Soil breeding and planting evaluation method based on Internet of Things

Abstract

The invention discloses a soil breeding planting evaluation method based on the Internet of Things, and relates to the technical field of soil breeding, and the method comprises the steps: determining a core index of the evaluation method based on a soil breeding target, and carrying out the real-time collection of the index through an Internet of Things sensor technology; according to the core indexes and the collected data, soil health condition evaluation is carried out; crop variety selection is carried out according to the soil health score; generating a soil improvement suggestion according to a crop variety selection result; the planting environment is optimized according to the soil improvement suggestion; dynamically monitoring the soil quality according to the optimization scheme; performing crop growth simulation according to the dynamic soil quality data; and performing crop yield prediction and optimization according to a growth simulation result. According to the method, a dynamic feature selection mechanism, a time decay-based reward function and a multi-dimensional feedback mechanism are introduced, so that the model is more efficient, and the real-time performance, the accuracy and the adaptive capacity to changes of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of breeding and seedling cultivation, and in particular to a soil breeding and planting evaluation method based on the Internet of Things. Background Art

[0002] With the continuous development of agricultural production, traditional planting management methods have gradually failed to meet the needs of modern agriculture for precision and intelligence. Accurate decision-making in soil health assessment and crop breeding requires real-time monitoring and intelligent analysis of multi-dimensional soil data. In recent years, with the application of Internet of Things technology, agricultural management has begun to develop in the direction of automation and digitalization, especially soil health monitoring and crop planting assessment methods, which have received widespread attention. These systems use soil sensors, climate monitoring, and crop growth status sensing equipment to collect key data such as soil moisture, temperature, pH value, nutrient content, etc. in real time, and use advanced algorithms for analysis to help agricultural producers make accurate decisions based on real-time soil conditions and crop needs.

[0003] In the existing technology, reinforcement learning algorithms have shortcomings: soil health management issues involve long-term, multi-dimensional decision-making processes, and existing reinforcement learning algorithms usually ignore the complex interactions between soil and crop status. Reinforcement learning algorithms usually require a large amount of interaction data and time to optimize strategies, but due to the complexity of the agricultural environment and the scarcity of data, existing reinforcement learning methods are difficult to fully learn effective soil management strategies in a short period of time. In addition, the reinforcement learning model is not accurate enough in the state space design of the soil system, ignoring the complex interrelationships between different soil characteristics, resulting in a slow convergence of the strategy, which ultimately affects the real-time and accuracy of the soil management strategy. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a soil breeding and planting evaluation method based on the Internet of Things to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides a soil breeding and planting evaluation method based on the Internet of Things, comprising the following steps: S1. Based on the soil breeding goals, determine the core indicators of the evaluation method, use the Internet of Things sensor technology to collect the indicators in real time, and obtain the collected data; S2. Evaluate soil health status based on core indicators and collected data to obtain a soil health score; S3, selecting crop varieties according to the soil health score, and obtaining crop variety selection results; S4. Generate soil improvement suggestions based on crop variety selection results; S5. Optimize the planting environment according to the soil improvement suggestions and obtain the optimization plan; S6. Perform dynamic soil quality monitoring according to the optimization plan to obtain dynamic soil quality data; S7, performing crop growth simulation according to the dynamic soil quality data to obtain a growth simulation result; S8. Predict and optimize crop yield based on growth simulation results.

[0006] To further optimize this technical solution, the core indicators in S1 include: Soil moisture, soil temperature, soil pH value, organic matter content, soil element content.

[0007] To further optimize this technical solution, the soil health status assessment in S2 includes: The random forest algorithm is used, core indicators and collected data are used, the impact of a single indicator is considered, and the interaction effects of multiple indicators are analyzed to monitor and evaluate the health of the soil in real time and obtain a soil health score.

[0008] To further optimize this technical solution, the crop variety selection in S3 includes: By using a decision tree algorithm, combined with soil health scores and crop growth requirements, and based on feedback from historical and real-time data, the crop variety selection strategy is gradually optimized to ensure the maximum match between the selected varieties and soil conditions, and to select the most suitable crop varieties.

[0009] To further optimize the technical solution, the soil improvement suggestion generated in S4 includes: Based on the selected crop varieties, the soil requirements and growth characteristics of the crop varieties are analyzed, and a linear regression model is used to combine the soil requirements, growth characteristics and existing soil conditions of the crop varieties to generate targeted soil improvement suggestions.

[0010] To further optimize the technical solution, the optimized planting environment in S5 includes: After improving the soil through soil improvement suggestions, genetic algorithms are used to simulate the natural selection process and find the optimal planting environment configuration based on the improved soil conditions and various planting management factors, including optimizing the irrigation system, fertilization plan, and crop planting density.

[0011] To further optimize the technical solution, the soil quality dynamic monitoring in S6 includes: Data fusion technology is used to integrate and analyze the collected soil data, monitor changes in soil quality in real time, and obtain accurate dynamic soil quality data monitoring information.

[0012] To further optimize the technical solution, the crop growth simulation in S7 includes: Based on dynamic soil quality data, a growth simulation model is used to comprehensively consider factors including soil quality, climate conditions, and crop growth factors to demonstrate the long-term impact of different management measures on crop growth, predict future growth trends of crops, and simulate the growth process of crops.

[0013] To further optimize the technical solution, the crop yield prediction and optimization in S8 includes: Based on the growth simulation results, a reinforcement learning algorithm is used to respond to environmental changes in real time, simulate the effects of different strategies, optimize crops through real-time feedback, adjust planting strategies, and predict crop yields.

[0014] To further optimize the technical solution, the reinforcement learning algorithm includes: This algorithm continuously optimizes planting strategies based on soil health, crop needs, and historical feedback, and dynamically adjusts through actual crop growth feedback to achieve the purpose of optimizing planting strategies and predicting crop yields; Model construction: ; in: :Soil status The value function at time t represents the expected reward from this state and following a certain strategy. : Discount factor, which indicates the decay of the reward and is set according to the optimization requirements; : The state of the soil after the action is performed, including the core indicators and time factors of the soil, including: ; in, is soil moisture, is the soil temperature, is the soil pH value, is the organic matter content, , , is the nitrogen, phosphorus and potassium content in the soil, is the current time; : The action performed at time t+k is set according to the soil improvement measures implemented; :In action Get Status The immediate reward obtained after , r is the reward function, including: ; in, For crops in soil state The output under Improvement of soil health, including pH adjustment and nutrient supplementation; is the weight coefficient, which is set according to the actual situation.

[0015] In a second aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of a soil breeding and planting evaluation method based on the Internet of Things as described in the first aspect of the present invention are implemented.

[0016] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of a soil breeding and planting evaluation method based on the Internet of Things as described in the first aspect of the present invention are implemented.

[0017] Compared with the prior art, the present invention provides a soil breeding and planting evaluation method based on the Internet of Things, which has the following beneficial effects: This IoT-based soil breeding and planting evaluation method introduces a time-decayed reward function and a multi-dimensional feedback mechanism, which enables the reinforcement learning model to learn and converge efficiently in a short period of time, while considering multi-dimensional factors and feedback in the long-term decision-making process of soil and crop management. Through the precisely designed reward function, the reinforcement learning algorithm can quickly adapt to changes in soil conditions and provide real-time and effective soil improvement and crop variety selection strategies under different environmental conditions, effectively improving the real-time and accuracy of the system and optimizing the long-term benefits of soil health management decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic diagram of a process of a soil breeding and planting evaluation method based on the Internet of Things proposed by the present invention; Figure 2 This is a flow chart of a random forest algorithm for a soil breeding and planting evaluation method based on the Internet of Things proposed in the present invention. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Embodiment 1: Reference Figure 1-2 , which is the first embodiment of the present invention, and provides a soil breeding and planting evaluation method based on the Internet of Things, comprising the following steps: S1. Based on the soil breeding goals, determine the core indicators of the evaluation method, use the Internet of Things sensor technology to collect the indicators in real time, and obtain the collected data.

[0024] In this embodiment, the core indicators include: Soil moisture, soil temperature, soil pH value, organic matter content, soil element content.

[0025] IoT sensor technology (such as soil sensors) is used to collect these key indicators in real time. The collected data needs to be able to reflect the biophysical state of the soil, thereby providing a reliable basis for soil breeding decisions.

[0026] S2. Assess soil health status based on core indicators and collected data to obtain a soil health score.

[0027] In this embodiment, soil health status assessment includes: The random forest algorithm is used, core indicators and collected data are used, the impact of a single indicator is considered, and the interaction effects of multiple indicators are analyzed to monitor and evaluate the health of the soil in real time and obtain a soil health score. The output of the soil health score not only helps to reflect the immediate health status of the soil, but also predicts the future change trend of the soil and provides a scientific management basis.

[0028] The key to the random forest algorithm model is to build a dynamic response random forest model by dynamically updating and integrating multi-dimensional soil data, which can evaluate the time-varying characteristics of soil health.

[0029] The model is constructed as follows: ; in: : Output of soil health assessment, representing soil health score; : A function based on the random forest algorithm, indicating that the model performs calculations by integrating multiple decision trees; : They are soil moisture, soil temperature, soil pH value, organic matter content, and nitrogen, phosphorus and potassium content in the soil; : Time factor, which represents the time dimension of the soil health score. The model will dynamically adjust according to the passage of time and update the weight of each feature; The time-weighted update mechanism is introduced. The soil health status depends not only on the current status (such as humidity, pH value, etc.), but also needs to take into account the changes in soil characteristics over the past period of time. The dynamic impact of this change is adjusted by a time factor, the formula is: ; in: : soil health score at time t; : Input dimension features, n=7, including ; : Time factor, which represents the weighted effect of time on each feature and is obtained through historical data training; : The weight of the i-th feature, indicating the contribution of each feature to soil health assessment; : The input value of the i-th feature; Model usage: Data collection: IoT sensors collect core soil indicators in real time as input to the model.

[0030] Weight adjustment: As soil conditions change, the model adjusts the weights of each feature based on the time factor. For example, in seasons with low soil moisture, the effect of moisture may be aggravated, while in seasons with high soil temperature, the effect of temperature may be more prominent.

[0031] Model training: Based on historical data, the random forest algorithm optimizes feature weights and time weighting factors through multiple decision tree splits. Through training, the model can adaptively update these parameters under different seasons, soil types, and environmental conditions.

[0032] Evaluation output: Ultimately, the algorithm calculates a soil health score by dynamically weighting various soil characteristics to provide a basis for agricultural decision-making. This evaluation result not only reflects the current health status of the soil, but also predicts possible future changes and helps agricultural producers optimize soil management strategies.

[0033] S3. Select crop varieties according to the soil health score to obtain crop variety selection results.

[0034] By using a decision tree algorithm, combined with soil health scores and crop growth requirements, and based on feedback from historical and real-time data, the crop variety selection strategy is gradually optimized to ensure the maximum match between the selected varieties and soil conditions, and to select the most suitable crop varieties.

[0035] Decision Tree Algorithm: Input features and target variables: The input features of the decision tree model include soil characteristics such as soil health score, soil moisture, temperature, pH value, organic matter content, and crop growth requirements (such as water requirements, temperature adaptation range, light requirements, etc.). The output target variable is the choice of crop variety.

[0036] The process of building a decision tree: A decision tree is composed of a series of "nodes" and "branches". In the process of building a decision tree, the algorithm divides the data set into different subsets by selecting the best split point (i.e. the best feature and split value) from the input features. Each split will make a decision based on a feature (such as the pH value or temperature of the soil) to ensure that each type of branch contains similar input data as much as possible. Each node will make a judgment based on a certain feature, for example: whether the soil moisture is greater than a certain threshold. Each branch assigns data to different child nodes based on the judgment criteria. This process continues until a preset termination condition is reached (such as the maximum depth of the tree, or the number of samples in each leaf node is lower than a certain threshold).

[0037] "Leaf nodes" of a decision tree: At each terminal node (leaf node) of the tree, the algorithm will give a crop variety choice. Each leaf node represents a specific crop variety, and the division of these leaf nodes is based on the characteristic conditions in the input data and the growth performance of the crop variety under these conditions.

[0038] Optimization process: During the training of the decision tree, the algorithm will learn from historical data how to make the best decision based on soil health scores and crop needs. Each time the tree is built, the algorithm will evaluate the effect of each segmentation point to ensure that the data subsets after segmentation are as "pure" as possible, that is, the sample categories (crop varieties) in the subsets are as similar as possible. Through step-by-step training, the decision tree can gradually optimize the decision rules in multiple segmentation processes and improve prediction accuracy.

[0039] Real-time feedback and updates: Through the real-time feedback mechanism, the decision tree can dynamically adjust the crop variety selection according to the current soil conditions and crop growth data. For example, when the growth status of the crop or the soil health score changes, the decision tree will use the latest data for retraining or adjustment, thereby continuously optimizing the crop variety selection and ensuring that the selected variety is the best match with the soil health status.

[0040] S4. Generate soil improvement recommendations based on crop variety selection results.

[0041] In this embodiment, generating soil improvement suggestions includes: Based on the selected crop varieties, the soil requirements and growth characteristics of the crop varieties are analyzed, and a linear regression model is used to combine the soil requirements, growth characteristics and existing soil conditions of the crop varieties to generate targeted soil improvement suggestions.

[0042] Soil requirements of crops: Different crops have different requirements for soil, such as pH value, nitrogen, phosphorus, potassium content, organic matter content, etc. These requirements are usually determined based on the growth characteristics of the crop and the optimal growth environment.

[0043] Soil status: Soil data collected by sensors.

[0044] Output target: Through the linear regression model, the output generated is specific soil improvement recommendations, such as adding specific fertilizers or adjusting soil pH.

[0045] The core formula of the linear regression model is as follows: ; in: : Output soil improvement recommendations, indicating the intensity or specific quantitative value of soil improvement measures (e.g., type and amount of fertilizer, dosage for pH adjustment, etc.); : Input characteristics, including soil pH, nitrogen, phosphorus, potassium content, organic matter content, etc.; : intercept term, representing the baseline value when there is no soil improvement measure; : regression coefficient, indicating the influence of each characteristic on soil improvement suggestions; : Error term, used to represent the unpredictable part of the model.

[0046] S5. Optimize the planting environment according to soil improvement suggestions and obtain an optimized plan.

[0047] In this embodiment, optimizing the planting environment includes: After improving the soil through soil improvement suggestions, genetic algorithms are used to simulate the natural selection process and find the optimal planting environment configuration based on the improved soil conditions and various planting management factors, including optimizing the irrigation system, fertilization plan, and crop planting density. The advantage of genetic algorithms lies in their global search capabilities, which can find the best combination strategy among multiple environmental factors.

[0048] The optimization plan specifically includes: Clarification of soil improvement plan: Through the soil improvement suggestions obtained in step S4, the system obtains key health issues in the soil, such as soil acidity (pH value), nutrient deficiency, loose structure, etc. The soil improvement plan is clarified according to the specific improvement measures in the soil improvement suggestions. For example, the soil organic matter content can be improved by adding organic fertilizers, or the pH value can be adjusted by applying lime to further improve the air permeability and water retention of the soil.

[0049] Multi-factor soil optimization: Soil optimization and improvement programs need to consider multiple factors and balance them. For example, the relationship between soil moisture content and aeration is very important. If the soil is too moist, it may cause root hypoxia and affect crop growth; if the soil is too dry, the root system will not be supplied with enough water, which will also affect crop growth. Therefore, the improvement program should not only improve the chemical properties of the soil (such as fertilizer application and pH adjustment), but also consider the physical properties (such as improving soil structure and increasing water permeability). On this basis, the role of genetic algorithms is to optimize the entire planting environment according to the characteristics of the soil, including integrated water and fertilizer management programs.

[0050] Dynamic adjustment of the management system after soil optimization: As soil conditions improve, planting management measures also need to be adjusted dynamically. By matching the optimized soil improvement plan with the crop growth cycle, the genetic algorithm can adjust the irrigation and fertilization strategy in real time to optimize the crop planting density. For example, after soil optimization, if the soil's water retention capacity is enhanced, the frequency and amount of irrigation may be reduced, and the amount of fertilizer can be dynamically adjusted according to the soil's nutrient status, thereby improving resource utilization efficiency and ensuring that crops grow under optimal conditions.

[0051] Applications of Genetic Algorithms: Genetic algorithm is a global search method that simulates the natural selection process. Its advantage is that it can find the optimal solution to complex multi-dimensional problems. After soil improvement, genetic algorithm is used to optimize the planting environment. The specific process is as follows: Initialize the population: Generate multiple initial planting configurations based on the characteristics of the improved soil. Each configuration includes settings for factors such as irrigation, fertilization, and planting density.

[0052] Fitness evaluation: Each plan is evaluated based on criteria such as crop yield, soil moisture retention, fertilizer utilization, etc. The fitness function can calculate the pros and cons of each plan based on these factors.

[0053] Selection, crossover and mutation: The genetic algorithm selects the better solution according to the fitness function for crossover and mutation to generate a new planting configuration. By repeating this process, the algorithm gradually approaches the optimal solution.

[0054] Optimization goal: The ultimate goal of genetic algorithm optimization is to find a balance point to ensure the best combination between soil health, crop growth efficiency and resource consumption. For example, by optimizing the irrigation system to reduce water waste while ensuring that crops get enough water; by adjusting the fertilization plan, reduce excessive fertilizer use, avoid excessive accumulation of soil nutrients, and increase crop yields.

[0055] Comprehensive management program: Soil optimization and improvement programs are not limited to immediate soil improvement, but also involve long-term sustainable management strategies. Based on the optimization results of genetic algorithms, soil health management strategies can be adjusted dynamically. For example, some soil improvement measures are more effective in specific seasons or environmental conditions. Genetic algorithms can combine weather forecasts, seasonal changes and other factors to adjust irrigation and fertilization plans for different growth stages and climatic conditions to achieve the best crop growth results.

[0056] S6. Perform dynamic soil quality monitoring according to the optimization plan to obtain dynamic soil quality data.

[0057] In this embodiment, soil quality dynamic monitoring includes: Data fusion technology is used to integrate and analyze the collected soil data, monitor changes in soil quality in real time, and obtain accurate dynamic soil quality data monitoring information.

[0058] Data fusion technology combines sensor data, historical data, and model predictions, and uses algorithms such as Kalman filtering to smooth data and reduce noise, thereby providing accurate information for dynamic monitoring of soil quality. This step can help the system detect the changing trend of soil quality in real time and adjust soil improvement measures based on data feedback.

[0059] S7. Perform crop growth simulation based on the dynamic soil quality data to obtain growth simulation results.

[0060] In this embodiment, the crop growth simulation includes: Based on dynamic soil quality data, a growth simulation model is used to comprehensively consider factors including soil quality, climate conditions, and crop growth factors to demonstrate the long-term impact of different management measures on crop growth, predict future growth trends of crops, and simulate the growth process of crops. The growth simulation model can demonstrate the long-term impact of different management measures on crop growth, thereby supporting long-term agricultural planning.

[0061] Growth simulation models usually consist of the following main parts: State variables: represent the state of the system at a certain moment, such as soil pH, humidity, nutrient concentration, etc., and crop growth status (such as leaf area index, crop yield, etc.).

[0062] Flow variable: describes the rate of change of state variables, such as water loss rate, nutrient absorption rate, crop growth rate, etc.

[0063] Controlled variables: controllable factors in the system, such as irrigation amount, fertilizer amount, planting density, etc.

[0064] The formula of the growth simulation model is: 1. Soil health status model Soil health status Affected by factors such as fertilization, irrigation, temperature, etc. Assume that the rate of change of soil health is proportional to these factors, and consider the natural decline process of soil health: ; in: : soil health status at time t; : are the amount of fertilizer and irrigation respectively; : Influence coefficient, set according to actual conditions; : The natural decline coefficient of soil, which indicates the rate of decline of soil health when no management measures are taken, is set according to actual conditions; 2. Crop Growth Model The growth status of crops Affected by soil health, climate conditions (such as temperature, humidity) and fertilization and irrigation. The growth rate of crops is usually closely related to the relationship between soil health and environmental factors, which is expressed by the following equation: ; in: : Crop growth status, including leaf area index and crop yield; : The efficiency coefficient of crop growth, reflecting the role of soil health in promoting crop growth; : Maximum growth potential of crops; 3. Resource Management Model Resource management involves controlling the amount of irrigation and fertilizer applied, usually in relation to crop demand and soil health. Set up a resource allocation function to adjust the amount of irrigation and fertilizer applied: ; ; in: : The current available amount of resources; : A function that adjusts resource usage based on soil health and crop growth status; 4. Feedback mechanism and delay effect There is a strong feedback mechanism between soil health and crop growth. The effects of fertilization and irrigation are reflected in crop growth, but it usually takes some time to manifest. This time delay effect is described by introducing a hysteresis function: ; ; in: : The time delay from fertilization and irrigation to crop growth, in days; 5. Integrated system model Finally, all modules are combined to form a complete growth simulation model that describes the dynamic interaction between soil health, crop growth, and resource management: ; Model application and analysis Real-time adjustments: Based on the system model, decision makers can optimize soil health and crop growth by adjusting resource allocation plans such as fertilizer and irrigation.

[0065] Long-term prediction: By simulating the dynamic processes of soil and crop growth, the system can predict long-term soil health changes and help develop sustainable agricultural management strategies.

[0066] Sensitivity analysis: Through sensitivity analysis, we can identify which factors (such as fertilizer application, irrigation amount, soil health level, etc.) have the greatest impact on crop yield and soil health, thereby providing a basis for resource allocation.

[0067] S8. Predict and optimize crop yield based on growth simulation results.

[0068] In this embodiment, crop yield prediction and optimization includes: Based on the growth simulation results, reinforcement learning algorithms are used to respond to environmental changes in real time, simulate the effects of different strategies, optimize crops, adjust planting strategies, and predict crop yields through real-time feedback. The use of reinforcement learning algorithms can not only predict crop yields, but also respond to environmental changes in real time and automatically adjust planting strategies to obtain the best results. Through this feedback mechanism, agricultural producers can obtain continuously optimized planting plans, increase crop yields, and achieve sustainable agricultural management.

[0069] Furthermore, the reinforcement learning algorithm includes: This algorithm continuously optimizes planting strategies based on soil health, crop needs, and historical feedback, and dynamically adjusts through actual crop growth feedback to achieve the purpose of optimizing planting strategies and predicting crop yields; The core of this algorithm model includes the following three aspects: Multi-dimensional input: Traditional reinforcement learning algorithms usually use simple state and action spaces, but this algorithm model introduces multiple soil health characteristics (such as moisture, pH, temperature, nutrients, etc.) as state inputs, and incorporates time factors into the state space to consider the time-varying characteristics of soil characteristics.

[0070] Adaptive strategy: Use the policy function in reinforcement learning to dynamically adjust soil management plans, such as updating fertilization, irrigation and other management decisions in real time based on soil changes, crop needs and environmental feedback.

[0071] Long-term optimization: This model does not just optimize the decision at a certain moment, but considers long-term benefits, that is, training the agent through time-delayed reward feedback.

[0072] Model construction: ; in: :Soil status The value function at time t represents the expected reward from this state and following a certain strategy. : Discount factor, which indicates the decay of the reward and is set according to the optimization requirements; : The state of the soil after the action is performed, including the core indicators and time factors of the soil, including: ; in, is soil moisture, is the soil temperature, is the soil pH value, is the organic matter content, , , is the content of nitrogen, phosphorus and potassium in the soil, is the current time; : The action performed at time t+k is set according to the soil improvement measures implemented; :In action Get Status The immediate reward obtained after , r is the reward function, including: ; in, For crops in soil state The output under Improvement of soil health, including pH adjustment and nutrient supplementation; is the weight coefficient, which is set according to the actual situation.

[0073] Decision-making process of reinforcement learning: State space: In this model, the state space is multidimensional, including various soil health indicators (moisture, temperature, pH, organic matter content, nutrient concentration, etc.), as well as time factors (current season, climate conditions, etc.). Therefore, the state space not only includes the current condition of the soil, but also takes into account the time-varying characteristics of the soil; Action space: The action space includes all possible soil management strategies, such as fertilizer application, irrigation, soil improvement measures (such as lime, phosphate fertilizer, etc.), which affect soil health and crop growth; Reward function: The reward function design takes into account the improvement of crop yield and soil health. The reward depends not only on the current soil health, but also on the long-term effect of the improvement measures. Therefore, the reward function is dynamic and will be adjusted over time and as the soil state changes; Strategy updating: Reinforcement learning updates strategies through interaction with the environment. By evaluating the rewards for taking actions in different states, the agent continuously adjusts decision rules and optimizes soil management strategies.

[0074] Use of the model: Initialization: Initialize the state and reward function through historical data and soil model; Training: During the training phase, the system performs actions based on the current state and obtains immediate rewards. Through repeated interactions, the reinforcement learning algorithm is used to optimize the strategy to maximize the long-term rewards. Strategy update and execution: After strategy learning is completed, the system dynamically adjusts soil improvement measures (such as fertilization, irrigation, etc.) according to the current health status of the soil and crop needs, and predicts the maximum crop yield and soil health status.

[0075] Embodiment 2: This embodiment also provides a computer device, which is suitable for a soil breeding and planting evaluation method based on the Internet of Things, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement a soil breeding and planting evaluation method based on the Internet of Things proposed in the above embodiment.

[0076] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, a soil breeding and planting evaluation method based on the Internet of Things as proposed in the above embodiment is implemented.

[0077] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0078] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0079] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0080] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk case (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0081] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A soil breeding and planting evaluation method based on the Internet of Things, characterized in that: The following steps are involved: S1. Based on the soil breeding goals, determine the core indicators of the evaluation method, use the Internet of Things sensor technology to collect the indicators in real time, and obtain the collected data; S2. Evaluate soil health status based on core indicators and collected data to obtain a soil health score; S3, selecting crop varieties according to the soil health score, and obtaining crop variety selection results; S4. Generate soil improvement suggestions based on crop variety selection results; S5. Optimize the planting environment according to the soil improvement suggestions and obtain the optimization plan; S6. Perform dynamic soil quality monitoring according to the optimization plan to obtain dynamic soil quality data; S7, performing crop growth simulation according to the dynamic soil quality data to obtain a growth simulation result; S8. Predict and optimize crop yield based on growth simulation results.

2. The method for evaluating soil breeding and planting based on the Internet of Things according to claim 1, characterized in that: The core indicators in S1 include: Soil moisture, soil temperature, soil pH value, organic matter content, soil element content.

3. The method for soil breeding and planting evaluation based on the Internet of Things according to claim 1, characterized in that: The soil health assessment in S2 includes: The random forest algorithm is used, core indicators and collected data are used, the impact of a single indicator is considered, and the interaction effects of multiple indicators are analyzed to monitor and evaluate the health of the soil in real time and obtain a soil health score.

4. The method for evaluating soil breeding and planting based on the Internet of Things according to claim 1, characterized in that: The crop variety selection in S3 includes: By using a decision tree algorithm, combined with soil health scores and crop growth requirements, and based on feedback from historical and real-time data, the crop variety selection strategy is gradually optimized to ensure the maximum match between the selected varieties and soil conditions, and to select the most suitable crop varieties.

5. The method for soil breeding and planting evaluation based on the Internet of Things according to claim 1, characterized in that: The soil improvement suggestions generated in S4 include: Based on the selected crop varieties, the soil requirements and growth characteristics of the crop varieties are analyzed, and a linear regression model is used to combine the soil requirements, growth characteristics and existing soil conditions of the crop varieties to generate targeted soil improvement suggestions.

6. The method for evaluating soil breeding and planting based on the Internet of Things according to claim 1, characterized in that: The optimized planting environment in S5 includes: After improving the soil through soil improvement suggestions, genetic algorithms are used to simulate the natural selection process and find the optimal planting environment configuration based on the improved soil conditions and various planting management factors, including optimizing the irrigation system, fertilization plan, and crop planting density.

7. The method for evaluating soil breeding and planting based on the Internet of Things according to claim 1, characterized in that: The dynamic monitoring of soil quality in S6 includes: Data fusion technology is used to integrate and analyze the collected soil data, monitor changes in soil quality in real time, and obtain accurate dynamic soil quality data monitoring information.

8. The method for soil breeding and planting evaluation based on the Internet of Things according to claim 1, characterized in that: The crop growth simulation in S7 includes: Based on dynamic soil quality data, a growth simulation model is used to comprehensively consider factors including soil quality, climate conditions, and crop growth factors to demonstrate the long-term impact of different management measures on crop growth, predict future growth trends of crops, and simulate the growth process of crops.

9. The method for soil breeding and planting evaluation based on the Internet of Things according to claim 1, characterized in that: The crop yield prediction and optimization in S8 includes: Based on the growth simulation results, a reinforcement learning algorithm is used to respond to environmental changes in real time, simulate the effects of different strategies, optimize crops through real-time feedback, adjust planting strategies, and predict crop yields.

10. The method for evaluating soil breeding and planting based on the Internet of Things according to claim 9, characterized in that: The reinforcement learning algorithm includes: This algorithm continuously optimizes planting strategies based on soil health, crop needs, and historical feedback, and dynamically adjusts through actual crop growth feedback to achieve the purpose of optimizing planting strategies and predicting crop yields; Model construction: ； in: :Soil status The value function at time t represents the expected reward from this state and following a certain strategy. : Discount factor, which indicates the decay of the reward and is set according to the optimization requirements; : The state of the soil after the action is performed, including the core indicators and time factors of the soil, including: ; in, is soil moisture, is the soil temperature, is the soil pH value, is the organic matter content, , , is the nitrogen, phosphorus and potassium content in the soil, is the current time; : The action performed at time t+k is set according to the soil improvement measures implemented; :In action Get Status The immediate reward obtained after , r is the reward function, including: ； in, For crops in soil state The output under Improvement of soil health, including pH adjustment and nutrient supplementation; is the weight coefficient, which is set according to the actual situation.

Images