Structure control optimization method for water distribution system of water energy accumulator

Abstract

The invention belongs to the field of agricultural intellectualization, and mainly relates to a structure control optimization method for a water distribution system of a water energy accumulator. According to the method, various sensors are installed in an agricultural scene, soil humidity, meteorological data and other information are collected in real time, and in combination with planting area video image data collected by an unmanned aerial vehicle, the growth condition of crops is monitored and analyzed in an omnibearing mode; the growth condition of crops is analyzed through an image recognition algorithm, and the water demand and growth state of the crops are accurately evaluated in combination with sensor data; using a deep learning algorithm to learn a rule of a crop growth cycle and an irrigation demand, and dynamically adjusting an irrigation behavior; the Internet of Things node judges irrigation requirements according to sensor data and a set threshold value and regulates and controls irrigation work in real time; according to the method, through multi-source data modeling and intelligent regulation and control, precise irrigation of different areas and different crops is achieved, efficient utilization of water resources is ensured, and excessive or insufficient irrigation is avoided.

Description

Technical Field

[0001] This invention belongs to the field of agricultural intelligence and mainly relates to a structural control optimization method for a water storage device water distribution system. Background Technology

[0002] In existing agricultural irrigation, water resource utilization efficiency is low. Many farmlands still rely on traditional irrigation methods, such as flood irrigation and timed irrigation. These methods cannot be adjusted in real time according to the actual needs of crops and environmental changes, which easily leads to water waste and insufficient irrigation precision.

[0003] In Chinese patent application CN202311093720.2, "Agricultural Irrigation System Based on Soil Moisture," this invention discloses an agricultural irrigation system based on soil moisture, comprising: a soil moisture module for real-time detection and calculation of the rate of change of soil moisture error values; an irrigation frequency converter for controlling the water pump by changing the output; a preprocessing module for constructing fuzzy subsets; a rule module for establishing a fuzzy rule base; and a controller module for outputting control signals. The working principle is as follows: acquiring soil moisture data, preprocessing to generate fuzzy subsets, using fuzzy inference to generate control quantities, and driving the frequency converter to implement irrigation. This system achieves accurate and adaptable irrigation through intelligent sensing and closed-loop control. Utilizing multi-source data modeling, knowledge and data integration, and dynamic regulation, it effectively solves the problems of traditional experience-based control, significantly improves yield and quality, and reduces labor costs. It is an effective solution for intelligent and refined agriculture.

[0004] The aforementioned patents have provided a relatively complete agricultural irrigation system. However, in agricultural scenarios, due to limitations in the layout of the water distribution system and the equipment itself, it is impossible to water an area evenly, resulting in imprecise irrigation, wasted water resources, and reduced irrigation effectiveness. Moreover, existing technologies typically focus on controlling soil moisture while neglecting the actual growth status of crops. Theoretically, crops should reach the expected growth stage after a specific irrigation cycle, but the actual situation is often more complex, causing crops to fail to achieve the expected growth effect through fixed irrigation methods. Furthermore, crops at different growth stages have different water requirements, which further affects crop yield and quality. Therefore, there is an urgent need for an agricultural irrigation method that controls irrigation conditions, senses overall soil moisture, and considers the actual growth status of crops. Summary of the Invention

[0005] This invention provides a structural control optimization method for a water storage device water distribution system, aiming to solve the problems of uneven irrigation and insufficient precision in existing agricultural irrigation systems; as well as the problem of focusing on soil moisture control while ignoring the differences in crop growth cycle and water requirements, resulting in crop growth not meeting expectations and affecting yield and quality.

[0006] To solve the above problems, the present invention employs the following technology:

[0007] A structural control optimization method for a water distribution system of a water storage tank:

[0008] In agricultural settings, an irrigation system is built; data acquisition modules are installed in planting areas, sensors are installed, and relevant data is collected; the data is encrypted and compressed, and then uploaded to IoT nodes; drones are used to collect video and image data of the planting area; and the video and image data is uploaded to the central processing unit.

[0009] IoT nodes receive sensor data, combine it with corresponding data threshold settings, regulate irrigation behavior, and simultaneously upload the data to the central processing unit.

[0010] The central processing unit preprocesses the obtained data; analyzes sensor data to determine the current irrigation status of the crop; uses image recognition algorithms to analyze video image data to assess the actual growth of the crop; records planting area data to construct an environmental data change map; and uses machine learning algorithms to perform deep learning on various types of data to learn irrigation patterns and adjust irrigation operations accordingly.

[0011] As a preferred implementation, the construction of an irrigation system in an agricultural setting specifically includes:

[0012] The type and layout of the irrigation system are determined based on the size of the agricultural region, crop types, climate conditions, and soil type.

[0013] Select appropriate equipment according to the design plan and make reasonable arrangements in the planting area so that the irrigation range covers all planting areas;

[0014] Record the deployment location, connection relationship and related parameters of the device nodes to the local database.

[0015] As a preferred implementation, the collection of relevant data specifically includes:

[0016] Several environmental data sensors were installed in the planting area to collect soil data and related data, and meteorological sensors were set up to collect meteorological data of the planting area.

[0017] The collected data is encrypted using an encryption algorithm and compressed using a compression algorithm before being transmitted to the corresponding IoT node.

[0018] As a preferred embodiment, the acquisition of video image data of the planting area specifically includes:

[0019] In non-rainy weather, use drones equipped with high-definition cameras to record video image data of the planting area along a specific route;

[0020] Adjust the frequency of video data collection by the drone according to the stability of the crop growth cycle;

[0021] The acquired video image data is uploaded to the central control system via a wireless network.

[0022] As a preferred embodiment, the regulation of irrigation behavior specifically includes:

[0023] The central control system sets the thresholds corresponding to the data and sends them to the corresponding IoT nodes;

[0024] The IoT nodes make judgments based on the data collected by the sensors and the thresholds, triggering irrigation work and synchronously uploading the sensor data to the central control system.

[0025] In a preferred embodiment, the preprocessing of the obtained data by the central processing unit specifically includes:

[0026] The data is cleaned by using Gaussian filtering to remove noise, aligning the timestamps of data from different sources, and then standardizing and normalizing them.

[0027] For video image data, image correction is performed to remove image distortion; Gaussian filters are used for image denoising; image enhancement is performed to highlight details; and image size is modified to meet algorithm standards.

[0028] As a preferred implementation, determining the current crop irrigation status specifically includes:

[0029] Based on sensor location information, a two-dimensional model diagram is constructed. Combined with sensor data, a cubic spline interpolation algorithm is used to fill the two-dimensional model diagram. The rationality of the irrigation system is analyzed by comprehensively analyzing the two-dimensional model diagram, and the corresponding optimization plan for the irrigation system is formulated and sent to the corresponding processing unit.

[0030] The current growth cycle is queried based on the crop growth time, the corresponding water use threshold is adjusted, and the data is sent to the IoT nodes.

[0031] By combining meteorological data for assessment, instructions are sent to IoT nodes in rainy weather to prevent irrigation from being triggered.

[0032] As a preferred implementation, the assessment of the actual growth of the crop specifically includes:

[0033] The deep learning model Convolutional Neural Network (CNN) is used to process image and video image data, and crop areas are identified by markers set up in the planting areas.

[0034] The model was trained using images of various crop growth conditions, and the model was used to evaluate crop growth from video image data.

[0035] As a preferred implementation, the construction of the environmental data change graph specifically includes:

[0036] Store historical environmental data, categorize it according to data type, and construct change graphs for various environmental data.

[0037] Multiple data change graphs are aligned by timestamp, and the actual growth of crops is linked to the data change graphs according to the time difference of video image data collection.

[0038] As a preferred implementation method, adjusting the irrigation process specifically includes:

[0039] Based on the planting requirements of crop theory, a basic water threshold is set, and a multimodal time-series adaptive network model is used to input various data change maps and the associated growth conditions into the model to learn the change characteristics between water and various data during the planting process and construct a change pattern with regional characteristics.

[0040] Based on the regional water variation patterns, combined with crop growth and existing data, the water threshold is adjusted to guide normal crop growth.

[0041] The beneficial effects of this invention are:

[0042] 1. Precision irrigation control: By combining soil moisture, meteorological data, crop growth status and environmental changes, the system can achieve uniform distribution of water within the region, avoid water waste, improve irrigation accuracy, and ensure that each region receives the appropriate amount of water according to its actual needs.

[0043] 2. Adaptive regulation mechanism: Based on multi-source data modeling and dynamic regulation technology, this method can adjust the irrigation strategy in real time according to the water requirements of different crops at each growth stage, optimize the irrigation strategy, avoid crop growth due to insufficient water or excessive irrigation, and improve crop yield and quality.

[0044] 3. Reduced manual intervention and costs: Through automated control and intelligent sensing, the need for manual intervention is reduced, labor costs are lowered, and water resource utilization efficiency is improved through precise irrigation regulation, providing an effective solution for intelligent agricultural management. Attached Figure Description

[0045] Figure 1 This is a flowchart of a structural control optimization method for a water distribution system of a water storage device according to the present invention;

[0046] Figure 2 This diagram shows a comparison of the effects of the structural control optimization method for a water storage tank distribution system in this invention with traditional methods. Detailed Implementation

[0047] To make the technical means, creative features, and achieved objectives and effects of this invention easier to understand, the invention is further described below with reference to specific embodiments. However, the following embodiments are merely preferred embodiments of this invention and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments described herein without creative effort are all within the protection scope of this invention. Unless otherwise specified, the experimental methods in the following embodiments are conventional methods, and the materials and reagents used in the following embodiments are commercially available unless otherwise specified.

[0048] Example 1 Figure 1 A flowchart of a structural control optimization method for a water distribution system of a water storage tank is shown. This embodiment provides a structural control optimization method for a water distribution system of a water storage tank, specifically including the following steps:

[0049] Step 1: Construct an irrigation system in the agricultural setting; build a data acquisition module in the agricultural setting, install various sensors to collect relevant data from the agricultural setting, and aggregate the obtained data to an IoT node; encrypt and compress the data during data transmission; periodically deploy drones equipped with image acquisition modules to collect video image data from the agricultural setting, and upload the agricultural video image data to the central control system via a wireless network.

[0050] Specifically, constructing a water distribution system in an agricultural setting refers to: determining the type and layout of the irrigation system, including drip irrigation or sprinkler irrigation, based on the scale of the agricultural area, crop type, climate conditions, and soil type; laying out sprinklers in the agricultural setting and connecting them to water storage tanks; selecting appropriate equipment according to the design plan, including water pumps, pipes, valves, sprinklers, or drippers; during installation, laying main pipes, branch pipes, and capillary pipes reasonably to ensure that water can evenly cover the entire irrigation area; and after completing the layout of the water distribution system, recording the deployment locations of the equipment nodes in the water distribution system, the connection relationships between the equipment, and the relevant parameters of the equipment in a local database.

[0051] Collecting various data in agricultural scenarios involves: deploying multiple sensors within the agricultural environment, including soil moisture and temperature sensors to monitor changes in soil moisture and temperature. These sensors should cover different areas of the entire agricultural scenario to accurately capture changes in different soil types and depths; deploying light sensors to measure light intensity; configuring meteorological sensors to collect environmental data such as air temperature, air humidity, wind speed, and precipitation to help comprehensively understand the impact of climate on the agricultural environment; rationally distributing sensors according to the size and layout of the agricultural scenario to ensure basic monitoring of environmental changes throughout the area. Particular attention must be paid to the uniformity of coverage in the deployment of soil moisture and temperature sensors, ensuring a reasonable and even distribution of sensors in each area. The precise location of the sensors within the agricultural scenario should be recorded, and the sensors should be associated with the corresponding crop type. Finally, IoT nodes should be set up according to the sensor layout structure, with multiple IoT nodes configured to achieve distributed processing and rapid response.

[0052] After agricultural environmental data sensors collect data, all sensor data is aggregated to IoT nodes for processing. Upon receiving the sensor data, the IoT nodes synchronously upload the data to the central control unit. To improve transmission efficiency and ensure data security, the data is encrypted before transmission using the AES encryption algorithm to ensure confidentiality and integrity during transmission and prevent external attacks or data leaks. In addition, to reduce network bandwidth pressure and improve data transmission speed, the data is compressed using the ZIP algorithm to reduce data size and ensure efficient and stable transmission even in environments with poor network conditions. The receiving port uses corresponding data processing methods to decompress and decrypt the data, restoring it to its original form.

[0053] Collecting video and image data of agricultural scenarios refers to the following: Based on the scale and monitoring needs of the agricultural scenario, drones are used regularly in non-rainy weather conditions such as sunny and cloudy days to collect video and image data of the agricultural scenario. Drone parameters, including flight speed, flight route, and flight altitude, are set. Corresponding markers, such as directional signs and buildings, are placed along the drone's video and image collection route. The drone is equipped with high-definition cameras or multispectral sensors to ensure the acquisition of high-definition video or high-resolution image data of the agricultural scenario, covering areas containing crops. The specific drone data collection frequency is set according to specific circumstances and flexibly according to the crop growth cycle. For example, during critical growth stages, the frequency of video and image collection is increased to once a day, while during stable growth stages, the frequency is reduced to once every two weeks. Drones are not used for data collection during rainy weather, but data collection should be conducted as soon as possible after the rain stops. The drone automatically collects video or images in real time according to the specific flight route and parameters. The collected data is stored in real time in the drone's storage device. After collection ends, the image data is transmitted to the central control system via wireless network (such as 4G / 5G, Wi-Fi, etc.).

[0054] Step 2: The distributed IoT nodes make real-time judgments based on sensor data and adjust the operation of the irrigation system according to the real-time data; the sensor data is simultaneously uploaded to the central control system.

[0055] Specifically, adjusting the operation of the irrigation system based on real-time data means that: IoT nodes receive data thresholds from the central control system. After receiving data from the data sensors, the IoT nodes first determine the threshold. If the current data is lower than the IoT node's current threshold, specifically indicating that the current soil moisture is too low and cannot meet the basic growth needs of crops in the current area, an instruction is immediately triggered to the corresponding irrigation control equipment to trigger a watering action in the area with low moisture. The specific data threshold is set by the central control system based on the actual situation. The obtained data is initially judged by the IoT sensors and simultaneously uploaded to the central control system.

[0056] Step 3: The central control system preprocesses the obtained data; analyzes the irrigation situation in the current agricultural environment by integrating sensor data; analyzes the video image data using image recognition algorithms, and confirms the irrigation situation by integrating the data analysis results.

[0057] Specifically, the preprocessing of the data received by the central control system involves the following steps: After receiving data uploaded by IoT nodes, the central control system first performs preliminary data cleaning, using a Gaussian filter algorithm to remove noise and ensure data accuracy and reliability; it then synchronizes data collected by different sensors to ensure data from different sources are aligned with the same timestamp for easier subsequent analysis; based on crop growth cycles and water requirements, different types of data are standardized and normalized for comprehensive analysis; the acquired video image data is preprocessed, including image correction after receiving video image data from drones or other devices, removing image distortion caused by lens distortion or flight angle to ensure the image accurately reflects the agricultural scene; a Gaussian filter is used to denoise the image, removing noise caused by ambient light, wind, sand, raindrops, etc., ensuring image clarity and quality; image enhancement is performed to improve contrast and brightness, making details more prominent for easier subsequent analysis, especially for images taken under low-light conditions; and the image size is adjusted according to analysis requirements to ensure rapid transmission and adaptability to algorithm processing needs.

[0058] Comprehensive sensor data analysis of irrigation conditions in the current agricultural environment refers to: constructing a two-dimensional model of soil moisture, temperature, and other data, combined with sensor placement, labeling sensor locations and data, and using cubic spline interpolation to fill the data in the two-dimensional model to obtain complete soil moisture and temperature data for the farm environment. The moisture map is used to determine if there are any unreasonable aspects to the water distribution system, such as if the soil moisture in one area is significantly lower or higher than other areas after a watering event; this unreasonable area is marked. A layout map is constructed based on the location records of all devices, and a reasonable coverage area is determined according to device parameters. The coverage area is then modified to match the actual soil moisture conditions. Attempts are made to improve the coverage area using layout modifications; otherwise, new irrigation equipment is added, and the data is sent to the corresponding processing unit. The system adjusts the quantity or layout of irrigation equipment based on actual conditions; it integrates environmental data and crop types from different regions, and queries the corresponding crop growth demand database to determine irrigation requirements for each crop based on its growth time. Based on recorded crop growth time, it preliminarily determines the current crop growth cycle and sends the corresponding growth cycle and related sensor data to the corresponding IoT nodes, replacing the original IoT node's judgment threshold. Additional evaluation is conducted using meteorological data, including current temperature and rainfall. If the current weather is rainy and the rainfall level basically meets or even exceeds crop growth requirements, an additional instruction to prohibit irrigation is added to the corresponding IoT node, thereby adjusting the crop's irrigation situation.

[0059] The analysis of video image data using image recognition algorithms refers to the following steps: After the central control system receives video image data from drones or other devices, it first extracts the video frames one by one using image processing technology, ensuring that each frame can be input into the image recognition algorithm as independent data. For each frame, a deep learning model, the Convolutional Neural Network (CNN), is used to extract features from the image. Corresponding agricultural scene markers are set in the database, and these markers are then identified and extracted from the image to mark the actual location of the current image area. The model is then trained on crop appearance morphology using a large number of crop growth images from agricultural scenes, including images of healthy, unhealthy, and different growth stages. This allows the model to distinguish the crop's growth stage, water shortage, and health status from the images. The model then identifies the morphological features of the crop in the current image frame, including leaf color, shape, size, and overall growth status. Based on these features, it determines whether the crop has problems such as pests, diseases, water shortage, or abnormal growth. When pests or diseases are found, corresponding warnings are triggered to the relevant units. The analysis results are then stored in a local database.

[0060] Step 4: Record the obtained data and construct an agricultural environment data change map; use machine learning algorithms to perform deep learning on the changes in agricultural environment data, summarize the changing patterns in the current agricultural scenario, and adjust irrigation work accordingly.

[0061] Constructing agricultural environmental data change maps involves long-term storage of data collected from various sensors to accumulate environmental change data, classifying it by data type, and gradually building various environmental change maps. This data includes soil moisture and temperature, ambient temperature and humidity, and meteorological data such as wind speed, precipitation, and air temperature. First, each data point is marked with a timestamp. Then, change maps are drawn for each data type to show environmental change trends at different points in time. For example, soil moisture sensor data can generate a soil moisture change map, showing changes in soil moisture over different time periods. Furthermore, combining image recognition algorithms to identify crop growth, and based on the time differences in video image data uploaded by drones, data points in all corresponding time windows are labeled with crop growth status, including the model's assessment of current crop growth, health status, and growth cycle.

[0062] Adjusting irrigation practices involves: first, setting a basic water threshold based on the crop's growth cycle. For example, tomatoes have specific soil moisture requirements during their growth. Research indicates that the ideal soil moisture for tomatoes is typically between 60% and 70%. When soil moisture falls below 60%, watering is triggered. In the absence of supporting data, this threshold is sent to the corresponding IoT node. A multimodal temporal adaptive network is used to perform deep learning on multiple data change maps and corresponding labels, establishing connections and mappings between various data types, including relationships such as light-soil temperature-soil moisture, air humidity-soil moisture, and rainfall-soil moisture. The corresponding data changes are calculated, such as the rate of change of soil temperature and moisture under sunny, high-sunlight conditions. The resulting data change maps and calculated data are input into the multimodal temporal adaptive network. A temporal convolutional network is used to extract local temporal patterns from the environmental data. A multi-head attention mechanism is used to analyze the interrelationships between different environmental variables, performing deep learning on the relationships between various data changes. Crop type and growth stage information are converted into feature vectors, establishing a correspondence between the growth cycle and environmental response, and different crops at different growth stages are associated with each other. The water demand characteristics are encoded into the input network, converting geographic location information into a numerical representation and establishing a correlation pattern between geographic location and environmental response. Then, a multimodal temporal adaptive network automatically weighs the importance of different data sources based on the input data, dynamically adjusting the contribution weights of each modal feature according to the specific task, and fusing multiple features. This allows analysis of soil water retention and heat preservation capabilities in different sensor regions, as well as crop growth in different regions, from data changes. It also analyzes the minute changes in crop water demand under different weather conditions, forming a correlation between water and crop growth that integrates soil characteristics in different regions. For example, on cloudy days, with high humidity and low temperature, the soil in this area has inherently lower water loss characteristics (possibly due to insufficient local sunlight or thicker soil layers), further reducing water loss. Furthermore, crop growth in this area may differ slightly, with some over-watering. Finally, combining the analyzed regional soil characteristics and crop growth, a dynamic long-term irrigation strategy is formulated. For example, data analysis can reveal that prolonged rain has led to persistently wet soil in a certain area, resulting in poor crop growth. The system can then automatically generate corresponding control strategies: optimizing the irrigation trigger threshold of IoT nodes in that area by appropriately lowering the lower limit of soil moisture for irrigation, prompting the system to irrigate only when the soil is drier. Through this precise control, excess soil moisture can be eliminated, guiding crop growth back to normal.

[0063] like Figure 2The diagram shows a comparison of the structural control optimization method of the water storage device water distribution system in this invention with traditional methods. The black bars in the diagram represent the effect of this invention, and the gray bars represent the effect of traditional methods. After adopting the method described in this invention, the crop water resource utilization rate is greatly improved compared with traditional methods, and it is less likely to experience abnormal water conditions such as too much or too little water.

[0064] Example 2: A hydroelectric power plant water distribution system was used on a farm for precise irrigation control. The farm was divided into multiple irrigation zones, each planted with different types of crops such as tomatoes and corn. To optimize irrigation efficiency and ensure healthy crop growth, corresponding water distribution systems were designed based on the water requirements of different crops.

[0065] First, the farm's irrigation system employs drip irrigation and micro-sprinkler irrigation technologies. For the tomato growing area, considering the crop's precise water requirements, a drip irrigation system suitable for root irrigation is installed to ensure that water can penetrate evenly into the root zone and avoid wasting water resources. The corn growing area uses a micro-sprinkler irrigation system, as it is more drought-resistant and has a deeper root system, allowing the sprinkler system to cover a wider planting area more efficiently and reduce water evaporation loss.

[0066] When implementing the water distribution system, different types of sensors were installed in various areas of the farm, including soil moisture sensors, soil temperature sensors, weather sensors, and light sensors. Soil moisture sensors primarily monitor soil moisture levels to ensure sufficient water for crop roots; soil temperature sensors monitor soil temperature changes to prevent excessive evaporation; weather sensors measure environmental conditions such as air temperature, humidity, and precipitation; and light sensors monitor light intensity in different crop areas to assess the photosynthetic efficiency of the crops.

[0067] All data collected by sensors is transmitted to the central control system via IoT nodes, where the system performs comprehensive analysis of this real-time data. For example, if the soil moisture in the corn growing area falls below a set threshold, irrigation will be initiated based on the difference between the soil moisture in that area and the expected moisture level to replenish the required water. Simultaneously, if weather sensors detect rainfall information in the weather forecast, the system will automatically adjust the irrigation amount based on real-time data to avoid unnecessary water waste. For tomato growing areas, the system similarly uses precise control based on changes in soil moisture and temperature to ensure the soil remains at an appropriate level of moisture and avoids over-irrigation. Furthermore, crop growth in each area is regularly monitored using drones. These drones are equipped with image acquisition modules to capture high-resolution images of the farm, which are then uploaded to the central control system.

[0068] The central control system preprocesses the collected data, including data cleaning and time synchronization, to ensure the accuracy of all data. It analyzes crop health and water requirements using image recognition algorithms; simultaneously, through deep learning and data analysis, combined with crop type and growth cycle, it can automatically learn the water requirement patterns of each crop and make dynamic adjustments. In farms with multiple crops, the system not only considers the needs of individual crops but also comprehensively analyzes soil, climate, and crop growth conditions in different areas to formulate optimal irrigation strategies.

[0069] Ultimately, through long-term data accumulation and model optimization, the system can automatically generate irrigation plans for each region, ensuring that different crops receive optimal water support at different growth stages, improving crop yield and quality, while maximizing water resource utilization. In the tomato area, the system adjusts the irrigation strategy according to the water requirements of the fruit growth stage; in the corn area, the system adjusts the irrigation intensity according to weather changes and soil moisture variations to ensure stable crop growth. Through this precise irrigation control, the farm's water resource utilization efficiency has been significantly improved.

[0070] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for structural control optimization of a water distribution system for a water storage tank, characterized in that: In agricultural settings, an irrigation system is built; data acquisition modules are installed in planting areas, sensors are installed, and relevant data is collected; the data is encrypted and compressed, and then uploaded to IoT nodes; drones are used to collect video and image data of the planting area; and the video and image data is uploaded to the central processing unit. IoT nodes receive sensor data, combine it with corresponding data threshold settings, regulate irrigation behavior, and simultaneously upload the data to the central processing unit. The central processing unit preprocesses the received data; The system analyzes sensor data to determine the current irrigation status of crops; uses image recognition algorithms to analyze video image data to assess the actual growth of crops; records data from the planting area to construct an environmental data change map; and employs machine learning algorithms to perform deep learning on various types of data to learn irrigation patterns and adjust irrigation operations accordingly.

2. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The construction of irrigation systems in agricultural scenarios specifically includes: The type and layout of the irrigation system are determined based on the size of the agricultural region, crop types, climate conditions, and soil type. Select appropriate equipment according to the design plan and make reasonable arrangements in the planting area so that the irrigation range covers all planting areas; Record the deployment location, connection relationship and related parameters of the device nodes to the local database.

3. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The collected relevant data specifically includes: Several environmental data sensors were installed in the planting area to collect soil data and related data, and meteorological sensors were set up to collect meteorological data of the planting area. The collected data is encrypted using an encryption algorithm and compressed using a compression algorithm before being transmitted to the corresponding IoT node.

4. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The video image data of the planting area collected specifically includes: In non-rainy weather, use drones equipped with high-definition cameras to record video image data of the planting area along a specific route; Adjust the frequency of video data collection by the drone according to the stability of the crop growth cycle; The acquired video image data is uploaded to the central control system via a wireless network.

5. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The specific measures for regulating irrigation behavior include: The central control system sets the thresholds corresponding to the data and sends them to the corresponding IoT nodes; The IoT nodes make judgments based on the data collected by the sensors and the thresholds, triggering irrigation work and synchronously uploading the sensor data to the central control system.

6. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The preprocessing of the obtained data by the central processing unit specifically includes: The data is cleaned by using Gaussian filtering to remove noise, aligning the timestamps of data from different sources, and then standardizing and normalizing them. For video image data, image correction is performed to remove image distortion; Gaussian filters are used for image denoising; image enhancement is performed to highlight details; and image size is modified to meet algorithm standards.

7. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The determination of the current crop irrigation status specifically includes: Based on sensor location information, a two-dimensional model diagram is constructed. Combined with sensor data, a cubic spline interpolation algorithm is used to fill the two-dimensional model diagram. The rationality of the irrigation system is analyzed by comprehensively analyzing the two-dimensional model diagram, and the corresponding optimization plan for the irrigation system is formulated and sent to the corresponding processing unit. The current growth cycle is queried based on the crop growth time, the corresponding water use threshold is adjusted, and the data is sent to the IoT nodes. By combining meteorological data for assessment, instructions are sent to IoT nodes in rainy weather to prevent irrigation from being triggered.

8. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The assessment of the actual growth of the crop specifically includes: The deep learning model Convolutional Neural Network (CNN) is used to process image and video image data, and crop areas are identified by markers set up in the planting areas. The model was trained using images of various crop growth conditions, and the model was used to evaluate crop growth from video image data.

9. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The constructed environment data change graph specifically includes: Store historical environmental data, categorize it according to data type, and construct change graphs for various environmental data. Multiple data change graphs are aligned by timestamp, and the actual growth of crops is linked to the data change graphs according to the time difference of video image data collection.

10. The structural control optimization method for a water distribution system of a water storage tank according to claim 1, characterized in that: The adjustments to irrigation work specifically include: Based on the planting requirements of crop theory, a basic water threshold is set, and a multimodal time-series adaptive network model is used to input various data change maps and the associated growth conditions into the model to learn the change characteristics between water and various data during the planting process and construct a change pattern with regional characteristics. Based on the regional water variation patterns, combined with crop growth and existing data, the water threshold is adjusted to guide normal crop growth.

Images