[0048] As can be seen above, the selection optimization algorithm not only optimizes the allocation of water resources, but also improves irrigation efficiency and crop productivity. This process significantly enhances the adaptability of the irrigation system through real-time data analysis and intelligent decision-making, enabling it to flexibly respond to various environmental changes. In addition, by automatically selecting the most appropriate irrigation method, the algorithm reduces water waste and supports sustainable agricultural practices.

[0049] Specifically, the steps of retaining the action with the highest action value include: in, The action with the highest action value; is the action space set, and Allocate water for irrigation, for water irrigation methods; is the number of the irrigation area, and is a positive integer from 1 to N; is a state set; is the parameter used to estimate the expected reward of each action; is the action value.

[0050] As can be seen from the above, it involves constructing an action space set containing different irrigation methods and water configurations for each irrigation area. The action space set for each irrigation area consists of two parts: one is the different irrigation methods, and the other is the water configuration corresponding to each irrigation method. Provide a comprehensive decision-making basis for the algorithm, including optional irrigation methods and corresponding water quotas. Next, the value of each action under a given local state set is evaluated through the local network. The local state set includes key parameters such as soil moisture, crop demand, and weather conditions, which are most suitable for determining which irrigation method is most appropriate. The local network evaluates the potential benefits of each action based on these parameters and calculates the value of each action. Finally, the action with the highest action value is retained, and the current irrigation area is adjusted according to the irrigation water allocation and water irrigation method of this action. This means that the system automatically executes the irrigation configuration with the highest action value to ensure that each irrigation area uses water resources in the most efficient way.

[0051] By intelligently analyzing the specific needs and environmental conditions of each irrigation area and then selecting the most appropriate irrigation method and water allocation, it not only improves the efficiency of water resource use, but also ensures that crops can grow under optimal conditions, thereby improving the overall benefits and sustainability of agricultural production.

[0052] Furthermore, the system status is , and each local state Includes: Soil moisture , Current water supply , Crop water requirement , precipitation , and historical irrigation volume . Global State Includes: Channel Status , Pump station status and other water source conditions .

[0053] Furthermore, irrigation water allocation Indicates the irrigation water allocation strategy for the area; scheduling method selection Select one of the four scheduling methods: channel water source priority, route optimal calculation, water source optimal calculation and water volume optimal calculation.

[0054] Further, the selection optimization algorithm includes the local networks of all irrigation areas and the global network constructed by mixing the local networks of all irrigation areas, and has the following situations: In the first scenario, the comprehensive action value is obtained based on the action value of all local networks at each unit time, and the global network is updated through the comprehensive action value.

[0055] In the second scenario, the update of the local network in the next time unit takes into account the influence of the global network on the current local network.

[0056] As can be seen above, each irrigation area has a local network that calculates the value of different irrigation actions based on the specific environmental conditions and needs of the area, such as soil moisture, crop type, weather conditions, etc. The local network provides an action value to guide irrigation decision-making by evaluating the potential benefits of each irrigation method and its supporting water volume. The global network integrates the action values ​​of all local networks to build a comprehensive perspective and evaluate the comprehensive action value of the entire irrigation area. This not only considers the optimal solution for a single area, but also considers how to adjust the resource allocation of the entire system to achieve the optimal use of water resources in the entire irrigation area.

[0057] In view of the description of scenario 1, at each unit time, the global network summarizes the action values ​​of all local networks for their respective irrigation areas. These action values ​​are calculated based on the adaptability and efficiency of each irrigation method, including the availability of channel water sources, the optimal irrigation path, the available water sources, and the optimal configuration of water volume. The global network calculates a comprehensive action value through these action values, which guides the irrigation strategy of the entire irrigation area to ensure the highest water resource utilization efficiency and crop growth support throughout the area.

[0058] In the description of scenario 2, the local networks not only reflect the immediate status and needs of their respective regions, but also consider the feedback from the global network at each time unit. This design enables each local decision to be optimized under the guidance of the global network, ensuring that local decisions are consistent with the overall water resource management strategy. If the global network indicates that a region needs to reduce water use to support the stability of the entire system, the corresponding local network will adjust its strategy, possibly switching from optimal water source call to optimal water volume call to adapt to this change.

[0059] Specifically, at each unit time, the local network of each irrigation area is updated by the action value, including: in, is the learning rate; is the discount factor; Adopt the benefit value of the current action for the irrigation area; is the state set for the next unit time; The action for the next unit of time; The maximum action value among all actions in the next state set; is the adjustment coefficient, which is used to control the influence of the global network on the local network update; is the gradient of the global network relative to the local network, which is used to indicate the influence of the global network on the local network; is the comprehensive action value of the global network; is the current action value.

[0060] In response to the above specific description, the decision of the irrigation area is updated at each time unit. Specifically, this includes calculating and updating the expected effect of each irrigation area, that is, the action value of the area, which is evaluated based on the current state and possible irrigation actions. This calculation takes into account not only the immediate reward (such as the direct impact of the current action on crop growth or water resource efficiency), but also the expected reward in the future, which is achieved by predicting the state of the next time unit and selecting the irrigation action that can bring the greatest expected benefit.

[0061] In addition, the algorithm specifically introduces an innovative adjustment term, the global network influence factor. This factor adjusts local decisions based on the evaluation of the global network, so that the irrigation strategy of each irrigation area not only responds to its local conditions, but also takes into account the water resource management and efficiency optimization of the entire irrigation area. Specifically, if the global network indicates that the irrigation action in a certain area may have a greater impact on the water resource balance of the entire system, this factor will adjust the action value calculation of the area to encourage it to make decisions that are more conducive to overall resource management.

[0062] As can be seen above, the core of the whole process lies in the collaborative work of two main parts: the local network ensures that each irrigation area can make the most appropriate decision based on real-time data; while the global network monitors and adjusts the actions of the entire system to ensure that all decisions jointly promote the maximum overall benefit of the entire irrigation area. In this way, the operation of the irrigation system becomes more intelligent and automated, greatly improving the efficiency of water resource use while also optimizing crop growth conditions.

[0063] Furthermore, the action value calculation formula is: in, is a neural network with parameters ; Input features are: ; The action input is: .

[0064] Specifically, the comprehensive action value is obtained as follows: in, is the weight matrix, used to linearly transform the action value of the local network; is a physical state measurement value vector and includes at least one item in the local state set; is the weight matrix used to adjust Impact in the calculation of comprehensive action value; A one-hot encoded vector representing the water irrigation pattern of the local network; is a weight matrix used to adjust the influence of water irrigation mode in the calculation of comprehensive action value; is the bias term, used in the first layer of the global network; is an activation function used to introduce nonlinearity to help the global network capture complex relationships; is the weight matrix of the second layer, and is a bias term, which is used together to convert the output of the hidden layer h into a comprehensive action value.

[0065] As can be seen above, first, the algorithm model constructs a hidden layer, the purpose of which is to effectively integrate input data from different sources. These data include the expected benefit values ​​for each irrigation area, which reflect the expected irrigation effect under specific operating conditions. In addition, physical state monitoring data such as soil moisture and air temperature are also included, which provide real-time environmental background information for the model. At the same time, the model also takes into account operating modes, which may involve specific irrigation strategies or preset behavior modes, such as automatic or manual control of the irrigation system.

[0066] In this hidden layer, the expected benefit value of each irrigation area is first weighted and synthesized with the corresponding physical state monitoring data and operation mode data. Each type of data has its corresponding weight, which is obtained through training to adjust the influence of each input on the final output. After weighting, the data is aggregated and transformed nonlinearly through an activation function, usually the ReLU function. The role of the ReLU function is to enhance the model's ability to handle nonlinear problems, so that the model can better simulate complex irrigation environments and decision-making processes.

[0067] Next, the output of the hidden layer will be further processed to generate a global irrigation strategy. This step involves another weight matrix, which transforms the output of the hidden layer into the final global irrigation decision value. This transformation process also includes a bias term, the purpose of which is to adjust the output value to adapt to different operational requirements or optimization goals.

[0068] Specifically, at each unit time, the global network is updated by integrating the action value, including: Calculate the difference between the comprehensive action value of the current time unit and the predicted comprehensive action value of the next time unit.

[0069] The corrected difference value is used to update the network parameters of the global network through the gradient descent method using the loss function.

[0070] The gradient of the loss function with respect to each parameter is calculated through the back-propagation algorithm to reduce the loss.

[0071] In view of the above specific description, at each unit of time, the system first needs to evaluate the effect of the current decision and predict the future situation. This involves calculating the comprehensive action value of the irrigation strategy implemented in the current time unit and predicting the possible results in the next time unit if the current strategy continues. By comparing these two values, a difference value can be obtained, which reflects the deviation between the immediate effect of the irrigation strategy and the future expectations.

[0072] To optimize the performance of the global network, the difference values ​​will be used as input to the loss function. The purpose of the loss function is to quantify the error between the current irrigation strategy and the ideal strategy so that the system can correct itself. By applying the gradient descent method, the system will adjust the network parameters, that is, automatically fine-tune the weights and biases that affect the quality of predictions and decisions, to minimize the value of the loss function.

[0073] Back propagation passes error information through the network layer by layer, calculating the contribution of each parameter to the final error. This calculation process ensures that each update is targeted and can effectively improve the decision-making ability of the network, thereby maximizing the action value of the entire system.

[0074] As can be seen above, through the above detailed update process, the global network is able to adapt to complex environmental changes and changing irrigation needs. For example, when encountering sudden drought events, the global network can quickly adjust and optimize the water resource allocation strategy to cope with the impact of reduced precipitation. This rapid response capability is the result of continuous learning and adaptation to the actual irrigation effect and its difference from the expected target.

[0075] Specifically, the loss function includes the following formula: in, is the loss function, which is used to characterize the deviation between the predicted comprehensive action value and the actual obtained value; For global rewards; is the discount factor; is the expected comprehensive action value under all possible actions in the next state set; is the comprehensive action value estimate for the current state set; are the network parameters of the current global network; are the parameters of the target network.

[0076] As can be seen above, the loss function is constructed based on the difference between the irrigation effects predicted by the global network in the current and future states. Specifically, the function calculates the square of the difference between the actual effect after the current irrigation decision is executed and the optimal effect that may be achieved in the future. This design enables the loss function to reflect the deviation between the immediate performance of the irrigation strategy and its long-term sustainability.

[0077] In each evaluation cycle, the system calculates an effect value based on the current irrigation strategy (including the selected irrigation method and water allocation). At the same time, the system also predicts the highest effect that can be achieved if the current optimal strategy is continued in the next cycle. By comparing these two effect values, a difference value can be obtained, which indicates the degree of deviation between the current strategy and the potential optimal strategy.

[0078] To optimize the global network’s predictive power and decision quality, the system uses gradient descent to adjust the network’s parameters. This process involves calculating the gradient of the loss function with respect to the global network’s parameters and then adjusting those parameters to reduce the loss. In this way, the network can gradually learn how to more accurately predict irrigation outcomes and allocate resources more efficiently.

[0079] The back-propagation algorithm is used here to efficiently calculate the gradient of the loss function with respect to each network parameter. This algorithm propagates error information backward through the network layers to find key parameter adjustments that improve prediction accuracy and decision quality. This not only helps the system improve the current irrigation strategy, but also optimizes the response to future irrigation needs.

[0080] Furthermore, global rewards Reflects the benefit of the entire irrigation system, defined as the sum of the rewards for each region: The rewards for each area are defined based on the following physical quantities: in, is the importance coefficient of soil moisture; is the importance coefficient of crop water requirement; is the water resource use cost coefficient; is the positive impact coefficient of precipitation on reward; is the penalty coefficient for violating physical constraints; is a penalty term based on physical constraints (such as exceeding water source capacity).

[0081] Furthermore, the comprehensive action value is obtained by weighted aggregation of the action values ​​of all local networks, and the aggregate weight of each action value in the next time unit is determined by the corresponding local state set in the current time unit.

[0082] As can be seen above, at each time unit, the local network of each irrigation area calculates an action value based on the current irrigation conditions and environmental status. These action values ​​reflect the expected benefits of taking a specific irrigation action under the current conditions. For example, when a region faces drought conditions, the local network may calculate that the action value of increasing irrigation is higher.

[0083] The action values ​​calculated by all local networks will be combined to form a comprehensive action value. The calculation of this comprehensive value is not just a simple addition, but a weighted method, where the weight of each action value is determined by the corresponding local state set. This means that if a certain area has an urgent need for irrigation or has a greater impact on the overall irrigation effect, the action value of this area will receive a higher weight when combined.

[0084] The collective weight of each action value is determined based on the set of local states in the current time unit. This includes information such as soil moisture, crop growth stage, weather forecast, etc. This approach ensures that the weight allocation can reflect the actual needs and priorities of each area in real time, making the global irrigation strategy more in line with the actual situation.

[0085] The present invention provides an automated water resource optimization scheduling management method based on an irrigation demonstration area. Traditional systems usually rely on fixed water sources and irrigation patterns and lack the ability to cope with sudden climate changes. The present invention greatly increases the flexibility of the system by constructing a scheduling set containing multiple irrigation methods and establishing these scheduling sets for each irrigation area. Each area can choose the most suitable irrigation method according to actual conditions, so as to effectively cope with different environmental changes. In the case where global resources may not be sufficient to meet all needs, the selection optimization algorithm can evaluate and determine the best irrigation strategy for each area in real time. Not only based on the immediate needs of each irrigation area, but also taking into account the water resource status of the entire irrigation area, it ensures that water resources are reasonably allocated and utilized where they are most needed.

[0086] By acquiring and analyzing the local state set of each irrigation area in real time, the system can accurately control the distribution of water and reduce water waste. This refined management not only improves the overall efficiency of water resources, but also ensures the effective use of water resources under extreme climatic conditions.

[0087] Allows the irrigation system to dynamically adjust irrigation strategies based on local conditions such as actual soil moisture, crop growth conditions, and weather forecasts. This adaptive capability enables the irrigation system to better respond to climate change and reduce crop losses caused by improper irrigation.

[0088] Irrigation is a key link in agricultural production. This invention helps agricultural producers increase crop yields and reduce environmental risks by ensuring that each decision is based on the latest data and algorithm optimization. It not only increases agricultural output, but also helps achieve sustainable agricultural development.

[0089] In the description of the present invention, it should be understood that the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0090] The above embodiments are only descriptions of the preferred modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. The automated water resources optimization scheduling management method based on the irrigation demonstration area is characterized by: The steps include: Obtain N irrigation areas in the current focused irrigation area, and respectively establish a scheduling set including multiple water-using irrigation methods for each of the irrigation areas; Constructing a selection optimization algorithm for selecting the water irrigation mode for all the irrigation areas one by one at each moment; Acquire a local state set of each of the irrigation areas in real time, and input all the local state sets into the selection optimization algorithm to select water irrigation methods for all the irrigation areas at the same time; The method includes: constructing an action space set for each irrigation area through the type of irrigation water allocation and water-use irrigation method, and constructing an independent local network for each irrigation area; obtaining the state set of the irrigation area according to the local state set, and calculating the action value of all actions of the irrigation area in the action space set through the local network; retaining the action with the highest action value, and adjusting water to the current irrigation area according to the irrigation water allocation and water-use irrigation method contained in the retained action.

2. The automated water resource optimization scheduling management method according to claim 1 is characterized in that: The water irrigation methods of the scheduling set include channel water source calling, path optimal calling, water source optimal calling and water volume optimal calling; The channel water source calling step includes: obtaining all channel paths in the current irrigation area, opening the currently obtained channel paths, until the channel water levels of all current channel paths are consistent; The step of optimally adjusting the path includes: obtaining the optimal path between the water pump and all grid fields in the current irrigation area, and irrigating the grid fields according to the optimal path; The step of optimally calling water sources includes: obtaining a callable irrigation area adjacent to the current irrigation area, and irrigating the current irrigation area through the callable irrigation area; The step of optimally calling water volume includes: obtaining the required water volume and the available water volume of all the irrigation areas, and irrigating all the irrigation areas through an external water source according to the difference between the required water volume and the available water volume.

3. The automated water resource optimization scheduling management method according to claim 1 is characterized in that: The step of retaining the action with the highest action value comprises: Among them, the is the action with the highest action value; is the action space set, and the Allocate water for irrigation, For water irrigation; is the number of the irrigation area, and is a positive integer from 1 to N; is the state set; is the parameter used to estimate the expected reward of each action; is the value of the action.

4. The automated water resource optimization scheduling management method according to claim 1 is characterized in that: The selection optimization algorithm includes the local networks of all the irrigation areas and the global network constructed by mixing the local networks of all the irrigation areas, and has the following situations: Scenario 1: Obtain a comprehensive action value according to the action values ​​of all local networks at each unit time, and update the global network according to the comprehensive action value; In scenario 2, the update of the local network in the next time unit takes into account the influence of the global network on the current local network.

5. The automated water resource optimization scheduling management method according to claim 4 is characterized in that: At each unit time, the local network of each irrigation area is updated by the action value, specifically including: Among them, the is the learning rate; is the discount factor; The benefit value of the current action is adopted for the irrigation area; is the state set for the next unit time; is the action for the next unit time; is the maximum action value among all actions in the next state set; is the adjustment coefficient, which is used to control the influence of the global network on the local network update; is the gradient of the global network relative to the local network, which is used to indicate the influence of the global network on the local network; is the comprehensive action value of the global network; is the current action value.

6. The automated water resource optimization scheduling management method according to claim 5 is characterized in that: The comprehensive action value is obtained as follows: Among them, the is a weight matrix used to linearly transform the action value of the local network; is a physical state measurement value vector, and includes at least one item in the local state set; is the weight matrix used to adjust The impact on the calculation of the comprehensive action value; A one-hot encoding vector representing the water irrigation method of the local network; is a weight matrix used to adjust the influence of water irrigation mode in the calculation of comprehensive action value; is a bias term used in the first layer of the global network; is an activation function used to introduce nonlinearity to help the global network capture complex relationships; is the weight matrix of the second layer, and is a bias term, which is used together to convert the output of the hidden layer h into a comprehensive action value.

7. The automated water resource optimization scheduling management method according to claim 4 is characterized in that: At each unit time, the global network is updated by the comprehensive action value, specifically including: Calculate the difference between the comprehensive action value of the current time unit and the predicted comprehensive action value of the next time unit; Using the corrected difference value to update the network parameters of the global network through a gradient descent method using a loss function; The gradient of the loss function with respect to each parameter is calculated through the back-propagation algorithm to reduce the loss.

8. The automated water resource optimization scheduling management method according to claim 7, characterized in that: The loss function includes the following formula: Among them, the is a loss function used to characterize the deviation between the predicted comprehensive action value and the actual obtained value; is a global reward; is the discount factor; is the expected comprehensive action value under all possible actions in the next state set; is the comprehensive action value estimate of the current state set; is the network parameter of the current global network; are the parameters of the target network.

9. The automated water resource optimization scheduling management method according to claim 4, characterized in that: The comprehensive action value is obtained by weighted aggregation of the action values ​​of all local networks, and the aggregate weight of each action value in the next time unit is determined by the corresponding local state set in the current time unit.

Images