Intelligent adjusting device for strawberry garden irrigation

Abstract

The invention discloses an intelligent adjusting device for strawberry garden irrigation, and relates to the technical field of agricultural irrigation, the device comprises a data sensing module, a main control module, a man-machine interaction module and an execution module; the data sensing module is used for collecting soil moisture content, planting environment data and strawberry growth stage data in real time; the main control module is used for receiving the sensing data, generating a dynamic irrigation decision according to an irrigation analysis model, and sending an irrigation decision control signal to the execution module; the man-machine interaction module is used for acquiring strawberry planting parameter information input by a user, querying soil and environment monitoring data and irrigation state information, and performing remote interaction; and the execution module is used for controlling the electromagnetic valve array to independently irrigate according to the strawberry planting ridges according to the irrigation decision control signal. Water resource waste of a traditional irrigation mode is avoided, accurate irrigation is conducted according to the soil humidity and the strawberry growth stage, the water resource utilization rate is increased, and automation of the irrigation process is achieved.

Description

TECHNICAL FIELD

[0001] The application relates to the field of agricultural irrigation technology, in particular to an intelligent adjusting device for strawberry field irrigation. BACKGROUND

[0002] As a high-value economic crop, the growth of strawberries is strictly dependent on irrigation conditions. Because the root system of strawberries is shallow and densely distributed, the water requirement varies greatly at different growth stages. Traditional drip irrigation is prone to local water and fertilizer excess or deficiency due to terrain undulations. Manual control relies on experience and cannot respond to sudden changes in soil moisture in real time, for example, the root rot rate increases due to failure to turn off the irrigation system after heavy rain. It is also difficult to accurately meet the water requirements of strawberries at different growth stages, which may lead to water waste, yield reduction and unstable quality. In addition, the water use efficiency of flood irrigation is low, and cold water irrigation may cause fluctuations in soil temperature, affecting root activity.

[0003] In the prior art, although some automatic irrigation equipment based on soil moisture sensors has appeared, most of them use fixed threshold triggering mechanism and do not consider the water requirement characteristics of strawberries at different growth stages, nor do they realize dynamic optimization of irrigation amount. At the same time, the data collection dimension is relatively single, usually only monitoring soil moisture parameters, lacking comprehensive perception of environmental factors and crop growth status, resulting in insufficient scientificity of irrigation decision. In addition, the energy supply mode of the existing system is single, which is easily limited by power supply conditions in remote planting areas, and lacks special water and fertilizer mixing and precise drip irrigation technical solutions for strawberry planting. SUMMARY

[0004] Therefore, the application provides an intelligent adjusting device for strawberry field irrigation, which solves the technical problems of local water and fertilizer excess or deficiency caused by traditional drip irrigation, inability to respond to sudden changes in soil moisture in real time, low water use efficiency, easy influence on root activity, lack of dynamic optimization of irrigation amount, single energy supply mode, insufficient irrigation precision and low efficiency.

[0005] To achieve the above-mentioned purpose, the application provides the following technical solutions: The application mainly forms a closed-loop control system by integrating water temperature regulation, water and fertilizer mixing and AI decision system, and realizes efficient irrigation management of strawberry planting. An intelligent irrigation device and system capable of comprehensively perceiving the environment and crop status, dynamically generating irrigation decisions and realizing zoned precise irrigation are provided. The device comprises a data perception module, a main control module, a human-computer interaction module and an execution module.

[0006] The data perception module is used for real-time acquisition of soil moisture, planting environment data and strawberry growth stage data, and sends the acquired perception data to the main control module, and comprises a distributed soil monitoring unit, an environment perception unit, an image analysis unit and a fault monitoring unit.

[0007] The input end of the main control module is connected with the output end of the data sensing module, for receiving the sensing data, generating a dynamic irrigation decision according to an irrigation analysis model, and sending an irrigation decision control signal to the execution module.

[0008] The human-computer interaction module is used for acquiring the strawberry planting parameter information input by a user, and querying soil and environment monitoring data and irrigation state information, and remote interaction.

[0009] The execution module is used for controlling the electromagnetic valve array and the irrigation equipment to independently irrigate according to the strawberry planting ridge partition according to the irrigation decision control signal.

[0010] Further, the data sensing module comprises a distributed soil monitoring unit and an environment sensing unit.

[0011] The environment sensing unit comprises an illumination intensity sensor, an air temperature and humidity sensor, a rain gauge and a micro weather station.

[0012] Further, the data sensing module further comprises an image analysis unit and a fault monitoring unit. The image analysis unit is used to preprocess strawberry plant growth images acquired by a high-definition camera. Preprocessing includes normalizing the image to 512×512 pixels, cropping based on the region of interest, Gaussian denoising (σ=1.5), histogram equalization, and contrast stretching with a stretching coefficient of 1.2. It also extracts the 128-dimensional color histogram and 3-dimensional color moment features of the strawberry plant in HSV space, calculates the gray-level co-occurrence matrix using an 8×8 window to extract 16-dimensional texture features such as energy and entropy, obtains the shape features of the strawberry plant through Canny edge detection and contour extraction, and quantizes them using 10 shape descriptors such as roundness and perimeter-area ratio. Principal component analysis is then used to select principal component features with a cumulative contribution rate ≥95% from the extracted features, reducing the feature dimension to 20-30 dimensions. An image classification model is built based on the YOLOv8 network to identify the current growth stage information of the strawberry plant. Transfer learning is used to initialize the weights. The training set contains 5000 images each of the strawberry seedling stage, flowering stage, early fruiting stage, peak fruiting stage, and ripening stage.

[0013] The fault monitoring unit includes a flow detector, a pipeline pressure sensor, and a current transformer, which work together to monitor faults in the irrigation system. The flow detector is an electromagnetic flow meter with a measurement range of 0-50 m³ / h and an accuracy of ±0.5%. The flow detector is used to detect drip irrigation flow information, the pipeline pressure sensor is used to detect pipeline pressure information, and the current transformer is used to detect water pump current information.

[0014] Furthermore, the main control module includes an evapotranspiration prediction unit and an optimization analysis unit.

[0015] The evapotranspiration prediction unit is used to predict the reference evapotranspiration (ET0) sequence for the next 24 hours based on collected historical meteorological sensing data using an LSTM network-based prediction model. The prediction model, built on an LSTM network, consists of an input layer (5 neurons), LSTM layer 1 (64 neurons, ReLU activation function), Dropout layer 1 (dropoutrate=0.2), LSTM layer 2 (32 neurons, tanh activation function), Dropout layer 2 (dropoutrate=0.2), a fully connected layer (16 neurons), and an output layer (24 neurons, linear activation). The model input is a time series of historical meteorological data from the past 3 days, containing 5 feature parameters: daily minimum temperature T_min (°C), daily maximum temperature T_max (°C), daily average relative humidity RH_avg (%), wind speed at 2m altitude u2 (m / s), and solar radiation Rs (MJ / m²・d). The output is the reference evapotranspiration (ET0) sequence for the next 24 hours. During model training, each feature was individually Z-score standardized (μ=0, σ=1), the Adam optimizer was used (learning rate 0.001), the loss function was mean squared error (MSE), and the training iterations were 500.

[0016] The input data for the LSTM network prediction model includes historical meteorological data (T_min, T_max, RH_avg, u2, Rs) time series from the past 3 days. The output layer outputs 24 nodes, representing the reference evapotranspiration ET0 data for the next 24 hours. During model training, each feature in the training data is first standardized using Z-score.

[0017] The optimization analysis unit is connected to the evapotranspiration prediction unit. The optimization analysis unit calculates the irrigation timing and amount based on strawberry growth stage information, predicted ET0, and soil moisture data. It also optimizes the irrigation amount for each stage based on the Jensen model, the expression of which is: in, This represents the actual evapotranspiration during the i-th growth stage. This represents the maximum evapotranspiration during the i-th growth stage. Let Y be the water shortage sensitivity index for the i-th growth stage, Y be the actual yield, and Ym be the maximum yield. The optimization objective is to ensure... Under the premise of ≥0.95, minimize the amount of irrigation water used in each stage. The process of calculating whether to irrigate and the amount of irrigation is as follows: According to the formula: Calculate the soil dynamic moisture threshold. The average ET0 for the next 24 hours is predicted by a prediction model based on an LSTM network. The average ET0 over the past 7 days The ET0 response coefficient is set to 0.05 for strawberries. This represents the basic threshold for field water holding capacity at the corresponding growth stage of strawberries; According to the formula: Calculate the risk of water shortage in the next 24 hours, among which, , This represents the water consumption for the next 24 hours. Let t be the ET0 value predicted by the LSTM network-based prediction model at hour t, and Kc(t) be the dynamic crop coefficient at hour t. For effective rainfall, For the effective water storage capacity of the strawberry root zone, =0.6 × root distribution depth × The strawberry root system is distributed at a depth of 30cm. , This represents the soil moisture stress coefficient. Current soil moisture; Irrigation is triggered when soil moisture falls below the dynamic soil moisture threshold or the water shortage risk value exceeds 0.8; otherwise, irrigation is not performed. When irrigation is triggered, the following formula applies: Calculate the irrigation amount, and the final irrigation amount satisfies: ,in, The maximum irrigation volume per cycle is given by Kc, which is the dynamic crop coefficient. Evapotranspiration per hour (Etm) is calculated as Etm = Kc * [Etm's value]. The total evapotranspiration for the next 24 hours is calculated by summing the hourly evapotranspiration. The amount of evapotranspiration that needs to be replenished is the total evapotranspiration for the next 24 hours. ETm,k is the total evapotranspiration for the next 24 hours, which is also the maximum water demand. A is the irrigated area. For system irrigation efficiency, drip irrigation is taken as 0.9, Rt is the effective rainfall, ΔSt is the soil water replenishment, and ΔSt=( - )*SWC, Target soil moisture, The current soil moisture is given, and SWC represents the total water holding capacity of the soil. The calculation determines the amount of irrigation required for the next day during the current growth stage, and the calculation results are converted into irrigation commands and sent to the execution module.

[0018] Furthermore, the main control module also includes a fault self-diagnosis unit and a dual-mode power supply unit; The fault self-diagnosis unit is used to determine whether there is a blockage fault based on the real-time detected flow information and pipeline pressure information using the vector machine classification method. It is also used to perform Fourier transform on the detected water pump current information and extract feature information. The dual-mode power supply unit includes a solar power supply module, an automatic switching circuit, and a mains power supply module. The solar power supply module includes a solar photovoltaic panel, an MPPT solar controller, and a 24V / 200Ah lithium iron phosphate battery. The mains power supply module includes an AC-DC converter, a voltage regulator, and a DC-DC converter. The automatic switching circuit is electrically connected to the solar power supply module and the mains power supply module to switch the power supply mode.

[0019] Furthermore, the execution module includes a water and fertilizer mixing unit, a water temperature regulation unit, and a pressure-compensating drip irrigation pipe; The water and fertilizer mixing unit is used to mix water and fertilizer using a combination of a Venturi fertilizer applicator and a screw pump. The screw pump is connected to the main control module via a frequency converter. The water temperature control unit includes an electric heating element, a circulating air heat exchanger, and a three-way proportional valve. The electric heating element is installed in the warm water tank, and the three-way proportional valve is used to mix cold water and warm water. The pressure-compensating drip irrigation pipe uses drip irrigation tape with an embedded silicone pressure compensation sheet.

[0020] Furthermore, it also includes a drainage module, which includes a drainage pump connected to the main control module. The drainage pump is used to drain water according to control signals.

[0021] As can be seen from the above technical solution, the advantages of the present invention are: 1. This invention utilizes multi-source data fusion from distributed soil monitoring, environmental sensing, and image analysis to comprehensively analyze the actual growth needs and environmental changes of strawberries. Based on LSTM network-based evapotranspiration prediction and Jensen model optimization, combined with the water requirements of strawberries at different growth stages, it dynamically adjusts irrigation timing and amount, ensuring balanced and timely water supply during strawberry growth. This achieves precision irrigation while improving water resource utilization efficiency, thereby increasing strawberry yield and quality. It avoids water waste associated with traditional irrigation methods. Furthermore, it automates the irrigation process, reducing manual operation and lowering labor intensity and costs. Through a human-computer interaction module, users can easily adjust irrigation strategies and view and analyze historical data, enabling intelligent management of strawberry garden irrigation.

[0022] 2. Achieve precise control and remote management of irrigation water volume, water temperature, and water-fertilizer ratio. Through real-time monitoring of flow rate, pressure, and current, and fault self-diagnosis, system anomalies can be detected and handled in a timely manner, reducing maintenance costs. Furthermore, it adopts automatic switching between solar power and mains power to meet the energy needs of remote planting areas. Attached Figure Description

[0023] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an undue limitation of this application.

[0024] Figure 1 This is a schematic diagram of the composition of the intelligent regulating device for strawberry garden irrigation according to this application.

[0025] Figure 2 This is a schematic diagram of the framework of the dynamic irrigation decision-making process in the strawberry garden in this embodiment.

[0026] Figure 3 This is a schematic diagram of the prediction model built based on the LSTM network in this embodiment.

[0027] Figure 4 This is a schematic diagram illustrating the steps of the irrigation calculation process in this embodiment. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and their descriptions are used to explain this application, but are not intended to limit it.

[0029] Traditional irrigation methods rely heavily on manual experience, failing to accurately meet the water needs of strawberries at different growth stages. This often leads to water waste, reduced yields, and inconsistent quality. While existing intelligent irrigation systems offer automated irrigation control, they still fall short in terms of comprehensive data sensing, precise irrigation decisions, and flexible execution, making it difficult to meet the precision irrigation requirements of strawberry cultivation. Figures 1 to 4 This embodiment provides an intelligent regulating device for strawberry garden irrigation, capable of precise control and remote management of irrigation water volume, water temperature, and water-fertilizer ratio. It can adjust irrigation strategies by analyzing changes in irrigation water volume and soil moisture at different growth stages, further improving irrigation accuracy and water resource utilization efficiency. Figure 1 As shown, the device includes a data sensing module, a main control module, a human-machine interaction module, and an execution module. The data sensing module collects real-time data on soil moisture, planting environment, and strawberry growth stages, and sends the collected data to the main control module. The input of the main control module is connected to the output of the data sensing module. The main control module receives the sensing data, generates dynamic irrigation decisions based on an irrigation analysis model, and sends the irrigation decision control signals to the execution module. The human-machine interaction module obtains strawberry planting parameters input by the user, queries soil and environmental monitoring data and irrigation status information, and enables remote interaction. The execution module irrigates independently according to the strawberry planting ridges based on the irrigation decision control signals.

[0030] The data sensing module includes a distributed soil monitoring unit and an environmental sensing unit. The distributed soil monitoring unit comprises multiple wireless sensor nodes, each integrating a soil moisture sensor, a temperature sensor, an EC sensor, and a LoRa wireless transmission module. The soil moisture sensor, temperature sensor, and EC sensor are each connected to the LoRa wireless transmission module. The soil moisture sensor monitors soil moisture information in the root zone, i.e., soil volumetric water content. The temperature sensor monitors the real-time temperature of irrigation water to prevent cold water from directly stimulating strawberry roots. The EC sensor detects the soil solution conductivity in real time, reflecting the fertilizer concentration. Data from each sensor is transmitted to the main control module via the LoRa wireless transmission module. The transmission interval can be dynamically adjusted according to the growth stage, such as 30 minutes / time during the seedling stage and 15 minutes / time during the fruiting stage. The environmental sensing unit includes a light intensity sensor, an air temperature and humidity sensor, a rain gauge, and a mini weather station. The light intensity sensor uses a photosynthetically active radiometer with a detection range of 0-2000 μmol / m²·s to detect changes in light intensity in the strawberry garden and analyze the relationship between strawberry photosynthetic intensity and water requirement. The air temperature and humidity sensor uses an SHT30 chip to achieve high-precision detection of temperature from -40℃ to 85℃ and humidity from 0-100%RH. The rain gauge uses a tipping bucket structure with a resolution of 0.2 mm to calculate effective rainfall. The mini weather station integrates the detection functions of meteorological parameters such as wind speed, wind direction, and solar radiation, providing basic data for evapotranspiration prediction.

[0031] Light intensity sensors are used to detect changes in light intensity in strawberry fields, air temperature and humidity sensors are used to detect environmental temperature and humidity in strawberry fields, rain gauges are used to detect rainfall, and mini weather stations are used to acquire meteorological data.

[0032] In this embodiment, the data perception module further includes an image analysis unit and a fault monitoring unit. The image analysis unit is used to preprocess strawberry plant growth images acquired by a high-definition camera. Preprocessing includes normalizing the image to 512×512 pixels, cropping based on the region of interest, Gaussian denoising (σ=1.5), histogram equalization, and contrast stretching with a stretching coefficient of 1.2. It also extracts the 128-dimensional color histogram and 3-dimensional color moment features of the strawberry plant in HSV space, calculates the gray-level co-occurrence matrix using an 8×8 window to extract 16-dimensional texture features such as energy and entropy, obtains the shape features of the strawberry plant through Canny edge detection and contour extraction, and quantizes them using 10 shape descriptors such as roundness and perimeter-area ratio. Principal component analysis is then used to select principal component features with a cumulative contribution rate ≥95% from the extracted features, reducing the feature dimension to 20-30 dimensions. An image classification model is built based on the YOLOv8 network to identify the current growth stage information of the strawberry plant. Transfer learning is used to initialize the weights. The training set contains 5000 images each of the strawberry seedling stage, flowering stage, early fruiting stage, peak fruiting stage, and ripening stage.

[0033] The growth stages of a strawberry plant can be categorized into five phases: seedling stage, flowering stage, initial fruiting stage, peak fruiting stage, and dormancy stage.

[0034] The fault monitoring unit includes a flow detector, a pipeline pressure sensor, and a current transformer, which work together to monitor faults in the irrigation system. The flow detector is an electromagnetic flow meter with a measurement range of 0-50 m³ / h and an accuracy of ±0.5%. The flow detector is used to detect drip irrigation flow information, the pipeline pressure sensor is used to detect pipeline pressure information, and the current transformer is used to detect water pump current information.

[0035] The existing irrigation system lacks real-time monitoring of its operating status, and the failure to troubleshoot faults in a timely manner will affect the irrigation effect.

[0036] The main control module includes an evapotranspiration prediction unit and an optimization analysis unit. The evapotranspiration prediction unit is used to predict the reference evapotranspiration ETO sequence for the next 24 hours based on collected historical meteorological sensing data and a prediction model based on an LSTM network. First, a prediction model is built based on an LSTM network, such as... Figure 3As shown, the structure includes an input layer (5 neurons), LSTM layer 1 (64 neurons, ReLU activation function), Dropout layer 1 (dropoutrate=0.2), LSTM layer 2 (32 neurons, tanh activation function), Dropout layer 2 (dropoutrate=0.2), a fully connected layer (16 neurons), and an output layer (24 neurons, linear activation). The model input is a time series of historical meteorological data from the past 3 days, containing 5 feature parameters: daily minimum temperature T_min (°C), daily maximum temperature T_max (°C), daily average relative humidity RH_avg (%), wind speed at 2m altitude u2 (m / s), and solar radiation Rs (MJ / m²・d). The output is a reference evapotranspiration ET0 (mm / h) sequence for the next 24 hours. During model training, each feature was individually Z-score standardized (μ=0, σ=1), using the Adam optimizer (learning rate 0.001) and the mean squared error (MSE) loss function. The training iterations were 500, and the validation set accuracy was ≥92%. The input data for the LSTM network prediction model included historical meteorological time series data from the past 3 days, with 5 features (T_min, T_max, RH_avg, u2, Rs). The output layer output 24 nodes, representing the reference evapotranspiration ETO data for the next 24 hours.

[0037] The optimization analysis unit is connected to the evapotranspiration prediction unit. The optimization analysis unit calculates the irrigation timing and amount based on strawberry growth stage information, predicted ET0, and soil moisture data. It also optimizes the irrigation amount for each stage based on the Jensen model, the expression of which is: in, This represents the actual evapotranspiration during the i-th growth stage. This represents the maximum evapotranspiration during the i-th growth stage. Let λ be the water shortage sensitivity index for the i-th growth stage, λ=0.3 for strawberry seedlings, λ=0.6 for flowering, and λ=0.8 for fruiting. Y represents the actual yield, and Ym represents the maximum yield. The optimization objective is to ensure... Under the premise of ≥0.95, the amount of irrigation water used in each stage is minimized.

[0038] like Figure 2As shown in the diagram, the framework for generating a dynamic irrigation regime for a strawberry garden based on LSTM prediction of the ET0 sequence and combined with the Jensen model is illustrated. The irrigation decision-making process of this application is as follows: a. Obtain meteorological data (5 features) for the past 72 hours; b. Standardize the data (using the same standardized parameters as during training); c. Predict ET0 for the next 24 hours using the LSTM model; d. Calculate the cumulative ET0 value for the next 24 hours; e. Obtain the current soil moisture, growth stage, and rainfall forecast (here, the rainfall forecast is set to 0); f. Calculate the dynamic soil moisture threshold; g. Determine whether irrigation should be triggered; h. If triggered, calculate the irrigation amount.

[0039] like Figure 4 As shown, the process for calculating whether irrigation is needed and the amount of irrigation is as follows: Step A: According to the formula: Calculate the soil dynamic moisture threshold. The average ET0 for the next 24 hours is predicted by a prediction model based on an LSTM network. The average ET0 over the past 7 days The ET0 response coefficient is set to 0.05 for strawberries. This sets the basic threshold for field water holding capacity at the corresponding growth stage of strawberries, with 60% for the seedling stage. 70% of the results were achieved. , Field holding capacity; Step B: According to the formula: Calculate the risk of water shortage in the next 24 hours, among which, , This represents the water consumption for the next 24 hours. Let t be the ET0 value predicted by the LSTM network-based prediction model at hour t, and Kc(t) be the dynamic crop coefficient at hour t. For effective rainfall, For the effective water storage capacity of the strawberry root zone, =0.6 × root distribution depth × The strawberry root system is distributed at a depth of 30cm. , This represents the soil moisture stress coefficient. Current soil moisture; Step C: Irrigation conditions are triggered when soil moisture is below the dynamic soil moisture threshold or the water shortage risk value is above 0.8; otherwise, irrigation is not performed. When irrigation conditions are triggered, the following formula applies: Calculate the irrigation amount, and the final irrigation amount satisfies: ,in, The maximum irrigation amount in a single instance is given by Kc, a dynamic crop coefficient of 0.6 for seedlings, 0.8 for flowering, and 1.0 for fruiting. Evapotranspiration per hour (Etm) is calculated as Etm = Kc * [Etm's value]. The total evapotranspiration for the next 24 hours is calculated by summing the hourly evapotranspiration. The amount of evapotranspiration that needs to be replenished is the total evapotranspiration for the next 24 hours. ETm,k is the total evapotranspiration for the next 24 hours, which is also the maximum water demand. A is the irrigated area. For system irrigation efficiency, drip irrigation is taken as 0.9, Rt is the effective rainfall, ΔSt is the soil water replenishment, and ΔSt=( - )*SWC, Target soil moisture, The current soil moisture is denoted as SWC, and the total water holding capacity of the soil is denoted as SWC. The irrigation amount required for the next day at the current growth stage is calculated, and the calculation results are converted into irrigation instructions and sent to the execution module. The irrigation instructions include the solenoid valve opening duration and the water pump frequency control instructions.

[0040] In this embodiment, the main control module further includes a fault self-diagnosis unit and a dual-mode power supply unit. The fault self-diagnosis unit is used to determine whether there is a blockage fault based on the real-time detected flow information and pipeline pressure information using a vector machine classification method. It is also used to perform Fourier transform on the detected water pump current information and extract feature information, and compare the extracted feature information with preset values ​​to determine whether a water pump operation fault has occurred.

[0041] The dual-mode power supply unit includes a solar power module, an automatic switching circuit, and a mains power module. The solar power module includes a solar photovoltaic panel, an MPPT solar controller, and a 24V / 200Ah lithium iron phosphate battery. The mains power module includes an AC-DC converter, a voltage regulator, and a DC-DC converter. The automatic switching circuit is electrically connected to both the solar power module and the mains power module to switch between power supply modes. The automatic switching circuit uses a MOSFET switching circuit, prioritizing solar power and switching to mains power for supplemental supply during periods of continuous cloudy / rainy weather, increased load, or when the battery voltage is below 10.5V.

[0042] The execution module of this application controls the solenoid valve array and irrigation equipment to independently irrigate strawberry planting ridges according to irrigation decision control signals. Specifically, it includes a water-fertilizer mixing unit, a water temperature regulation unit, and pressure-compensating drip irrigation pipes. The water-fertilizer mixing unit uses a combination of a Venturi fertilizer applicator and a screw pump to mix water and fertilizer. The screw pump is connected to the main control module via a frequency converter and can adjust the fertilizer application rate in real time based on feedback from an EC sensor. The water temperature regulation unit includes an electric heating element, a circulating air heat exchanger, and a three-way proportional valve. The electric heating element is placed in a warm water tank. The three-way proportional valve is used to mix cold and warm water. Water temperature control includes winter mode, summer mode, and transition mode. Winter mode uses electric heating and circulating heat preservation, with a target temperature of 18±2℃. Summer mode uses a three-way valve for cold water mixing, with an adjustment range of 10-25℃. Transition mode uses natural convection heat dissipation. The pressure-compensating drip irrigation pipes use drip tape with embedded silicone pressure compensation pads to ensure uniform drip irrigation across different planting ridges.

[0043] The human-computer interaction module in this embodiment adopts a 7-inch touchscreen and a wireless communication module architecture. The wireless communication module includes a 4G / NB-IoT module, used to acquire user-input parameters such as strawberry variety, planting density, and target yield, supporting manual input and preset template selection. Users can query real-time soil moisture (water content, temperature, EC value), environmental data (sunlight, temperature and humidity, rainfall), and irrigation status (on / off, flow rate, pressure) through the interface. Historical data is stored for one year and supports export in Excel format. Remote interaction is achieved through a mobile terminal APP, supporting operations such as modifying irrigation parameters, manual irrigation control, and receiving fault alarms.

[0044] This embodiment also includes a drainage module, which includes a drainage pump connected to the main control module. The drainage pump is used to drain water according to a control signal when the soil moisture is >90%. If rainfall exceeds 50mm / 24h, the main control module triggers the drainage pump to operate until the soil moisture drops to 70%. The following are methods to prevent strawberry roots from rotting due to water accumulation.

[0045] In this embodiment, the system was deployed in a 10-mu (approximately 1.65 acres) strawberry plantation, using the "Zhangji" variety, with raised-ridge cultivation (ridge width 80cm, ridge height 30cm, ridge spacing 50cm), dividing the plantation into 50 ridges. During data sensing module deployment, one wireless sensor node was placed between every two ridges (25 in total), with a node spacing of 10m and a burial depth of 25cm. The environmental sensing unit was installed in the center of the plantation, 1.5m above the ground. One high-definition camera was installed at each of the four corners of the plantation, with the lens facing the ridge surface, capturing images covering all ridges. The flow detector and pressure sensor of the fault monitoring unit were installed at the main pipeline inlet, and the current transformer was mounted on the water pump power supply line. The main control module used an STM32H743 microprocessor as the main control chip, integrating an LSTM neural network acceleration module, and was installed in a waterproof control cabinet placed at the edge of the plantation. The touchscreen of the human-machine interface module was mounted on the surface of the control cabinet, and the wireless communication module was connected to the local 4G network. The water and fertilizer mixing unit of the execution module is located at the irrigation front-end hub, with a warm water tank capacity of 500L; each planting ridge is equipped with one pressure-compensated drip irrigation pipe and one solenoid valve, and the main pipeline uses Φ50mm PE pipe. When deploying the drainage unit, two submersible pumps are installed in the low-lying areas of the plantation.

[0046] The irrigation control process was based on the peak fruiting period of strawberries (λ=0.8, =70% Taking irrigation control as an example, the specific process is as follows: Data Acquisition: The data sensing module collects data every 15 minutes, and the soil moisture sensor detects that the current soil moisture content is 55%. The image analysis unit identified that the strawberries were in their peak fruiting period, while the environmental perception unit recorded the average of the past 7 days. =4.2mm / d, the micro weather station collected meteorological data from the past 3 days. Evapotranspiration forecast: The evapotranspiration forecasting unit inputs the meteorological data of the past 3 days (T_min=8℃, T_max=22℃, RH_avg=65%, u2=1.5m / s, Rs=15MJ / m²・d) into the LSTM model to predict the average ET0_P=5.0mm / d for the next 24 hours, with the hourly ET0(t) sequence ranging from 0.1-0.3mm / h (night) to 0.5-0.7mm / h (noon). Parameter calculation: Step A: Calculate the soil dynamic moisture threshold: The irrigation triggering conditions are met.

[0047] Step B: Calculate the water shortage risk (WRI), where the average ET0(t) for the next 24 hours is 0.35 mm / h, Kc(t) is taken as the peak daily average coefficient of 1.0, and Ks = Therefore There was no rainfall that day, so the effective rainfall Re=0, and the root distribution depth was 30cm=0.3m. Taking 30mm / m, the effective water storage in the root zone S1 = 0.6 × 0.3 × 30 = 5.4mm, and the water shortage risk WRI = (4.62 - 0) / 5.4 ≈ 0.855 > 0.8, which meets the irrigation triggering condition.

[0048] Step C: Calculate the total evapotranspiration ETm over the next 24 hours, k = 24 × 0.35 = 8.4 mm, irrigated area A = 10 mu = 6666.7 m², Kc = 1.0, η = 0.9, ΔSt = ( - )*30mm / m×0.3m=1.35mm, Rt=0, ≈62.23m³, maximum single irrigation volume f_max=20m³ / mu×10mu=200m³, =min(62.23,200)=62.23m³, therefore the irrigation volume for this operation is determined to be 62.23m³.

[0049] The main control module converts irrigation volume into control signals. The solenoid valve opening time is approximately 150 seconds per row (62.23 m³ ÷ (total flow rate 3 m³ / h × 50 rows) × 3600 s). The screw pump frequency is 40 Hz (corresponding to a fertilizer application rate of 0.8 L / h), and the target water temperature for the electric heating element is 20℃. Upon receiving the module's response command, the water and fertilizer mixing unit adjusts the screw pump speed based on feedback from the EC sensor. The water temperature regulation unit stabilizes the outlet water temperature at 20 ± 1℃ via a three-way proportional valve. The 50 solenoid valves open in rotation (10 rows per rotation, with a 5-minute interval). The pressure-compensated drip irrigation pipe maintains a working pressure of 0.2 MPa. During irrigation, the fault monitoring unit collects data every 5 seconds; flow rate, pressure, and pump current are all within normal ranges. When the 30th row has been irrigated for 120 seconds, the soil moisture sensor indicates that the moisture content of that row has reached 70%. The main control module closes the corresponding solenoid valves in advance to avoid over-irrigation. After all ridges are irrigated, the execution module closes all solenoid valves, screw pumps, and heating pipes. The human-machine interface module records the irrigation data (time, water volume, energy consumption) and uploads it to the cloud. The data sensing module resumes the normal collection frequency and continues to monitor soil information.

[0050] This invention provides a strawberry irrigation device based on intelligent sensing and decision-making. It integrates multiple sensors to achieve comprehensive sensing of the strawberry growing environment and growth status, and uses a data analysis model to generate precise irrigation decisions. An intelligent execution module enables independent irrigation in designated areas, improving irrigation efficiency and quality for strawberry cultivation. It also features intelligent fault diagnosis and flexible power supply capabilities.

[0051] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. In some cases, the actions or steps recorded in the specification and claims can be performed in a different order than that shown in the embodiments, and the desired result can still be achieved. In addition, the processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result; in some embodiments, multitasking and parallel processing are also feasible or advantageous.

[0052] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing the relevant hardware to implement them. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0053] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to the embodiments of this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An intelligent regulating device for strawberry garden irrigation, characterized in that, include: The module consists of a data sensing module, a main control module, a human-computer interaction module, and an execution module. The data sensing module is used to collect soil moisture, planting environment data and strawberry growth stage data in real time, and send the collected sensing data to the main control module. The input terminal of the main control module is connected to the output terminal of the data sensing module. The main control module is used to receive sensing data, generate dynamic irrigation decisions based on the irrigation analysis model, and send irrigation decision control signals to the execution module. The human-computer interaction module is used to obtain strawberry planting parameter information input by the user, query soil and environmental monitoring data and irrigation status information, and perform remote interaction; The execution module is used to control the solenoid valve array and irrigation equipment to irrigate independently according to the strawberry planting ridges based on the irrigation decision control signal.

2. The intelligent regulating device for strawberry garden irrigation according to claim 1, characterized in that, The data sensing module includes a distributed soil monitoring unit and an environmental sensing unit; The distributed soil monitoring unit includes multiple wireless sensor nodes. Each node integrates a soil moisture sensor, a temperature sensor, an EC sensor, and a LoRa wireless transmission module. The soil moisture sensor, the temperature sensor, and the EC sensor are respectively connected to the LoRa wireless transmission module. The soil moisture sensor is used to monitor soil moisture information in the root distribution area, the temperature sensor is used to monitor irrigation water temperature information, and the EC sensor is used to detect fertilizer concentration in real time. The environmental sensing unit includes a light intensity sensor, an air temperature and humidity sensor, a rain gauge, and a mini weather station. The light intensity sensor is used to detect changes in light intensity in the strawberry garden, the air temperature and humidity sensor is used to detect the ambient temperature and humidity in the strawberry garden, the rain gauge is used to detect rainfall, and the mini weather station is used to acquire meteorological data.

3. The intelligent regulating device for strawberry garden irrigation according to claim 2, characterized in that, The data sensing module also includes an image analysis unit and a fault monitoring unit; The image analysis unit is used to perform normalization, cropping, Gaussian denoising, histogram equalization, and contrast stretching preprocessing on the strawberry plant growth images acquired by the camera. It also extracts the color histogram and color moment features of the strawberry plant in the HSV space, extracts texture features using gray-level co-occurrence matrix and local binary mode methods, obtains the shape features of the strawberry through edge detection and contour extraction methods, and performs quantitative description in combination with shape descriptors. Then, it uses principal component analysis to select the most discriminative feature combination from the extracted features to reduce the feature dimensionality. Finally, it uses an image classification model based on the YOLOv8 network to identify the current growth stage information of the strawberry plant. The fault monitoring unit includes a flow detector, a pipeline pressure sensor, and a current transformer. The flow detector is used to detect the flow rate of the drip tube, the pipeline pressure sensor is used to detect the pipeline pressure, and the current transformer is used to detect the pump current.

4. The intelligent regulating device for strawberry garden irrigation according to claim 1, characterized in that, The main control module includes an evapotranspiration prediction unit and an optimization analysis unit; The evapotranspiration prediction unit is used to predict the reference evapotranspiration ETO sequence for the next 24 hours based on the collected historical meteorological sensing data and a prediction model based on an LSTM network. The prediction model based on an LSTM network includes an input layer, LSTM layer 1, Droput layer 1, LSTM layer 2, Droput layer 2, a fully connected layer, and an output layer. The optimization analysis unit is connected to the evapotranspiration prediction unit. The optimization analysis unit is used to calculate whether irrigation is needed and the amount of irrigation based on strawberry growth stage information, predicted ET0, and soil moisture data, using the Jensen model. Optimize irrigation water volume during the optimization phase, among which, This represents the actual evapotranspiration during the i-th growth stage. This represents the maximum evapotranspiration during the i-th growth stage. Let Y be the water shortage sensitivity index for the i-th growth stage, Y be the actual yield, and Ym be the maximum yield, while ensuring... Under the premise of ≥0.95, the process of calculating whether irrigation is needed and the amount of irrigation is as follows: According to the formula: Calculate the soil dynamic moisture threshold. The average ET0 for the next 24 hours is predicted by a prediction model based on an LSTM network. The average ET0 over the past 7 days The ET0 response coefficient is set to 0.05 for strawberries. This represents the basic threshold for field water holding capacity at the corresponding growth stage of strawberries; According to the formula: Calculate the risk of water shortage in the next 24 hours, among which, , This represents the water consumption for the next 24 hours. Let t be the ET0 value predicted by the LSTM network-based prediction model at hour t, and Kc(t) be the dynamic crop coefficient at hour t. For effective rainfall, This represents the effective water storage capacity of the strawberry root zone. Irrigation is triggered when soil moisture falls below the dynamic soil moisture threshold or the water shortage risk value exceeds 0.8; otherwise, irrigation is not performed. When irrigation is triggered, the following formula applies: Calculate the irrigation amount, and the final irrigation amount satisfies: ,in, The maximum irrigation volume per cycle is given by Kc, which is the dynamic crop coefficient. Evapotranspiration per hour (Etm) is calculated as Etm = Kc * [Etm's value]. The total evapotranspiration for the next 24 hours is calculated by summing the hourly evapotranspiration. The amount of evapotranspiration that needs to be replenished is the total evapotranspiration for the next 24 hours. ETm,k is the total evapotranspiration for the next 24 hours, which is also the maximum water demand. A is the irrigated area. For system irrigation efficiency, drip irrigation is taken as 0.9, Rt is the effective rainfall, ΔSt is the soil water replenishment, and ΔSt=( - )*SWC, Target soil moisture, Given the current soil moisture and SWC (total soil water holding capacity), the system calculates the irrigation amount required for the next day during the current growth stage and converts the calculation results into an irrigation command, which is then sent to the execution module.

5. The intelligent regulating device for strawberry garden irrigation according to claim 4, characterized in that, The main control module also includes a fault self-diagnosis unit and a dual-mode power supply unit; The fault self-diagnosis unit is used to determine whether there is a blockage fault based on the real-time detected flow information and pipeline pressure information, and is also used to compare the detected water pump current information with a preset value to determine whether a water pump operation fault has occurred. The dual-mode power supply unit includes a solar power supply module, an automatic switching circuit, and a mains power supply module. The solar power supply module includes a solar photovoltaic panel, a solar controller, and a battery. The mains power supply module includes an AC-DC converter, a voltage regulator, and a DC-DC converter. The automatic switching circuit is electrically connected to the solar power supply module and the mains power supply module and is used to switch the power supply mode.

6. The intelligent regulating device for strawberry garden irrigation according to claim 1, characterized in that, The execution module includes a water and fertilizer mixing unit, a water temperature regulation unit, and a pressure-compensating drip irrigation pipe; The water and fertilizer mixing unit is used to mix water and fertilizer by combining a Venturi fertilizer applicator and a screw pump. The screw pump is connected to the main control module via a frequency converter. The water temperature regulating unit includes an electric heating element, a circulating air heat exchanger, and a three-way proportional valve. The electric heating element is installed in the warm water tank, and the three-way proportional valve is used to mix cold water and warm water. The pressure-compensating drip irrigation pipe uses drip irrigation tape with an embedded silicone pressure compensation sheet.

7. The intelligent regulating device for strawberry garden irrigation according to claim 1, characterized in that, It also includes a drainage module, which includes a drainage pump connected to the main control module. The drainage pump is used to drain water according to a control signal.

Images